GPS World » Algorithms & Methods

GNSS PPP Workshop Early Registration Extended to May 3

GPS World staff — Thu, 25 Apr 2013 23:02:24 +0000

The International Association of Geodesy, Natural Resources Canada, the International GNSS Service, and York University will be hosting GNSS Precise Point Positioning: Reaching Full Potential in Ottawa, Canada, June 12-14, 2013.

The primary objective of this workshop is to provide a forum for international experts from academia, government and industry to discuss PPP-related matters, including data processing, error modelling, data products, dissemination, applications, and associated policy.

The preliminary program is now available on the workshop website, along with details about accommodations and registration. Note that early registration has been extended until May 3, 2013.

Given recent rapid developments in PPP technology, the objectives of this workshop will be to:

Provide a forum for international experts from academia, government and industry to discuss PPP-related matters, including data processing, error modelling, data products, dissemination, applications, and associated policy.
Define the current state of PPP performance and communicate global PPP activities and applications in all sectors.
Identify and investigate the technical and non-technical issues that need to be addressed to improve the technology.
Suggest PPP performance and utility in the next five to ten years.

Innovation: Getting at the Truth: A Civilian GPS Position Authentication System

Richard Langley — Tue, 01 Jan 2013 20:32:51 +0000

By Zhefeng Li and Demoz Gebre-Egziabher

INNOVATION INSIGHTS with Richard Langley

My University, the University of New Brunswick, is one of the few institutes of higher learning still using Latin at its graduation exercises. The president and vice-chancellor of the university asks the members of the senate and board of governors present “Placetne vobis Senatores, placetne, Gubernatores, ut hi supplicatores admittantur?” (Is it your pleasure, Senators, is it your pleasure, Governors, that these supplicants be admitted?). In the Oxford tradition, a supplicant is a student who has qualified for their degree but who has not yet been admitted to it. Being a UNB senator, I was familiar with this usage of the word supplicant. But I was a little surprised when I first read a draft of the article in this month’s Innovation column with its use of the word supplicant to describe the status of a GPS receiver.

If we look up the definition of supplicant in a dictionary, we find that it is “a person who makes a humble or earnest plea to another, especially to a person in power or authority.” Clearly, that describes our graduating students. But what has it got to do with a GPS receiver? Well, it seems that the word supplicant has been taken up by engineers developing protocols for computer communication networks and with a similar meaning. In this case, a supplicant (a computer or rather some part of its operating system) at one end of a secure local area network seeks authentication to join the network by submitting credentials to the authenticator on the other end. If authentication is successful, the computer is allowed to join the network. The concept of supplicant and authenticator is used, for example, in the IEEE 802.1X standard for port-based network access control.

Which brings us to GPS. When a GPS receiver reports its position to a monitoring center using a radio signal of some kind, how do we know that the receiver or its associated communications unit is telling the truth? It’s not that difficult to generate false position reports and mislead the monitoring center into believing the receiver is located elsewhere — unless an authentication procedure is used. In this month’s column, we look at the development of a clever system that uses the concept of supplicant and authenticator to assess the truthfulness of position reports.

“Innovation” is a regular feature that discusses advances in GPS technology andits applications as well as the fundamentals of GPS positioning. The column is coordinated by Richard Langley of the Department of Geodesy and Geomatics Engineering, University of New Brunswick. He welcomes comments and topic ideas. Contact him at lang @ unb.ca.

This article deals with the problem of position authentication. The term “position authentication” as discussed in this article is taken to mean the process of checking whether position reports made by a remote user are truthful (Is the user where they say they are?) and accurate (In reality, how close is a remote user to the position they are reporting?). Position authentication will be indispensable to many envisioned civilian applications. For example, in the national airspace of the future, some traffic control services will be based on self-reported positions broadcast via ADS-B by each aircraft. Non-aviation applications where authentication will be required include tamper-free shipment tracking and smart-border systems to enhance cargo inspection procedures at commercial ports of entry. The discussions that follow are the outgrowth of an idea first presented by Sherman Lo and colleagues at Stanford University (see Further Reading).

For illustrative purposes, we will focus on the terrestrial application of cargo tracking. Most of the commercial fleet and asset tracking systems available in the market today depend on a GPS receiver installed on the cargo or asset. The GPS receiver provides real-time location (and, optionally, velocity) information. The location and the time when the asset was at a particular location form the tracking message, which is sent back to a monitoring center to verify if the asset is traveling in an expected manner. This method of tracking is depicted graphically in FIGURE 1.

FIGURE 1. A typical asset tracking system.

The approach shown in Figure 1 has at least two potential scenarios or fault modes, which can lead to erroneous tracking of the asset. The first scenario occurs when an incorrect position solution is calculated as a result of GPS RF signal abnormalities (such as GPS signal spoofing). The second scenario occurs when the correct position solution is calculated but the tracking message is tampered with during the transmission from the asset being tracked to the monitoring center. The first scenario is a falsification of the sensor and the second scenario is a falsification of the transmitted position report.

The purpose of this article is to examine the problem of detecting sensor or report falsification at the monitoring center. We discuss an authentication system utilizing the white-noise-like spreading codes of GPS to calculate an authentic position based on a snapshot of raw IF signal from the receiver.

Using White Noise as a Watermark

The features for GPS position authentication should be very hard to reproduce and unique to different locations and time. In this case, the authentication process is reduced to detecting these features and checking if these features satisfy some time and space constraints. The features are similar to the well-designed watermarks used to detect counterfeit currency.

A white-noise process that is superimposed on the GPS signal would be a perfect watermark signal in the sense that it is impossible reproduce and predict. FIGURE 2 is an abstraction that shows how the above idea of a superimposed white-noise process would work in the signal authentication problem. The system has one transmitter, T_x , and two receivers, R_s and R_a. R_s is the supplicant and R_a is the authenticator. The task of the authenticator is to determine whether the supplicant is using a signal from T_x or is being spoofed by a malicious transmitter, T_m. R_a is the trusted source, which gets a copy of the authentic signal, V_x(t) (that is, the signal transmitted by T_x). The snapshot signal, V_s(t), received at R_s is sent to the trusted agent to compare with the signal, V_a(t), received at R_a. Every time a verification is performed, the snapshot signal from R_s is compared with a piece of the signal from R_a. If these two pieces of signal match, we can say the snapshot signal from R_s was truly transmitted from T_x. For the white-noise signal, match detection is accomplished via a cross-correlation operation (see Further Reading). The cross-correlation between one white-noise signal and any other signal is always zero. Only when the correlation is between the signal and its copy will the correlation have a non-zero value. So a non-zero correlation means a match. The time when the correlation peak occurs provides additional information about the distance between R_a and R_s.

Unfortunately, generation of a white-noise watermark template based on a mathematical model is impossible. But, as we will see, there is an easy-to-use alternative.

FIGURE 2. Architecture to detect a snapshot of a white-noise signal.

An Intrinsic GPS Watermark

The RF carrier broadcast by each GPS satellite is modulated by the coarse/acquisition (C/A) code, which is known and which can be processed by all users, and the encrypted P(Y) code, which can be decoded and used by Department of Defense (DoD) authorized users only. Both civilians and DoD-authorized users see the same signal. To commercial GPS receivers, the P(Y) code appears as uncorrelated noise. Thus, as discussed above, this noise can be used as a watermark, which uniquely encodes locations and times. In a typical civilian GPS receiver’s tracking loop, this watermark signal can be found inside the tracking loop quadrature signal.

The position authentication approach discussed here is based on using the P(Y) signal to determine whether a user is utilizing an authentic GPS signal. This method uses a segment of noisy P(Y) signal collected by a trusted user (the authenticator) as a watermark template. Another user’s (the supplicant’s) GPS signal can be compared with the template signal to judge if the user’s position and time reports are authentic. Correlating the supplicant’s signal with the authenticator’s copy of the signal recorded yields a correlation peak, which serves as a watermark. An absent correlation peak means the GPS signal provided by the supplicant is not genuine. A correlation peak that occurs earlier or later than predicted (based on the supplicant’s reported position) indicates a false position report.

System Architecture

FIGURE 3 is a high-level architecture of our proposed position authentication system. In practice, we need a short snapshot of the raw GPS IF signal from the supplicant. This piece of the signal is the digitalized, down-converted, IF signal before the tracking loops of a generic GPS receiver. Another piece of information needed from the supplicant is the position solution and GPS Time calculated using only the C/A signal. The raw IF signal and the position message are transmitted to the authentication center by any data link (using a cell-phone data network, Wi-Fi, or other means).

FIGURE 3. Architecture of position authentication system.

The authentication station keeps track of all the common satellites seen by both the authenticator and the supplicant. Every common satellite’s watermark signal is then obtained from the authenticator’s tracking loop. These watermark signals are stored in a signal database. Meanwhile, the pseudorange between the authenticator and every satellite is also calculated and is stored in the same database.

When the authentication station receives the data from the supplicant, it converts the raw IF signal into the quadrature (Q) channel signals. Then the supplicant’s Q channel signal is used to perform the cross-correlation with the watermark signal in the database. If the correlation peak is found at the expected time, the supplicant’s signal passes the signal-authentication test. By measuring the relative peak time of every common satellite, a position can be computed. The position authentication involves comparing the reported position of the supplicant to this calculated position. If the difference between two positions is within a pre-determined range, the reported position passes the position authentication.

While in principle it is straightforward to do authentication as described above, in practice there are some challenges that need to be addressed. For example, when there is only one common satellite, the only common signal in the Q channel signals is this common satellite’s P(Y) signal. So the cross-correlation only has one peak. If there are two or more common satellites, the common signals in the Q channel signals include not only the P(Y) signals but also C/A signals. Then the cross-correlation result will have multiple peaks. We call this problem the C/A leakage problem, which will be addressed below.

C/A Residual Filter

The C/A signal energy in the GPS signal is about double the P(Y) signal energy. So the C/A false peaks are higher than the true peak. The C/A false peaks repeat every 1 millisecond. If the C/A false peaks occur, they are greater than the true peak in both number and strength. Because of background noise, it is hard to identify the true peak from the correlation result corrupted by the C/A residuals.

To deal with this problem, a high-pass filter can be used. Alternatively, because the C/A code is known, a match filter can be designed to filter out any given GPS satellite’s C/A signal from the Q channel signal used for detection. However, this implies that one match filter is needed for every common satellite simultaneously in view of the authenticator and supplicant. This can be cumbersome and, thus, the filtering approach is pursued here.

In the frequency domain, the energy of the base-band C/A signal is mainly (56 percent) within a ±1.023 MHz band, while the energy of the base-band P(Y) signal is spread over a wider band of ±10.23 MHz. A high-pass filter can be applied to Q channel signals to filter out the signal energy in the ±1.023 MHz band. In this way, all satellites’ C/A signal energy can be attenuated by one filter rather than using separate match filters for different satellites.

FIGURE 4 is the frequency response of a high-pass filter designed to filter out the C/A signal energy. The spectrum of the C/A signal is also plotted in the figure. The high-pass filter only removes the main lobe of the C/A signals. Unfortunately, the high-pass filter also attenuates part of the P(Y) signal energy. This degrades the auto-correlation peak of the P(Y) signal. Even though the gain of the high-pass filter is the same for both the C/A and the P(Y) signals, this effect on their auto-correlation is different. That is because the percentage of the low-frequency energy of the C/A signal is much higher than that of the P(Y) signal. This, however, is not a significant drawback as it may appear initially. To see why this is so, note that the objective of the high-pass filter is to obtain the greatest false-peak rejection ratio defined to be the ratio between the peak value of P(Y) auto-correlation and that of the C/A auto-correlation. The false-peak rejection ratio of the non-filtered signals is 0.5. Therefore, all one has to do is adjust the cut-off frequency of the high-pass filter to achieve a desired false-peak rejection ratio.

FIGURE 4. Frequency response of the notch filter.

The simulation results in FIGURE 5 show that one simple high-pass filter rather than multiple match filters can be designed to achieve an acceptable false-peak rejection ratio. The auto-correlation peak value of the filtered C/A signal and that of the filtered P(Y) signal is plotted in the figure. While the P(Y) signal is attenuated by about 25 percent, the C/A code signal is attenuated by 91.5 percent (the non-filtered C/A auto-correlation peak is 2). The false-peak rejection ratio is boosted from 0.5 to 4.36 by using the appropriate high-pass filter.

FIGURE 5. Auto-correlation of the filtered C/A and P(Y) signals.

Position Calculation

Consider the situation depicted in FIGURE 6 where the authenticator and the supplicant have multiple common satellites in view. In this case, not only can we perform the signal authentication but also obtain an estimate of the pseudorange information from the authentication. Thus, the authenticated pseudorange information can be further used to calculate the supplicant’s position if we have at least three estimates of pseudoranges between the supplicant and GPS satellites. Since this position solution of the supplicant is based on the P(Y) watermark signal rather than the supplicant’s C/A signal, it is an independent and authentic solution of the supplicant’s position. By comparing this authentic position with the reported position of the supplicant, we can authenticate the veracity of the supplicant’s reported GPS position.

FIGURE 6. Positioning using a watermark signal.

The situation shown in Figure 6 is very similar to double-difference differential GPS. The major difference between what is shown in the figure and the traditional double difference is how the differential ranges are calculated. Figure 6 shows how the range information can be obtained during the signal authentication process. Let us assume that the authenticator and the supplicant have four common GPS satellites in view: SAT1, SAT2, SAT3, and SAT4. The signals transmitted from the satellites at time t are S₁(t), S₂(t), S₃(t), and S₄(t), respectively. Suppose a signal broadcast by SAT1 at time t₀ arrives at the supplicant at t₀ + ν₁^s where ν₁^s is the travel time of the signal. At the same time, signals from SAT2, SAT3, and SAT4 are received by the supplicant. Let us denote the travel time of these signals as ν₂^s, ν₃^s, and ν₄^s, respectively. These same signals will be also received at the authenticator. We will denote the travel times for the signals from satellite to authenticator as ν₁^a, ν₂^a, ν₃^a, and ν₄^a. The signal at a receiver’s antenna is the superposition of the signals from all the satellites. This is shown in FIGURE 7 where a snapshot of the signal received at the supplicant’s antenna at time t₀ + ν₁^s includes GPS signals from SAT1, SAT2, SAT3, and SAT4. Note that even though the arrival times of these signals are the same, their transmit times (that is, the times they were broadcast from the satellites) are different because the ranges are different. The signals received at the supplicant will be S₁(t₀), S₂(t₀ + ν₁^s – ν₂^s), S₃(t₀ + ν₁^s – ν₃^s), and S₄(t₀ + ν₁^s – ν₄^s). This same snapshot of the signals at the supplicant is used to detect the matched watermark signals from SAT1, SAT2, SAT3, and SAT4 at the authenticator. Thus the correlation peaks between the supplicant’s and the authenticator’s signal should occur at t₀ + ν₁^a, t₀ + ν₁^s – ν₂^s + ν₂^a, t₀ + ν₁^s – ν₃^s + ν₃^a, and t₀ + ν₁^s – ν₄^s + ν₄^a.

Referring to Figure 6 again, suppose the authenticator’s position (x_a, y_a, z_a) is known but the supplicant’s position (x_s, y_s, z_s) is unknown and needs to be determined. Because the actual ith common satellite (x_i , y_i , z_i ) is also known to the authenticator, each of the ρ_i^a, the pseudorange between the ith satellite and the authenticator, is known. If ρ_i^s is the pseudorange to the ith satellite measured at the supplicant, the pseudoranges and the time difference satisfies equation (1):

ρ₂^s – ρ₁^s= ρ₂^a – ρ₁^a – ct₂₁ + cχ₂₁(1)

where χ₂₁ is the differential range error primarily due to tropospheric and ionospheric delays. In addition, c is the speed of light, and t₂₁ is the measured time difference as shown in Figure 7. Finally, ρ_i^s for i = 1, 2, 3, 4 is given by:

(2)

FIGURE 7. Relative time delays constrained by positions.

If more than four common satellites are in view between the supplicant and authenticator, equation (1) can be used to form a system of equations in three unknowns. The unknowns are the components of the supplicant’s position vector r_s = [x_s, y_s, z_s]^T. This equation can be linearized and then solved using least-squares techniques. When linearized, the equations have the following form:

Aδr_s= δm (3)

where δr_s = [δx_s,δy_s,δz_s]^T, which is the estimation error of the supplicant’s position. The matrix A is given by

where is the line of sight vector from the supplicant to the ith satellite. Finally, the vector δm is given by:

(4)

where δr_iis the ith satellite’s position error, δρ_i^a is the measurement error of pseudorange ρ_i^a or pseudorange noise. In addition, δt_ij is the time difference error. Finally, δχ_ij is the error of χ_ij defined earlier.

Equation (3) is in a standard form that can be solved by a weighted least-squares method. The solution is

δr_s= ( A^T R^-1 A)^-1 A^T R^-1δm (5)

where R is the covariance matrix of the measurement error vector δm. From equations (3) and (5), we can see that the supplicant’s position accuracy depends on both the geometry and the measurement errors.

Hardware and Software

In what follows, we describe an authenticator which is designed to capture the GPS raw signals and to test the performance of the authentication method described above. Since we are relying on the P(Y) signal for authentication, the GPS receivers used must have an RF front end with at least a 20-MHz bandwidth. Furthermore, they must be coupled with a GPS antenna with a similar bandwidth. The RF front end must also have low noise. This is because the authentication method uses a noisy piece of the P(Y) signal at the authenticator as a template to detect if that P(Y) piece exists in the supplicant’s raw IF signal. Thus, the detection is very sensitive to the noise in both the authenticator and the supplicant signals. Finally, the sampling of the down-converted and digitized RF signal must be done at a high rate because the positioning accuracy depends on the accuracy of the pseudorange reconstructed by the authenticator. The pseudorange is calculated from the time-difference measurement. The accuracy of this time difference depends on the sampling frequency to digitize the IF signal. The high sampling frequency means high data bandwidth after the sampling.

The authenticator designed for this work and shown in FIGURE 8 satisfies the above requirements. A block diagram of the authenticator is shown in Figure 8a and the constructed unit in Figure 8b. The IF signal processing unit in the authenticator is based on the USRP N210 software-defined radio. It offers the function of down converting, digitalization, and data transmission. The firmware and field-programmable-gate-array configuration in the USRP N210 are modified to integrate a software automatic gain control and to increase the data transmission efficiency. The sampling frequency is 100 MHz and the effective resolution of the analog-to-digital conversion is 6 bits. The authenticator is battery powered and can operate for up to four hours at full load.

FIGURE 8a. Block diagram of GPS position authenticator.

FIGURE 8b. Photo of constructed unit.

Performance Validation

Next, we present results demonstrating the performance of the authenticator described above. First, we present results that show we can successfully deal with the C/A leakage problem using the simple high-pass filter. We do this by performing a correlation between snapshots of signal collected from the authenticator and a second USRP N210 software-defined radio. FIGURE 9a is the correlation result without the high-pass filter. The periodic peaks in the result have a period of 1 millisecond and are a graphic representation of the C/A leakage problem. Because of noise, these peaks do not have the same amplitude. FIGURE 9b shows the correlation result using the same data snapshot as in Figure 9a. The difference is that Figure 9b uses the high-pass filter to attenuate the false peaks caused by the C/A signal residual. Only one peak appears in this result as expected and, thus, confirms the analysis given earlier.

FIGURE 9a. Example of cross-correlation detection results without high-pass filter.

FIGURE 9b. Example of cross-correlation with high-pass filter.

We performed an experiment to validate the authentication performance. In this experiment, the authenticator and the supplicant were separated by about 1 mile (about 1.6 kilometers). The location of the authenticator was fixed. The supplicant was then sequentially placed at five points along a straight line. The distance between two adjacent points is about 15 meters. The supplicant was in an open area with no tall buildings or structures. Therefore, a sufficient number of satellites were in view and multipath, if any, was minimal. The locations of the five test points are shown in FIGURE 10.

FIGURE 10. Five-point field test. Image courtesy of Google.

The first step of this test was to place the supplicant at point A and collect a 40-millisecond snippet of data. This data was then processed by the authenticator to determine if:

The signal contained the watermark. We call this the “signal authentication test.” It determines whether a genuine GPS signal is being used to form the supplicant’s position report.
The supplicant is actually at the position coordinates that they say they are. We call this the “position authentication test.” It determines whether or not falsification of the position report is being attempted.

Next, the supplicant was moved to point B. However, in this instance, the supplicant reports that it is still located at point A. That is, it makes a false position report. This is repeated for the remaining positions (C through E) where at each point the supplicant reports that it is located at point A. That is, the supplicant continues to make false position reports.

In this experiment, we have five common satellites between the supplicant (at all of the test points A to E) and the authenticator. The results of the experiment are summarized in TABLE 1. If we can detect a strong peak for every common satellite, we say this point passes the signal authentication test (and note “Yes” in second column of Table 1). That means the supplicant’s raw IF signal has the watermark signal from every common satellite. Next, we perform the position authentication test. This test tries to determine whether the supplicant is at the position it claims to be. If we determine that the position of the supplicant is inconsistent with its reported position, we say that the supplicant has failed the position authentication test. In this case we put a “No” in the third column of Table 1. As we can see from Table 1, the performance of the authenticator is consistent with the test setup. That is, even though the wrong positions of points (B, C, D, E) are reported, the authenticator can detect the inconsistency between the reported position and the raw IF data. Furthermore, since the distance between two adjacent points is 15 meters, this implies that resolution of the position authentication is at or better than 15 meters. While we have not tested it, based on the timing resolution used in the system, we believe resolutions better than 12 meters are achievable.

Table 1. Five-point position authentication results.

Conclusion

In this article, we have described a GPS position authentication system. The authentication system has many potential applications where high credibility of a position report is required, such as cargo and asset tracking. The system detects a specific watermark signal in the broadcast GPS signal to judge if a receiver is using the authentic GPS signal. The differences between the watermark signal travel times are constrained by the positions of the GPS satellites and the receiver. A method to calculate an authentic position using this constraint is discussed and is the basis for the position authentication function of the system. A hardware platform that accomplishes this was developed using a software-defined radio. Experimental results demonstrate that this authentication methodology is sound and has a resolution of better than 15 meters. This method can also be used with other GNSS systems provided that watermark signals can be found. For example, in the Galileo system, the encrypted Public Regulated Service signal is a candidate for a watermark signal.

In closing, we note that before any system such as ours is fielded, its performance with respect to metrics such as false alarm rates (How often do we flag an authentic position report as false?) and missed detection probabilities (How often do we fail to detect false position reports?) must be quantified. Thus, more analysis and experimental validation is required.

Acknowledgments

The authors acknowledge the United States Department of Homeland Security (DHS) for supporting the work reported in this article through the National Center for Border Security and Immigration under grant number 2008-ST-061-BS0002. However, any opinions, findings, conclusions or recommendations in this article are those of the authors and do not necessarily reflect views of the DHS. This article is based on the paper “Performance Analysis of a Civilian GPS Position Authentication System” presented at PLANS 2012, the Institute of Electrical and Electronics Engineers / Institute of Navigation Position, Location and Navigation Symposium held in Myrtle Beach, South Carolina, April 23–26, 2012.

Manufacturers

The GPS position authenticator uses an Ettus Research LLC model USRP N210 software-defined radio with a DBSRX2 RF daughterboard.

Zhefeng Li is a Ph.D. candidate in the Department of Aerospace Engineering and Mechanics at the University of Minnesota, Twin Cities. His research interests include GPS signal processing, real-time implementation of signal processing algorithms, and the authentication methods for civilian GNSS systems.

Demoz Gebre-Egziabher is an associate professor in the Department of Aerospace Engineering and Mechanics at the University of Minnesota, Twin Cities. His research deals with the design of multi-sensor navigation and attitude determination systems for aerospace vehicles ranging from small unmanned aerial vehicles to Earth-orbiting satellites.

Innovation: Getting Along: Collaborative Navigation in Transitional Environments

GPS World staff — Thu, 01 Nov 2012 22:02:08 +0000

By Dorota A. Grejner-Brzezinska, J.N. (Nikki) Markiel, Charles K. Toth and Andrew Zaydak

INNOVATION INSIGHTS with Richard Langley

COLLABORATION, n. /kəˌlæbəˈreɪʃən/, n. of action. United labour, co-operation; esp. in literary, artistic, or scientific work — according to the Oxford English Dictionary. Collaboration is something we all practice, knowingly or unknowingly, even in our everyday lives. It generally results in a more productive outcome than acting individually. In scientific and engineering circles, collaboration in research is extremely common with most published papers having multiple authors, for example.

The term collaboration can be applied not only to the endeavors of human beings or other living creatures but also to inanimate objects, too. Researchers have developed systems of miniaturized robots and unmanned vehicles that operate collaboratively to complete a task. These platforms must navigate as part of their functions and this navigation can often be made more continuous and accurate if each individual platform navigates collaboratively in the group rather than autonomously. This is typically achieved by exchanging sensor measurements by some kind of short-range wireless technology such as Wi-Fi, ultra-wide band, or ZigBee, a suite of communication protocols for small, low-power digital radios based on an Institute of Electrical and Electronics Engineers’ standard for personal area networks.

A wide variety of navigation sensors can be implemented for collaborative navigation depending on whether the system is designed by outdoor use, for use inside buildings, or for operations in a wide variety of environments. In addition to GPS and other global navigation satellite systems, inertial measurement units, terrestrial radio-based navigation systems, laser and acoustic ranging, and image-based systems can be used.

In this month’s article, a team of researchers at The Ohio State University discusses a system under development for collaborative navigation in transitional environments — environments in which GPS alone is insufficient for continuous and accurate navigation. Their prototype system involves a land-based deployment vehicle and a human operator carrying a personal navigator sensor assembly, which initially navigate together before the personal navigator transitions to an indoor environment. This system will have multiple applications including helping first responders to emergencies. Read on.

Collaborative navigation is an emerging field where a group of users navigates together by exchanging navigation and inter-user ranging information. This concept has been considered a viable alternative for GPS-challenged environments. However, most of the developed systems and approaches are based on fixed types and numbers of sensors per user or platform (restricted in sensor configuration) that eventually leads to a limitation in navigation capability, particularly in mixed or transition environments.

As an example of an applicable scenario, consider an emergency crew navigating initially in a deployment vehicle, and, when subsequently dispatched, continuing in collaborative mode, referring to the navigation solution of the other users and vehicles. This approach is designed to assure continuous navigation solution of distributed agents in transition environments, such as moving between open areas, partially obstructed areas, and indoors when different types of users need to maintain high-accuracy navigation capability in relative and absolute terms.

At The Ohio State University (OSU), we have developed systems that use multiple sensors and communications technologies to investigate, experimentally, the viability and performance attributes of such collaborative navigation. For our experiments, two platforms, a land-based deployment vehicle and a human operator carrying a personal navigator (PN) sensor assembly, initially navigate together before the PN transitions to the indoor environment.

In the article, we describe the concept of collaborative navigation, briefly describe the systems we have developed and the algorithms used, and report on the results of some of our tests. The focus of the study being reported here is on the environment-to-environment transition and indoor navigation based on 3D sensor imagery, initially in post-processing mode with a plan to transition to real time.

The Concept

Collaborative navigation, also referred to as cooperative navigation or positioning, is a localization technique emerging from the field of wireless sensor networks (WSNs). Typically, the nodes in a WSN can communicate with each other using wireless communications technology based on standards, such as Zigbee/IEEE 802.15.4. The communication signals in a WSN are used to derive the inter-nodal distances across the network. Then, the collaborative navigation solution is formed by integrating the inter-nodal range measurements among nodes (users) in the network using a centralized or decentralized Kalman filter, or a least-squares-based approach.

A paradigm shift from single to multi-sensor to multi-platform navigation is illustrated conceptually in Figure 1. While conventional sensor integration and integrated sensor systems are commonplace in navigation, sensor networks of integrated sensor systems are a relatively new development in navigation. Figure 2 illustrates the concept of collaborative navigation with emphasis on transitions between varying environments. In actual applications, example networks include those formed by soldiers, emergency crews, and formations of robots or unmanned vehicles, with the primary objective of achieving a sustained level of sufficient navigation accuracy in GPS-denied environments and assuring seamless transition among sensors, platforms, and environments.

Figure 1. Paradigm shift in sensor integration concept for navigation.

Figure 2. Collaborative navigation and transition between varying environments.

Field Experiments and Methodology

A series of field experiments were carried out in the fall of 2011 at The Ohio State University (OSU), and in the spring of 2012 at the Nottingham Geospatial Institute of the University of Nottingham, using the updated prototype of the personal navigator developed earlier at the OSU Satellite Positioning and Inertial Navigation Laboratory, and land-based multisensory vehicles. Note that the PN prototype is not a miniaturized system, but rather a sensor assembly put together using commercial off-the-shelf components for demonstration purposes only.

The GPSVan (see Figure 3), the OSU mobile research navigation and mapping platform, and the recently upgraded OSU PN prototype (see Figure 4) jointly performed a variety of maneuvers, collecting data from multiple GPS receivers, inertial measurement units (IMUs), imaging sensors, and other devices. Parts of the collected data sets have been used for demonstrating the performance of navigation indoors and in the transition between environments, and it is this aspect of our experiments that will be discussed in the present article.

Figure 3. Land vehicle, OSU GPSVan.

Figure 4. Personal navigator sensor assembly.

The GPSVan was equipped with navigation, tactical, and microelectromechanical systems (MEMS)-grade IMUs, installed in a two-level rigid metal cage, and the signals from two GPS antennas, mounted on the roof, were shared among multiple geodetic-grade dual-frequency GPS receivers. In addition, odometer data were logged, and optical imagery was acquired in some of the tests.

The first PN prototype system, developed in 2006–2007, used GPS, IMU, a digital barometer, a magnetometer compass, a human locomotion model, and 3D active imaging sensor, Flash LIDAR (an imaging light detection and ranging system using rapid laser pulses for subject illumination). Recently, the design was upgraded to include 2D/3D imaging sensors to provide better position and attitude estimates indoors, and to facilitate transition between outdoor and indoor environments. Consequently, the current configuration allows for better distance estimation among platforms, both indoors and outdoors, as well as improving the navigation and tracking performance in general.

The test area where data were acquired to support this study, shown in Figure 5, includes an open parking lot, moderately vegetated passages, a narrow alley between buildings, and a one-storey building for indoor navigation testing. The three typical scenarios used were:
1)   Sensor/platform calibration: GPSVan and PN are connected and navigate together.
2)   Both platforms moved closely together, that is, the GPSVan followed the PN’s trajectory.
3)   Both platforms moved independently.

Image-Based Navigation

The sensor of interest for the study reported here is an image sensor that actually includes two distinct data streams: a standard intensity image and a 3D ranging image, see Figure 6. The unit consists primarily of a 640 × 480 pixel array of infrared detectors. The operational range of the sensor is 0.8–10 meters, with a range resolution of 1 centimeter at a 2-meter distance.

Figure 6. PN captured 3D image sequence from inside the building.

In this study, the image-based navigation (no IMU) was considered. To overcome this limitation, the intensity images acquired simultaneously with the range data by the unit were leveraged to provide crucial information. The two intensity images were processed utilizing the Scale Invariant Feature Transform (SIFT) algorithm to identify matching features between the pair of 2D intensity images.

The SIFT algorithm has been primarily applied to 1D and 2D imagery to date; the authors are not aware of any research efforts to apply SIFT to 3D datasets for the expressed purpose of positioning. Analysis at our laboratory supported well-published results regarding the exceptional performance of SIFT with respect to both repeatability and extraction of the feature content. The algorithm is remarkably robust to most image corruption schema, although white noise above 5 percent does appear to be the primary weakness of the algorithm. The algorithm suffers in three critical areas with respect to providing a 3D positioning solution. First, the algorithm is difficult to scale in terms of the number of descriptive points; that is, the algorithm quickly becomes computationally intractable for a large number (>5,000) of pixels. Secondly, the matching process is not unique; it is exceptionally feasible for the algorithm to match a single point in one image to multiple points in another image. Finally, since the algorithm loses spatial positioning capabilities to achieve the repeatability, the ability to utilize matching features for triangulation or trilateration becomes impaired. Owing to the noted issues, SIFT was not found to be a suitable methodology for real-time positioning based on 3D Flash LIDAR datasets.

Despite these drawbacks, the intensity images offer the only available sensor input beyond the 3D ranging image. As such, the SIFT methodology provides what we believe to be a “best in class” algorithmic approach for matching 2D intensity images. The necessity of leveraging the intensity images will be apparent shortly, as the schema for deriving platform position is explained.

The algorithm has been developed and implemented by the second author (see Further Reading for details). The algorithm utilizes eigenvector “signatures” for point features as a means to facilitate matching. The algorithm is comprised of four steps:
1)   Segmentation
2)   Coordinate frame transformation
3)   Feature matching
4)   Position and orientation determination.

The algorithm utilizes the eigenvector descriptors to merge points likely to belong to a surface and identify the pixels corresponding to transitions between surfaces. Utilizing an initial coarse estimate from the IMU system, the results from the previous frame are transformed into the current coordinate reference frame by means of a Random Sampling Consensus or RANSAC methodology. Matching of static transitional pixels is accomplished by comparing eigenvector “signatures” within a constrained search window. Once matching features are identified and determined to be static, the closed form quaternion solution is utilized to derive the position and orientation of the acquisition device, and the result updates the inertial system in the same manner as a GPS receiver within the common GPS/IMU integration. The algorithm is unique in that the threshold mechanisms at each step are derived from the data itself, rather than relying upon a-priori limits. Since the algorithm only utilizes transitional pixels for matching, a significant reduction in dimensionality is generally accomplished and facilitates implementation on larger data frames.

The key point in this overview is the need to provide coarse positioning information to the 3D matching algorithm to constrain the search space for matching eigenvector signatures. Since the IMU data were not available, the matching SIFT features from the intensity images were correlated with the associated range pixel measurements, and these range measurements were utilized in Horn’s Method (see Further Reading) to provide the coarse adjustment between consecutive range image frames. The 3D-range-matching algorithm described above then proceeds normally.

The use of SIFT to provide the initial matching between the images entails the acceptance of several critical issues, beyond the limitations previously discussed. First, since the SIFT algorithm is matching 2D features on the intensity image; there is no guarantee that the matched features represent static elements in the field of view. As an example, SIFT can easily “match” the logo on a shirt worn by a moving person; since the input data will include the position of non-static elements, the resulting coarse adjustment may possess very large biases (in position). If these biases are significant, constraining the search space may be infeasible, resulting in either the inability to generate eigenvector matches (worst case) or a longer search time (best case). Since the 3D-range-matching algorithm checks the two range images for consistency before the matching process begins, this can be largely mitigated in implementation. Secondly, the SIFT features are located with sub-pixel location, thus the correlation to the range pixel image will inherently possess an error of ± 1 pixel (row and column). The impact of this error is that range pixels utilized to facilitate the coarse adjustment may in fact not be correct; the correct range pixel to be matched may not be the one selected. This will result in larger errors during the initial (coarse) adjustment process. Third, the uncertainty of the coarse adjustment is not known, so a-priori estimates of the error ellipse must be made to establish the eigenvector search space. The size and extent of these error ellipses is not defined on-the-fly by the data, which reduces one of the key elements of the 3D matching algorithm. Fourth, the limited range of the image sensor results in a condition where intensity features have no associated range measurement (the feature is out of range for the range device). This reduces the effective use of SIFT features for coarse alignment. However, using the intensity images does demonstrate the ability of the 3D-range-matching algorithm to generically utilize coarse adjustment information and refine the result to provide a navigation solution.

Data Analysis

In the experiment selected for discussion in this article, initially, the PN was initially riding in the GPSVan. After completing several loops in the parking lot (the upper portion of Figure 5), the PN then departed the vehicle and entered the building (see Figure 7), exited the facility, completed a trajectory around the second building (denoted as “mixed area” in Figure 5), and then returned to the parking lot.

Figure 7. Building used as part of the test trajectory for indoor and transition environment testing; yellow line: nominal personal navigator indoor trajectories; arrows: direction of personal navigator motion inside the building; insert: reconstructed trajectory section, based on 3D image-based navigation.

While minor GPS outages can occur under the canopy of trees, the critical portion of the trajectory is the portion occurring inside the building since the PN platform will be unable to access the GPS signal during this portion of the trajectory. Our efforts are therefore focused on providing alternative methods for positioning to bridge this critical gap.

Utilizing the combined intensity images (for coarse adjustment via SIFT) and the 3D ranging data, a trajectory was derived for travel inside the building at the OSU Supercomputing Facility. There is a finite interval between exiting the building and recovery of GPS signal lock during which the range acquisition was not available; thus the total extent of travel distance during GPS signal outage is not precisely identical to the travel distance where 3D range solutions were utilized for positioning. We estimate the distance from recovery of GPS signals to the last known 3D ranging-derived position to be approximately 3 meters. Based upon this estimate, the travel distance inside the building should be approximately 53.5 meters (forward), 9.5 meters (right), and 0.75 meters (vertical). Based upon these estimates, the total misclosure based upon 3D range-derived positions is provided in Table 1. The asterisk in the third row indicates the estimated nature of these values.

Table 1. Approximate positional results for the OSU Supercomputing Facility trajectory.

The average positional uncertainty reflects the relative, frame-to-frame error reported by the algorithm during the indoor trajectory. This includes both IMU and 3D ranging solutions. The primary reason for the rather large misclosure in the forward and vertical directions is the result of three distinct issues. First, the image ranging sensor has a limited range; during certain portions of the trajectory the sensor is nearly “blind” due to lack of measurable features within the range. During this period, the algorithm must default to the IMU data, which is known to be suspect, as previously discussed. Secondly, the correlation between SIFT features and range measurement pixels can induce errors, as discussed above. Third, the 3D range positions and the IMU data were not integrated in this demonstration; the range positions were used to substitute for the lost GPS signals and the IMU was drifting. Resolving this final issue would, at a minimum, reduce the IMU drift error and improve the overall solution.

A follow-up study conducted at a different facility was completed using the same platform and methodology. In this study, a complete traverse was completed indoors forming a “box” or square trajectory, which returned to the original entrance point. A plot of the trajectory results is provided in Figure 8. The misclosure is less than four meters with respect to both the forward (z) and right (x) directions. While similar issues exist with IMU drift (owing to lack of tight integration with the ranging data), a number of problems between the SIFT feature/range pixel correlation portion of the algorithm are evident; note the large “clumps’ of data points, where the algorithm struggles to reconcile the motions reported by the coarse (SIFT-derived) position and the range-derived position.

Figure 8. Indoor scenario: square (box) trajectory.

Conclusions

As demonstrated in this paper, the determination of position based upon 3D range measurements can be seen to have particular potential benefit for the problem of navigation during periods of operation in GPS-denied environments. The experiment demonstrates several salient points of use in our ongoing research activities. First, the effective measurement range of the sensor is paramount; the trivial (but essential) need to acquire data is critical to success. A major problem was the presence of matching SIFT features but no corresponding range measurement. Second, orientation information is just as critical as position; the lack of this information significantly extended the time required to match features (via eigenvector signatures). Third, there is a critical need for the sensor to scan not only forward (along the trajectory) but also right/left and up/down. Obtaining features in all axes would support efforts to minimize IMU drift, particularly in the vertical. Alternatively, a wider field of view could conceivably accomplish the same objective. Finally, the algorithm was not fully integrated as a substitute for GPS positioning and the IMU was free to drift. Since the 3D ranging algorithm cannot guarantee a solution for all epochs, accurate IMU positioning is critical to bridge these outages. Fully integrating the 3D ranging solution with a GPS/IMU/3D schema would significantly reduce positional errors and misclosure.

Our study indicates that leveraging 3D ranging images to achieve indoor relative (frame-to-frame) positioning shows great promise. The utilization of SIFT to match intensity images was an unfortunate necessity dictated by data availability; the method is technically feasible but our efforts would suggest there are significant drawbacks to this application, both in terms of efficiency and positional accuracy. It would be better to use IMU data with orientation solutions to derive the best possible solution. Our next step is the full integration within the IMU to enable 3D ranging solutions to update the ongoing trajectory, which we believe will reduce the misclosure and provide enhanced solutions supporting autonomous (or semi-autonomous) navigation.

Acknowledgments

This article is based on the paper “Cooperative Navigation in Transitional Environments,” presented at presented at PLANS 2012, the Institute of Electrical and Electronics Engineers / Institute of Navigation Position, Location and Navigation Symposium held in Myrtle Beach, South Carolina, April 23–26, 2012.

Manufacturers

The equipment used for the experiments discussed in this article included a NovAtel Inc. SPAN system consisting of a NovAtel OEMV GPScard, a Honeywell International Inc. HG1700 Ring Laser Gyro IMU, a Microsoft Xbox Kinect 3D imaging sensor, and a Casio Computer Co., Ltd. Exilim EX-H20G Hybrid-GPS digital camera.

DOROTA GREJNER-BRZEZINSKA is a professor and leads the Satellite Positioning and Inertial Navigation (SPIN) Laboratory at OSU, where she received her M.S. and Ph.D. degrees in geodetic science.

J.N. (NIKKI) MARKIEL is a lead geophysical scientist at the National Geospatial-Intelligence Agency. She obtained her Ph.D. in geodetic engineering at OSU.

CHARLES TOTH is a senior research scientist at OSU’s Center for Mapping. He received a Ph.D. in electrical engineering and geoinformation sciences from the Technical University of Budapest, Hungary.

ANDREW ZAYDAK is a Ph.D. candidate in geodetic engineering at OSU.

The Kinematic GPS Challenge: First Gravity Comparison Results

GPS World staff — Wed, 14 Mar 2012 17:53:36 +0000

By Theresa Diehl

The National Geodetic Survey (NGS) has issued a “Kinematic GPS Challenge” to the community in support of NGS’ airborne gravity data collection program, called Gravity for the Redefinition of the American Vertical Datum (GRAV-D). The “Challenge” is meant to provide a unique benchmarking opportunity for the kinematic GPS community by making available two flights of data from GRAV-D’s airborne program for their processing. By comparing the gravity products that are derived from a wide variety of kinematic GPS processing products, a unique quality assessment is possible.

GRAV-D has made available two flights over three data lines (one line was flown twice) from the Louisiana 2008 survey. For more information on the announcement of the Challenge and descriptions of the data provided, see Gerald Mader’s blog on November 29, 2011. The GRAV-D program routinely operates at long-baselines (up to 600 km), high altitudes (20,000 to 35,000 ft), and high speeds (up to 280 knots), a challenging data set from a GPS perspective. As of December 2011, ten groups of kinematic GPS processors have provided a total of sixteen position solutions for each flight. At two data lines per flight, this yielded 64 total position solutions. Only a portion of the December 2011 data is discussed here, but all test results will soon be available on when the Challenge website is completed.

Why use the application of airborne gravity to investigate the quality of kinematic GPS processing solutions? Because the gravity measurement itself is an acceleration, which is being recorded with a sensor on a moving platform, inside a moving aircraft, in a rotating reference frame (the Earth). The gravity results are completely reliant on our ability to calculate the motion of the aircraft— position, velocity, and acceleration. These values are used in several corrections that must be applied to the raw gravimeter measurement in order to recover a gravity value (Table 1). The corrections in Table 1 are simplified to assume that the GPS antenna and gravimeter sensor are co-located horizontally and offset vertically by a constant, known distance.

Table 1. GPS-Derived Values that are used in the Calculation of Free-Air Gravity Disturbances

All Challenge solutions are presented anonymously here, with f## designations. For each flight of data, the software that made the f01 solution is the same as for f16, f02 the same as f17, and so on.

Test #1: Are the solutions precise and accurate?

The first Challenge test compares each free-air gravity result versus the unweighted average of all the results, here called the ensemble average solution (Figure 1). This comparison highlights any GPS solutions whose gravity result is significantly different from the others, and will group together solutions that are similar to each other (precise). Precision is easy to test this way, but in order to tell which gravity results are accurate calculations of the gravity field, a “truth” solution is necessary. So, the Challenge data are also plotted alongside data from a global gravity model (EGM08) that is reliable, though not perfect, in this area.

Figure 1 shows two of the four data lines processed for the Challenge; these two data lines are actually the same planned data line, which was reflown (F15 L206, flight 15 Line 206) due to poor quality on the first pass (F06 L106, flight 6 Line 106). The 5-10 mGal amplitude spikes of medium frequency along L106 are due to turbulence experienced by the aircraft, turbulence that the GPS and gravity processing could not remove from the gravity signal.

Figure 1.

Figure 2.

Data from Flight 6, Line 106 (F06 L106, top) and Flight 15, Line 206 (F15, L206, bottom) for all Challenge solutions (anonymously labeled with f## designators). Figures 1 and 2. Comparison of Challenge free-air gravity disturbances (FAD) to the ensemble average gravity disturbance (dotted black line) and comparison to a reliable global gravity model, EGM08 (dotted red line).

Figure 3.

Figure 4.

Figures 3 and 4. Difference between the individual Challenge gravity disturbances and the ensemble average. The thin black lines mark the 2-standard deviation levels for the differences. For F15 L206, one solution (f23) was removed from the difference plot and statistics because it was an outlier. For both lines, the ensemble’s difference with EGM08 is not plotted because it is too large to fit easily on the plot.

The results of test #1 are surprising in several ways:

The data using the PPP technique (precise point positioning, which uses no base station data) and the data using the differential technique (which uses base stations) produce equivalent gravity data results, where any differences between the methods are virtually indistinguishable.
There was one outlier solution (f23) that was removed from the difference plots and is still under investigation. Also, on F15 L206, solution f28 had an unusually large difference from the average though it performed predictably on the other lines. Of the remaining solutions, four solutions stand out as the most different from all the others: f03/f18, f04/f19, f05/f20, and f07/f22.
The solutions on the difference plots (right panels) cluster closely together, with 2-standard deviation values shown as thin horizontal lines on the plots. The Challenge solutions meet the precision requirements for the GRAV-D program: +/- 1 mGal for 2-standard deviations.
However, the large differences between the Challenge gravity solutions and the EGM08 “truth” gravity (left panels) mean that none of the solutions come close to meeting the GRAV-D accuracy requirement, which is the more important criterion for this exercise.

Test #2: Does adding inertial measurements to the position solution improve results?

NGS operates an inertial measurement unit (IMU) on the aircraft for all survey flights. The IMU records the aircraft’s orientation (pitch, roll, yaw, and heading). Including the orientation information in the calculation of the position solution should yield a better position solution than GPS-only calculations, but it was not expected to be significantly better. Figure 2 shows the NGS best loosely-coupled GPS/IMU free-air gravity result versus the Challenge GPS-only results and Table 2 shows the related statistics.

Figure 5.

Figure 6.

Figures 5 and 6. F06 L105. (Figure 5) Comparison of Challenge FAD gravity solutions (ensemble=black dotted line) with EGM08 (red dotted line); (Figure 6) comparison of Challenge gravity solutions (all GPS-only; ensemble=black dotted line) with NGS’ coupled GPS/IMU gravity solution (red dotted line).

Table 2. Statistics for Comparison of GPS-only Challenge Ensemble Gravity and NGS GPS/IMU Gravity.

For all data lines, the GPS/IMU solution matches the EGM08 “truth” gravity solution more closely than any of the Challenge GPS-only solutions. In fact, the more motion that is experienced by the aircraft, the more that adding IMU information improves the solution. One conclusion from this test is that IMU data coupled with GPS data is a requirement, not optional, in order to obtain the best free-air gravity solutions.

Additional Testing and Future Research

Other testing has already been completed on the Challenge data and the results will be available on the Challenge website soon. Important results are:

Two Challenge participants’ solutions perform better than the rest, two perform worse, and one is a low quality outlier. The reasons for these differences are still under investigation.
A very small magnitude sawtooth pattern in the latitude-based gravity correction (normal gravity correction) is the result of a periodic clock reset for the Trimble GPS unit in the aircraft. This clock reset is uncorrected in the majority of Challenge solutions. The clock reset causes an instantaneous small change in apparent position, which results in a 1-2 mGal magnitude unreal spike in the gravity tilt correction at each epoch with a clock reset.
There are significant differences, as noted by Gerry Mader, in the ellipsoidal heights of the Challenge solutions and the differences result in unusual patterns and magnitude differences in the free-air gravity correction.

In order to further explore these Challenge results, IMU data will be released to the GPS Challenge participants in the spring of 2012 and GPS/IMU coupled solutions solicited in return. Additionally, basic information about the Challenge participants’ software and calculation methodologies will be collected and will form the basis of the benchmarking study.

We will still accept new Challenge participants through the end of February, when we will close participation in order to complete final analyses. Please contact Theresa Diehl and visit the Challenge website for data if you’re interested in participating.

Innovation: Filling in the Gaps: Improving Navigation Continuity Using Parallel Cascade Identification

Richard Langley — Sat, 01 Oct 2011 01:41:12 +0000

INNOVATION INSIGHTS with Richard Langley

THREE, TWO, ONE, ZERO! Can you still navigate with just a GPS receiver when the number of tracked GPS satellites drops from four to none? As we know, pseu- doranges from a minimum of four satellites, preferably well spaced out in the sky, are required for three-dimensional positioning. That’s because there are four unknowns to estimate: the three position coordinates (latitude, longitude, and height) and the offset of the receiver clock from GPS System Time. If we had a stable clock in the receiver, then we could hold the clock offset constant and have 3D navigation with just three satellites. But for every 3 nanoseconds of clock drift, we will have about 1 meter of pseudorange error, which will lead to several meters of position error depend- ing on the receiver-satellite geometry. On the other hand, we could hold the height coor- dinate constant (viable for navigation over slowly changing topography or at sea) and estimate the horizontal coordinates and the receiver clock offset. So far, so good.

What if the number of tracked satellites then drops to two? We can now only esti- mate two unknowns. They could be the two horizontal coordinates, if we hold both the receiver clock offset and the height fixed. Any errors in those fixed values will propagate into the estimated horizontal coordinates but the resulting position errors might still be acceptable. Instead of holding the clock offset
fixed, we could assume a constant heading and compute the position along the assumed trajectory. However, navigation will rapidly deteriorate as soon as we make the first turn. And one satellite? We would have to make assumptions about the vehicle trajectory, the height, and the clock offset, with likely very poor results. And with no satellites? We might be able to navigate over a short period of time by “dead reckoning,” assuming a constant trajectory and speed, but the resulting positions will be educated guesses at best.

Clearly, if we want to reliably navigate with fewer than four satellites we need to augment the GPS pseudoranges with measurements from some other sensors. A common approach is to use inertial measurement units or IMUs. A complete IMU consists of three accelerometers and three gyroscopes, and small, cost-effective microelectromechanical IMUs are readily available. For land navigation, however, it can be advantageous to use a reduced inertial sensor system or RISS, consisting of one single-axis gyroscope, two accelerometers, and the vehicle speedometer. We can also make use of GPS pseudorange rates along with the pseudoranges.

But what is the best way to combine the RISS measurements with the GPS measurements? The classic approach is to integrate the measurements in a conventional tightly coupled Kalman filter. But in this month’s column, we look at how a mathematical procedure called parallel cascade identification can improve the Kalman filter’s job, when navigating with three, two, or even one GPS satellite.

The Global Positioning System meets the requirements for numerous navigational applications when there is direct line-of-sight (LOS) to four or more GPS satellites. Vehicular navigation systems and personal positioning systems may suffer from satellite signal blockage as LOS to at least four satellites may not be readily available when operating in urban landscapes with high buildings, underpasses, and tunnels, or in the countryside with thick forested areas. In such environments, a typical GPS receiver will have difficulties attaining and maintaining signal tracking, which causes GPS outages resulting in degraded or non-existent positioning information. Due to these well-known limitations of GPS, multi-sensor system integration is often employed. By integrating GPS with complementary motion sensors, a solution can be obtained that is often more accurate than that of GPS alone.

Microelectromechanical systems (MEMS) inertial sensors have enabled production of lower-cost and smaller-size inertial measurement units (IMUs) with little power consumption. A complete IMU is composed of three accelerometers and three gyroscopes. These MEMS sensors have composite error characteristics that are stochastic in nature and difficult to model. In traditional low-cost MEMS-based IMU/GPS integration, a few minutes of degraded GPS signals can cause position errors, which can reach several hundred meters. For full 3D land vehicle navigation, we had earlier proposed a low-cost MEMS-based reduced inertial sensor system (RISS) based on only one single-axis gyroscope, two accelerometers, and the vehicle odometer, and we have integrated this system with GPS. RISS mitigates several error sources in the full-IMU solution; moreover, RISS reduces system cost further due to the use of fewer sensors. Another enhancement can be achieved by using tightly coupled integration, which can provide GPS aiding for a navigation solution when the number of visible satellites is three or lower, removing the foremost requirement of loosely coupled integration, which is visibility of at least four satellites. GPS aiding during limited GPS satellite availability can improve the operation of the navigation system for tightly coupled systems. Thus, in our reseach, a Kalman filter (KF) is used to integrate low-cost MEMS-based RISS with GPS in a tightly coupled scheme.

The KF employed in tightly coupled RISS/GPS integration utilizes pseudoranges and pseudorange rates measured by the GPS receiver. The accuracy of the position estimates is highly dependent on the accuracy of the range measurements. The tightly coupled solutions presented in the literature assume that the pseudorange measurement, after correcting for ionospheric and tropospheric delays, satellite clock errors, and ephemeris errors, only have errors due to the receiver clock and white noise. These latter two are the only errors modeled inside the measurement model in the tightly coupled solutions presented in the literature. Experimental investigation of the GPS pseudoranges for vehicle trajectories in different areas and for different scenarios showed that, in addition, there are residual correlated errors. Since it has been experimentally proven that there are residual correlated errors in addition to white noise and receiver clock errors, we have proposed using a nonlinear system identification technique called parallel cascade identification (PCI) to model such correlated errors in pseudorange measurements.

We propose that the PCI model for the residual pseudorange errors be cascaded with a KF since this nonlinear model cannot be included inside the KF measurement model. The normal operation of a KF is as follows: the difference between the measured pseudorange and pseudorange rate from the mth GPS satellite and the corresponding RISS-predicted estimates of pseudorange and pseudorange rate are used as a measurement update for the KF integration, which computes the estimated RISS errors and corrects the mechanization output. We propose the use of a PCI module, where the role of PCI is to model the pseudorange residual errors. When GPS is available, estimated full 3D position, velocity, and attitude are obtained by integrating the MEMS-based RISS with GPS. In parallel, as a background routine, a PCI model is built for each visible satellite to model its pseudorange error. The actual pseudorange of each visible satellite is used as the input to the model; the target output is the error between the corrected pseudorange (calculated based on corrected receiver position from the integrated solution) and the actual pseudorange. This target output provides the reference output to build the PCI model of the pseudorange residual error. Dynamic characteristics between system input and output help to identify a nonlinear PCI model and the algorithm can then achieve a residual pseudorange error model.

When fewer than four satellites are visible, the identified parallel cascades for the remaining visible satellites will be used to predict the pseudorange errors for these satellites and correct the pseudorange values to be provided to the KF. This improvement of pseudorange measurements will result in a more accurate aiding for RISS, and thus more accurate estimates of position and velocities.

We examined the performance of the proposed technique by conducting road tests with real-life trajectories using a low-cost MEMS-based RISS. The ultimate check for the proposed system’s accuracy is during GPS signal degradation and blockage. This work presents both downtown scenarios with natural GPS degradation and scenarios with simulated GPS outages where the number of visible satellites was varied from three to zero. The results are examined and compared with KF-only tightly coupled RISS/GPS integration without pseudorange correction using a PCI module. This comparison clearly demonstrates the advantage of using a PCI module for correcting pseudoranges for tightly coupled integration.

RISS/GPS Integration

Earlier, we proposed the reduced inertial sensor system to reduce the overall cost of a positioning system for land vehicles without appreciable performance compromise depending on the fact that land vehicles mostly stay in the horizontal plane. It is the gyroscope technology that contributes the most both to the overall cost of an IMU and to the performance of the INS. In RISS mechanization, the heading (azimuth) angle is obtained by integrating the gyroscope measurement, ω_z. Since this measurement includes the component of the Earth’s rotation as well as rotation of the local level frame on the Earth’s curved surface, these quantities are removed from the measurement before integration. Assuming relatively small pitch and roll angles for land vehicle applications, we can write the rate of change of the azimuth angle directly in the local level frame as:
(1)
where ω^e is the Earth’s rotation rate, φ is the latitude, v_e is the east velocity of the vehicle, h is the altitude of the vehicle and R_N is the normal (prime vertical) radius of curvature of the vehicle’s position on the reference ellipsoid.

The two horizontal accelerometers can be employed for obtaining the pitch and roll angles of the vehicle. Thus, a 3D navigation solution can be achieved to boost the performance of the solution. When the vehicle is moving, the forward accelerometer measures the forward vehicle acceleration as well as the component due to gravity, g. To calculate the pitch angle, the vehicle acceleration derived from the odometer measurements, a_od, is removed from the forward accelerometer measurements, f_y. Consequently, the pitch angle is computed as:

(2)

Similarly, the transversal accelerometer measures the normal component of the vehicle acceleration as well as the component due to gravity. Thus, to calculate the roll angle, the transversal accelerometer measurement, f_x, must be compensated for the normal component of acceleration. The roll angle is then given by:

(3)

The computed azimuth and pitch angles allow the transformation of the vehicle’s speed along the forward direction, v_od (obtained from the odometer measurements) to east, north, and up velocities (v_e, v_n, and v_u respectively) as follows:
(4)
where is the rotation matrix that transforms velocities in the vehicle body frame to the navigation frame. The east and north velocities are transformed and integrated to obtain position in geodetic coordinates (latitude, φ, and longitude, λ). The vertical component of velocity is integrated to obtain altitude, h. The following equation shows these operations:
(5)

where, in addition to the terms already defined, RM is the meridional radius of curvature. We have used the KF as the estimation technique for tightly coupled RISS/GPS integration in our approach. KF is an optimal estimation tool that provides a sequential recursive algorithm for the optimal least mean variance (LMV) estimation of the system states. In addition to its benefits as an optimal estimator, the KF provides real-time statistical data related to the estimation accuracy of the system states, which is very useful for quantitative error analysis. The filter generates its own error analysis with the computation of the error covariance matrix, which gives an indication of the estimation accuracy.

In tightly coupled RISS/GPS system architecture, instead of using the position and velocity solution from the GPS receiver as input for the fusion algorithm, raw GPS observations such as pseudoranges and Doppler shifts are used. The range measurement is usually known as a pseudorange due to the contamination of errors, such as atmospheric errors, as well as synchronization errors between the satellite and receiver clocks.

After correcting for the satellite clock error and the ionospheric and tropospheric errors, we can obtain a corrected pseudorange. The receiver clock error and the residual errors remaining in the corrected pseudorange, assumed as white Gaussian noise, are the only errors modeled inside the measurement model in the tightly coupled solutions presented in the literature. Experimental investigation of the GPS pseudoranges in trajectories in different areas and under different scenarios showed that the residual errors are not just white noise as assumed in the literature, but, in fact, are correlated errors. As the GPS observables are used to update the KF, a technique must be developed to adequately model these errors to improve the overall performance of the KF. We propose using PCI to model these correlated errors. A PCI module models these errors, and then provides corrections prior to sending the GPS pseudoranges to aid the KF during periods of GPS partial outages (when the number of visible satellites drops below four).

Parallel Cascade Identification

What is PCI? System identification is a procedure for inferring the dynamic characteristics between system input and output from an analysis of time-varying input-output data. Most of the techniques assume linearity due to the simplicity of analysis since nonlinear techniques make analysis much more complicated and difficult to implement than for the linear case. However, for more realistic dynamic characterization nonlinear techniques are suggested. PCI is a nonlinear system identification technique proposed by one of us [MJK]. This technique models the input/output behavior of a nonlinear system by a sum of parallel cascades of alternating dynamic linear (L) and static nonlinear (N) elements. The parallel array shown in Figure 1 can be built up one cascade at a time.

Figure 1. Block diagram of parallel cascade identification.

It has been proven that any discrete-time Volterra series with finite memory and anticipation can be uniformly approximated by a finite sum of parallel LNL cascades, where the static nonlinearities, N, are exponentials and logarithmic functions. [A Volterra series, named after the Italian mathematician and physicist Vito Volterra, is similar to the more familiar infinite Taylor series expansion of a function used, for example, in systems analysis, but the Volterra series can include system “memory” effects.] It has been shown that any discrete-time doubly finite (finite memory and order) Volterra series can be exactly represented by a finite sum of LN cascades where the N are polynomials. A key advantage of this technique is that it is not dependent on a white or Gaussian input, but the identified individual L and N elements may vary depending on the statistical properties of the input chosen. The cascades can be found one at a time and nonlinearities in the models are localized in static functions. This reduces the computation as higher order nonlinearities are approximated using higher degree polynomials in the cascades rather than higher order kernels in a Volterra series approximation.

The method begins by approximating the nonlinear system by a first such cascade. The residual (that is, the difference between the system and the cascade outputs) is treated as the output of a new nonlinear system, and a second cascade is found to approximate the latter system, and thus the parallel array can be built up one cascade at a time. Let y_k(n) be the residual after fitting the kth cascade, with y_o(n) = y(n). Let z_k(n) be the output of the kth cascade, so
(6)
where k = 1, 2, …

The dynamic linear elements in the cascades can be determined in a number of ways. The method we have employed uses cross correlations of the input with the current residual. Best fitting of the current residuals was used to find the polynomial coefficients of the static nonlinearities. These resulting cascades are such that they drive the cross-correlations of the input with the residuals to zero. However, a few basic parameters have to be specified in order to identify a parallel cascade model, including the memory length of the dynamic linear element that begins each cascade, the degree of the polynomial static nonlinearity that follows the linear element (this polynomial is best fit to minimize the mean-square error (MSE) of the approximation of the residual), the maximum number of cascades allowable in the model, and a threshold based on a standard correlation test for determining whether a cascade’s reduction of the MSE justifies its addition to the model.

Augmented Kalman Filter

In the previous section, the parallel cascade model was briefly presented, together with a simple method for building up the model to approximate the behavior of a dynamic nonlinear system, given only its input and output. In order to apply PCI to 3D RISS/GPS integration, we propose the use of a KF-PCI module, where the role of PCI is to model the residual errors of GPS pseudoranges.

In full GPS coverage when four or more satellites are available to the GPS receiver, the KF integrated solution provides an adequate position benefiting from both GPS and RISS readings, and the PCI builds the model for the pseudorange errors for each visible satellite. The input of each PCI module is the pseudorange of the visible mth GPS satellite, and the reference output is the difference between the observed pseudorange and the estimated pseudorange from the corrected navigation solution.

The reference output has no corrections during full GPS coverage. It is only used to build the PCI model. Dynamic characteristics between system input and output help to achieve a residual pseudorange error model as shown in the Figure 2.

Figure 2. Block diagram of augmented KF-PCI module for pseudoranges during GPS availability.

During partial GPS coverage, when there are fewer than four satellites available, the PCI modules for all satellites cease training, and the available PCI model for each visible satellite is used to predict the corresponding residual pseudorange errors, as shown in Figure 3. The KF operates as usual, but in this instance the GPS observed pseudorange is corrected by the output of the corresponding PCI. The pre-built PCI models, only for the visible satellites during the partial outage, predict the corresponding residual pseudorange errors to obtain a correction. Thus, the corrected pseudorange can then be obtained.

During a full GPS outage, when no satellites are available, the KF operates in prediction mode and the PCI modules neither provide corrections nor operate in training mode.

FIGURE 3 Block diagram of augmented KF-PCI module for pseudoranges during limited availability of GPS.

Experimental Setup

The performance of the developed navigation solution was examined with road test experiments in a land vehicle. The experimental data collection was carried out using a full-size passenger van carrying a suite of measurement equipment that included inertial sensors, GPS receivers, antennae, and computers to control the instruments and acquire the data as shown in the Figure 4. The inertial sensors used in our tests are packaged in a MEMS-grade IMU. The specifications of the IMU are listed in Table 1.

Figure 4. Data collection equipment mounted inside the road test vehicle.

Table 1. IMU specifications.

The vehicle’s forward speed readings were obtained from vehicle built-in sensors through the On-Board Diagnostics version II (OBD II) interface. The sample rate for the collection of speed readings was 1 Hz. The GPS receiver used in our integrated system was a high-end dual-frequency unit. Our results were evaluated with respect to a reference solution determined by a system consisting of another receiver of the same type, integrated with a tactical grade IMU.

This system provided the reference solution to validate the proposed method and to examine the overall performance during simulated GPS outages.
Several road test trajectories were carried out using the setup described above. The road test trajectory considered for this article was performed in the city of Kingston, Ontario, Canada, and is shown in Figure 5. This road test was performed for nearly 48 minutes of continuous vehicle navigation and a distance of around 22 kilometers. Ten simulated GPS outages of 60 seconds each were introduced in post-processing (they are shown as blue circles overlaid on the map in Figure 5) during good GPS availability. The trajectory was run four times with the simulated partial outages having three, two, one, and zero visible satellites, respectively. The case with no satellites seen is like a scenario that would occur in loosely coupled integration. The errors estimated by KF-PCI and KF-only solutions were evaluated with respect to the reference solution.

Figure 5. Trajectory of the test conducted in the city of Kingston.

Experimental Results

The results in Figure 6 and Figure 7 demonstrate the benefits of the proposed PCI module. The main benefit of using PCI for pseudorange correction is the modeling capability, which enables correction of the raw GPS measurements. The benefit of more satellite availability can also be seen from these results. Figures 6 and 7 clearly show that both the average maximum position error and the average root-mean-square (RMS) position error are lower with the KF-PCI approach compared to the conventional KF, even when data from only one satellite is available.

Figure 6. Bar graph showing average maximum positional errors for all outages.

Figure 7. Bar graph for RMS positional errors for all outages.

To gain more insight about the performance of the proposed technique to enhance the aiding of the KF by correcting the pseudoranges, we can look at the results of KF-PCI and KF approaches with different numbers of satellites visible to the receiver for one of the artificial outages. Figure 8 shows a map featuring the different compared solutions during outage number 8. Eight solutions are presented for the cases of three, two, one, and zero satellites observed for the standard KF and KF with PCI. To get some idea of the vehicle dynamics during this outage, we can examine Figure 9, which depicts the forward speed of the vehicle as well as its azimuth angle as obtained from the reference solution. There is a significant variation in speed, with only a small variation in azimuth.

Figure 8. Performance of tightly coupled 3D-RISS during outage #8.

Figure 9. Vehicle dynamics (speed and azimuth) during GPS outage #8.

Figure 10 illustrates the performance differences between the KF-PCI and KF-only solutions for different numbers of satellites for this outage. Similar to Figure 7, Figure 10 shows the average RMS position differences between the KF-PCI and KF-only solutions and the reference solution (without the artificial outages). While the differences increase as the number of available satellites decreases, the accuracies may still be acceptable for many navigation purposes.

And while the differences between the KF-PCI and KF-only approaches for this particular outage are small, the KF-PCI approach consistently provides better accuracy.

Figure 10. Performance of PCI-KF (shades of blue for different number of satellites) and KF (shades of green for different number of satellites) of tightly coupled 3D-RISS during outage #8.

Conclusion

In this article, we have described a novel design for a navigation system that augments a tightly coupled KF system with PCI modules using low-cost MEMS-based 3D RISS and GPS observations to produce an integrated positioning solution. A PCI module is built for each satellite during good signal availability where the integrated solution presents a good position estimate. The output of each PCI module provides corrections to the GPS pseudoranges of the corresponding visible satellite during GPS partial outages, thereby decreasing residual errors in the GPS observations. This KF-PCI module was tested with real road-test trajectories and compared to a KF-only approach and was shown to improve the overall maximum position error during GPS partial outages.

Future work with PCI for modeling the residual pseudorange errors will consider replacing the KF with a particle filter to provide more robust integration and a consistent level of accuracy.

Acknowledgments

The research discussed in this article was supported, in part, by grants from the Natural Sciences and Engineering Research Council of Canada, the Geomatics for Informed Decisions (GEOIDE) Network of Centres of Excellence, and Defence Research and Development Canada. The equipment was acquired by research funds from the Directorate of Technical Airworthiness and Engineering Support, the Canada Foundation for Innovation, the Ontario Innovation Trust, and the Royal Military College of Canada. The article is based on the paper “Modeling Residual Errors of GPS Pseudoranges by Augmenting Kalman Filter with PCI for Tightly Coupled RISS/GPS Integration” presented at ION GNSS 2010, the 23rd International Technical Meeting of the Satellite Division of The Institute of Navigation held in Portland, Oregon, September 21–24, 2010.

Manufacturers

The test discussed in this article used a NovAtel Inc. OEM4 dual-frequency GPS receiver and a Crossbow Technology Inc., now Moog Crossbow IMU300CC-100 MEMS-grade IMU. The On-Board Diagnostics data was accessed with a Davis Instruments CarChip Pro data logger. The reference solutions were provided by a NovAtel G2 Pro-Pack SPAN unit, interfacing a NovAtel OEM4 receiver with a Honeywell HG1700 tactical grade IMU.

Umar Iqbal is a doctoral candidate at Queen’s University, Kingston, Ontario, Canada. He received a master’s of electrical engineering degree in integrated positioning and navigation systems from Royal Military College (RMC) of Canada, Kingston, in 2008. He also holds an M.Sc. in electronics engineering from the Ghulam Ishaq Khan Institute of Engineering Sciences and Technology, Topi, Pakistan, and a B.Sc. in electrical engineering from the University of Engineering and Technology, Lahore, Pakistan. His research focuses on the development of enhanced performance navigation and guidance systems that can be used in several applications including car navigation.

Jacques Georgy received his Ph.D. degree in electrical and computer engineering from Queen’s University in 2010. He received B.Sc. and M.Sc. degrees in computer and systems engineering from Ain Shams University, Cairo, Egypt, in 2001 and 2007, respectively. He is working in positioning and navigation systems for vehicular, machinery, and portable applications with Trusted Positioning Inc., Calgary, Alberta, Canada. He is also a member of the Navigation and Instrumentation Research Group at RMC. His research interests include linear and nonlinear state estimation, positioning and navigation by inertial navigation system/global positioning system integration, autonomous mobile robot navigation, and underwater target tracking.

Michael J. Korenberg is a professor in the Department of Electrical and Computer Engineering at Queen’s University. He holds an M.Sc. (mathematics) and a Ph.D. (electrical engineering) from McGill University, Montreal, Quebec, Canada, and has published extensively in the areas of nonlinear system identification and time-series analysis.

Aboelmagd Noureldin is a cross-appointment associate professor with the Department of Electrical and Computer Engineering at Queen’s University and the Department of Electrical and Computer Engineering at RMC. He is also the founder and leader of the Navigation and Instrumentation Research Group at RMC. He received the B.Sc. degree in electrical engineering and the M.Sc. degree in engineering physics from Cairo University, Giza, Egypt, in 1993 and 1997, respectively, and the Ph.D. degree in electrical and computer engineering from The University of Calgary, Calgary, Alberta, Canada, in 2002. His research is related to artificial intelligence, digital signal processing, spectral estimation and de-noising, wavelet multiresolution analysis, and adaptive filtering, with emphasis on their applications in mobile multisensor system integration for navigation and positioning technologies.

Innovation: The Right Attitude: Experimenting with GPS on Board High-Altitude Balloons

GPS World staff — Thu, 01 Sep 2011 02:54:12 +0000

In this month’s column, we look at how a team of Dutch and Japanese researchers is using GPS to determine the attitude of a payload launched from a high-altitude balloon.

By Peter J. Buist, Sandra Verhagen, Tatsuaki Hashimoto, Shujiro Sawai, Shin-Ichiro Sakai, Nobutaka Bando, and Shigehito Shimizu

INNOVATION INSIGHTS with Richard Langley

It is not widely recognized that relative or differential positioning using GNSS carrier-phase measurements is an interferometric technique. In interferometry, the difference in the phase of an electromagnetic wave at two locations is precisely measured as a function of time. The phase differences depend, amongst other factors, on the length and orientation of the baseline connecting the two locations. The classic demonstration of interferometry, showing that light could be interpreted as a wave phenomenon, was the 1803 double-slit experiment of the English polymath, Thomas Young. Many of us recreated the experiment in high school or university physics classes. A collimated beam of light is shone through two small holes or narrow slits in a barrier placed between the light source and a screen. Alternating light and dark bands are seen on the screen. The bands are called interference fringes and result from the waves emanating from the two slits constructively and destructively interfering with each other. The colors seen on the surface of an audio CD, the colors of soap film, and those of peacock feathers and the wings of the Morpho butterfly are all examples of interference.

Interference fringes also reveal information about the source of the waves. In 1920, the American Nobel-prize-winning physicist, Albert Michelson, used an interferometer attached to a large telescope to measure the diameter of the star Betelgeuse. Radio astronomers extended the concept to radio wavelengths, using two antennas connected to a receiver by cables or a microwave link. Such radio interferometers were used to study the structure of various radio sources including the sun. Using atomic frequency standards and magnetic tape recording, astronomers were able to sever the real-time links between the antennas, giving birth to very long baseline interferometry (VLBI) in 1967. The astronomers used VLBI to study extremely compact radio sources such as the enigmatic quasars. But geodesists realized that high resolution VLBI could also be used to determine — very precisely — the components of the baseline connecting the antennas, even if they were on separate continents.

That early work in geodetic VLBI led to the concept developed by Charles Counselman III and others at the Massachusetts Institute of Technology in the late 1970s of recording the carrier phase of GPS signals with two separate receivers and then differencing the phases to create an observable from which the components of the baseline connecting the receivers’ antennas could be determined. This has become the standard high-precision GPS surveying technique. Later, others took the concept and applied it to short baselines on a moving platform allowing the attitude of the platform to be determined.

In this month’s column, we look at how a team of Dutch and Japanese researchers is using GPS to determine the attitude of a payload launched from a high-altitude balloon.

The Japan Aerospace Exploration Agency (JAXA) is developing a system to provide a high-quality, long duration microgravity environment using a capsule that can be released from a high-altitude balloon. Since 1981, an average of 100 million dollars is spent every year on microgravity research by space agencies in the United States, Europe, and Japan. There are many ways to achieve microgravity conditions such as (in order of experiment duration) drop towers, parabolic flights, balloon drops, sounding rockets, the Space Shuttle (unfortunately, no longer), recoverable satellites, and the International Space Station. The order of those options is also approximately the order of increasing experiment cost, with the exception of the balloon drop. Besides being cost-efficient, a balloon-based system has the advantage that no large acceleration is required before the experiment can be performed, which could be important for any delicate equipment that is carried aloft.

In this article, we will describe JAXA’s Balloon-based Operation Vehicle (BOV) and the experiments carried out in cooperation with Delft University of Technology (DUT) using GPS on the gondola of the balloon in 2008 (single baseline estimation) and 2009 (full attitude determination and relative positioning). The attitude calculated using observations from the onboard GPS receiver during the 2009 experiment is compared with that from sun and geomagnetic sensors as well as that provided by the GPS receiver itself.

Nowadays, GNSS is used for absolute and relative positioning of aircraft and spacecraft as well as determination of their attitude. What these applications have in common is that, in general, the orientation of the platform is changing relatively slowly and, to a large extent, predictably. Here, we will discuss a balloon-based application where the orientation of the platform, at times, varies very dynamically and unpredictably.

Balloon Experiments

Scientific balloons have been launched in Japan by the Institute of Space and Astronautical Science (ISAS), now a division of JAXA, since 1965, and it holds the world record for the highest altitude reached by a balloon — 53 kilometers. Recently, balloon launches have taken place from the Multipurpose Aviation Park (MAP) in Taiki on the Japanese island of Hokkaido. The balloons are launched using a so-called sliding launcher. The sliding launcher and the hanger at MAP are shown in FIGURE 1.

Figure 1. Takeoff of the Balloon-based Operation Vehicle (BOV) 2009 experiment. The BOV and the gondola hanging from the sliding launcher can be seen to the left, while the balloon can be seen in front of the hanger on the right.

Balloon-Based Operation Vehicle. As previously mentioned, JAXA’s BOV has been designed for microgravity research. The scenario of a microgravity experiment is illustrated in FIGURE 2. The vehicle is launched with a balloon, which carries it to an altitude of more than 40 kilometers, where it is released.

Figure 2. Microgravity experiment procedure.

After separation, the BOV is in free fall until the parachute is released so that the vehicle can make a controlled landing in the sea. The BOV is recovered by helicopter and can be reused. The capsule has a double-shell drag-free structure and it is controlled so as not to collide with the inner shell. The flight capsule, shown hanging at the sliding launcher in Figure 1, consists of a capsule body (the outer shell), an experiment module (the inner shell), and a propulsion system. The inner capsule shown in FIGURE 3 is kept in free-falling condition after release of the BOV from the balloon, and no disturbance force acts on this shell and the microgravity experiment it contains.

Figure 3. Balloon-based Operation Vehicle overview.

The outer shell has a rocket shape to reduce aerodynamic disturbances. The distance between the outer and inner shells is measured using four laser range sensors. Besides the attitude of the BOV, the propulsion system controls the outer shell so that it does not collide with the inner s
hell. The propulsion system uses 16 dry-air gas-jet thrusters of 60 newtons, each controlling it not only in the vertical direction but also in the horizontal direction to compensate disturbances from, for example, wind.

Flight experiments with the BOV were carried out in 2006 (BOV1) and in 2007 (BOV2), when a fine microgravity environment was established successfully for more than 7 and 30 seconds, respectively.

Attitude Determination. Balloon experiments are performed for a large number of applications, some of which require attitude control. Observations from balloon-based telescopes are an example of an application in which stratospheric balloons have to carry payloads of hundreds of kilograms to an altitude of more than 30 kilometers to be reasonably free of atmospheric disturbances. In this application, the typical requirement for the control of the azimuth angle of the platform is to within 0.1 degrees.

JAXA is developing the Attitude Determination Package (ADP, see TABLE 1) for a future version of the BOV, which contains Sun Aspect Sensors (SAS), the Geomagnetic Aspect Sensor (GAS), an inclinometer, and a gyroscope. Each SAS determines the attitude with a resolution of one degree around one axis and the ADP has four of these sensors pointing in different directions. Inherently, this type of sensor can only provide attitude information if the sun is within the field of view of the sensor. The GAS also determines one-axis attitude. The resolution of magnetic flux density measured by the GAS and applied to obtain an attitude estimate is 50 parts per million. This results in an attitude determination accuracy of the GAS of 1.5 degrees with dynamic bias compensation. The inclinometer determines two-axis attitude with a resolution of 0.2 degrees.

Table 1. Sensor specifications.

Background GPS Experiment. DUT is involved in a precise GPS-based relative positioning and attitude determination experiment onboard the BOV and the gondola of the balloon. Not only is the BOV a challenging environment, but so is the gondola itself, because of the rather rapidly varying attitude (due to wind and — especially at takeoff and separation — rotation) and the high altitude. For a GPS experiment, the altitude of around 40 kilometers is interesting as not many experiments have been performed at this height, which is higher than the altitude reachable by most aircraft but below the low earth orbits for spacecraft. An altitude of about 40 kilometers is a harsh environment for electrical devices because the pressure is about 1/1000 of an atmosphere and the temperature ranges from –60 to 0 degrees Celsius. Furthermore, the antennas are placed under the balloon, which affects the received GPS signals. Later on, we will describe in detail two experiments performed in 2008 and 2009, respectively.

The GPS receivers on the first flight in 2008 were a navigation-type receiver, not especially adapted for such an experiment. The data was collected on a single baseline with two dual-frequency receivers. The receivers were controlled by, and the data stored on, an ARM Linux board using an RS-232 serial connection.

For the second flight in 2009, we used a multi-antenna receiver, for which the Coordinating Committee for Multilateral Export Controls altitude restriction was explicitly removed. This receiver has three RF inputs that can be connected to three antennas, so the observations from the three antennas are time-synchronized by a common clock. The receiver has the option to store observations internally, which simplified the control of the GPS experiment. We used three antennas: one L1/L2 antenna as the main antenna and two L1 antennas as auxiliary antennas.

Theory of Attitude Determination

In this section, we will provide background information on the models applied in our GPS experiment. More details can be found in the publications listed in Further Reading.

Standard LAMBDA. Most GNSS receivers make use of two types of observations: pseudorange (code) and carrier phase. The pseudorange observations typically have a precision of decimeters, whereas carrier-phase observations have precisions up to the millimeter level.

Carrier-phase observations are affected, however, by an unknown number of integer-cycle ambiguities, which have to be resolved before we can exploit the higher precision of these observations. The observation equations for the double-difference (between satellites and between antennas/receivers) can be written for a single baseline as a system of linearized observation equations:

(1)

where E(y) is the expected value and D(y) is the dispersion of y. The vector of observed-minus-computed double-difference carrier-phase and code observations is given by y; z is the vector of unknown ambiguities expressed in cycles rather than distance units to maintain their integer character; b is the baseline vector, which is unknown for relative navigation applications but for which the length in attitude determination is generally known; A is a design matrix that links the data vector to the vector z; and B is the geometry matrix containing normalized line-of-sight vectors. The variance-covariance matrix of y is represented by the positive definite matrix Qyy, which is assumed to be known.

The least-squares solution of the linear system of observation equations as introduced in Equation (1) is obtained using from:

. (2)

The integer solution of this system can be obtained by applying the standard Least-Squares Ambiguity Decorrelation Adjustment (LAMBDA) method.

Constrained LAMBDA. In applications for which some of the baseline lengths are known and constant, for example GNSS-based attitude determination, we can exploit the so-called baseline-constrained model. Then, the baseline-constrained integer ambiguity resolution can make use of the standard GNSS model by adding the length constraint of the baseline, ||b|| = , where is known. The least-squares criterion for this problem reads:

.(3)

The solution can be obtained with the baseline-constrained (or C-)LAMBDA method, which is described in referred literature listed in Further Reading. Later on, we will refer to the attitude calculated by this approach simply as C-LAMBDA.

For platforms with more than one baseline, the C-LAMBDA method can be applied to each baseline individually, and the full attitude can be determined using those individual baseline solutions. For completeness, we also mention a recently developed solution of this problem, called the multivariate-constrained (MC-) LAMBDA, which integrally accounts for both the integer and attitude matrix. Both approaches are applied in the analyses of the BOV data.

Onboard Attitude Determination. In this article, we also use the onboard estimate of the attitude as provided by the multi-antenna receiver. The method applied in the receiver is based on a Kalman filter and the ambiguities are resolved by the standard LAMBDA method. The baseline length, if the information is provided to the receiver a priori, is used to validate the results. For baseline lengths of about 1 meter, the receiver’s pitch and roll accuracy is about 0.60 degrees, and heading about 0.30 degrees according to the receiver manual. We will refer to the attitude as provided by the receiver as KF.

Flight Experiments

In this section, we will discuss our analyses of the GPS data from two of the BOV experiments.

Gondola Experimental Flight 2008. In September 2008, we performed a test of the ADP for a future version of the BOV and a GPS system containing two navigation-grade GPS receivers. The goal of the experiment was to confirm nominal performance in the real environment of the ADP sensors and GPS receivers on the gondola; therefore, the BOV was not launched. The data from the single baseline was used to determine the pointing direction of the gondola, an application referred to as the GNSS compass. The receivers and the controller were stored in an airtight container (see FIGURE 4) and the antennas were sealed in waterproof bags. The location of the two GPS antennas on the gondola is indicated in Figure 4. The baseline length was 1.95 meters. Both receivers used their own individual clocks, so observations were not synchronized. The trajectory (altitude) of this flight is shown in the right-hand side of Figure 4, with the longitude and latitude shown in FIGURE 5. This is a typical flight profile for our application. The flight takes about three hours and reaches an altitude of more than 40 kilometers.

Figure 4A. Single baseline experiment performed in September 2008, the sensor configuration.

Figure 4B. Single baseline experiment performed in September 2008, the flight trajectory (altitude).

Figure 5. Flight trajectory of the 2008 experiment.

First, the balloon makes use of the wind direction in the lower layers of the atmosphere, which brings it eastwards. During this part of the flight, the balloon is kept at a maximum altitude of about 12 kilometers. After about 30 minutes, the altitude is increased to make use of a different wind direction that carries the balloon back in the westerly direction toward the launch base in order to ease the recovery of the capsule and/or the gondola.

At the end of the flight, there is a parachute-guided fall over 40 kilometers to sea level, for both the gondola and the BOV (if it is launched), which takes about 30 minutes. In this experiment, we could confirm the nominal operation of some of the sensors and reception of the GPS signals on the gondola under the large balloon.

Gondola Experimental Flight 2009. In May 2009, the third flight of the BOV was performed. The three GPS receiver antennas and the other attitude sensors were placed on an alignment frame for stiffness, which was then attached to the gondola. Furthermore, we used a ground station to demonstrate the combination of GPS-based attitude determination and relative positioning between the platform and the ground station. As the motion of the system is rather unpredictable, we used a kinematic approach for both attitude determination and relative positioning.

Preflight static test: Before the flight, we did a ground test using the actual antenna frame of the gondola (see FIGURE 6). The roll, pitch, and heading angles for this static test are shown on the right-hand side of this figure. Due to the geometry of the baselines, the heading angle is more accurate. For this static test, we can calculate the standard deviation of the three angles to confirm the accuracy achievable for the flight test. These results are summarized in TABLE 2. For the baselines with a length of about 1.4 meters, we achieved an accuracy of about 0.25 degrees for the roll and pitch angles and 0.1 degrees for heading, which is as expected from the lengths and geometry of the baselines. Using single-epoch data, we could resolve the ambiguities correctly for more than 99 percent of the epochs (see TABLE 3). Also, the standard deviation of the receiver’s Kalman-filter-based attitude estimate (KF) is included in the table. The accuracy is, after convergence of the filter, similar to our C-LAMBDA result, although the applied method is very different. The Kalman filter takes about 10 seconds to converge for this static experiment, whereas the C-LAMBDA method provides this accuracy from the very first epoch. For completeness, the instantaneous success rate of the standard LAMBDA and MC-LAMBDA methods are also included in Table 3.

Figure 6. Static experiment: Setup of the GPS antennas.

Figure 6. Static experiment: C-LAMBDA-based attitude estimates.

Table 2. Standard deviation of attitude angles for static test.

Table 3. Single-epoch, overall success rate for baseline 1-2 (static experiment).

Gondola nominal flight: Next, we applied the same GPS configuration on the gondola. An important difference with respect to the static field experiment is that the antennas were now placed under the balloon and inside waterproof bags (see the picture on the left-hand side of FIGURE 7). The right-hand side of Figure 7 shows the flight trajectory (altitude) of the experiment. At 21:05 UTC (07:05 Japan Standard Time), the balloon was released from the sliding launcher (Figure 1). In 2.5 hours, the balloon reached an altitude of more than 41 kilometers from which the BOV was dropped. At 23:55, the BOV was released from the Gondola, and at 23:59 the gondola was separated from the balloon. After the release of the BOV, the balloon and gondola ascended more than 2 kilometers because of the reduced mass of the system. For this flight, the attitude determination package and the GPS system were installed on the gondola to confirm the nominal performance of all the sensors.

Figure 7A. Full attitude experiment performed in May 2009, sensor configuration.

Figure 7B. Full attitude experiment performed in May 2009, flight trajectory (altitude).

Using the new GPS receiver with three antennas, we are able to calculate the full attitude of the gondola. The roll and pitch estimates, from both C-LAMBDA and KF, are shown in FIGURE 8. The heading angle from the GPS-based C-LAMBDA and KF, and that from the GAS and SAS sensors are shown in FIGURE 9. As explained in a previous section, the four SAS sensors will only output an attitude estimate if the sun is in the field of view of a sensor. Therefore we can distinguish four bands in the heading estimate of the SAS, corresponding to the individual sensors (indicated in Figure 7 as SAS1 to SAS4).

Figure 8A. GPS results for roll angles during nominal flight.

Figure 8B. GPS results for pitch angles during nominal flight.

Figure 9A. GPS results for heading angle during nominal flight.

Figure 9B. GAS and SAS results for heading angle during nominal flight.

The number of locked GPS satellites at the main antenna is shown on the right-hand side of Figure 7. Before takeoff, we saw that the number of locked channels varies rapidly due to obstructions, but after takeoff the number is rather constant until the BOV is separated from the gondola. Before takeoff, the GPS observations are affected by the obstruction of the sliding launcher and therefore ambiguity resolution is only possible on the second baseline (see Figure 8). Also, the GPS receiver itself does not provide an attitude estimation during this phase of the experiment. During takeoff, we see large variations in orientation of the gondola (up to 20 degrees (±10 degrees) for both roll and pitch), which can be estimated well by both C-LAMBDA and KF. Again, the Kalman filter takes a few epochs to converge (in this case, 15 seconds from takeoff), whereas the C-LAMBDA method provides an accurate solution from the very first epoch. After takeoff, the attitude of the gondola stabilizes and the C-LAMBDA and KF attitude estimates are very similar.

We investigated the difference between the attitude estimation from the different sensors during nominal flight. The mean and standard deviations of the differences are shown in TABLE 4. If we compare the C-LAMBDA and KF attitudes, we observe biases for all angles. This is something we have to investigate further, but the most likely cause for this bias is the time delay of the Kalman filter in response to changes in attitude, as we observed in the static experiment in the form of convergence time.

Table 4. Attitude differences (offset/standard deviation) for flight test of 2009.

The standard deviation for the difference in the estimates of roll, pitch, and heading is as expected. For the comparison with the other sensors, we use the C-LAMBDA attitude as the reference. Between C-LAMBDA and GAS/SAS, we observe a bias, most likely due to minor misalignment issues between the sensors. The standard deviations in Table 4 are in line with expectation based on the sensor specifications. During this part of the flight, we achieved a single-epoch, single-frequency empirical overall success rate for ambiguity resolution on the two baselines of 95.09 percent. As a reference, we also include in TABLE 5 the success rate for standard LAMBDA using observations from a single epoch. If we make use of the MC-LAMBDA method, the success rate is increased to 99.88 percent as shown in the table. The success rate is higher as the integrated model for all the baselines is stronger.

Table 5. Single-epoch, overall success rate for baseline 1-2 (flight experiment).

Gondola flight after BOV separation: After the separation of the BOV from the gondola, the gondola starts to ascend and sway. FIGURE 10 contains roll and pitch estimates for this part of the flight until the gondola separation. In the figure, we see large variations in the orientation of the gondola (up to 40 (±20) degrees for roll and 20 (± 10) degrees for pitch). It is interesting that after BOV separation, during the large maneuvers of the gondola caused by the separation, both KF and C-LAMBDA estimates are available but to a certain extent are different. Table 4 also contains standard deviations and biases between C-LAMBDA and KF for this part of the flight.

Figure 10A. GPS results for roll angles during nominal flight.

Figure 10B. GPS results for pitch angles during nominal flight.

We conclude that the differences (standard deviation but also bias) between C-LAMBDA and KF — both for roll and pitch — are increased compared to the nominal part of the flight. This confirms our expectation that the Kalman-filter-based result lags behind the true attitude in dynamic situations, whereas the C-LAMBDA result based on single-epoch data should be able to provide the same accurate estimate as during the other phases of the flight.

Future Work

For the final phase of the experiment program, we would like to collect multi-baseline data from a number of vehicles. The preferred option for the experiment is three antennas (two independent baselines) on the BOV, and two antennas (one baseline) on the gondola. Furthermore, similar to our 2009 experiment, a number of antennas at a reference station could be used. The goal of the final phase of the program is to collect data for offline relative positioning and attitude determination, though real-time emulation, between a number of vehicles that form a network.

Acknowledgments

Peter Buist thanks Professor Peter Teunissen for support with the theory behind ambiguity resolution and, including Gabriele Giorgi, for the pleasant cooperation during our research. The MicroNed-MISAT framework is kindly thanked for their support. The research of Sandra Verhagen is supported by the Dutch Technology Foundation STW, the Applied Science Division of The Netherlands Organisation for Scientific Research (NWO), and the Technology Program of the Ministry of Economic Affairs. This article is based on the paper “GPS Experiment on the Balloon-based Operation Vehicle” presented at the Institute of Electrical and Electronics Engineers / Institute of Navigation Position Location and Navigation Symposium 2010, held in Indians Wells, California, May 6–8, 2010, where it received a best-paper-in-track award.

Manufacturers

The Attitude Determination Package’s Sun Aspect Sensor is based on photodiodes manufactured by Hamamatsu Photonics K.K.; the Geomagnetic Aspect Sensor is based on magnetometers manufactured by Bartington Intruments Ltd.; the inclinometer is based on a module manufactured by Measurement Specialties; and the gyro is manufactured by Silicon Sensing Systems Japan, Ltd. For the 2009 experiment, we used a Septentrio N.V. PolaRx2@ multi-antenna receiver with S67-1575-96 and S67-1575-46 antennas from Sensor Systems Inc. Details on the receivers and antennas used for the 2008 experiment are not publicly available. A Trimble Navigation Ltd. R7 receiver and two NovAtel Inc. OEMV receivers were used at the reference ground station. The ARM-Linux logging computer is an Armadillo PC/104 manufactured by Atmark Techno, Inc.

Peter J. Buist is a researcher at Delft University of Technology in Delft, The Netherlands. Before rejoining DUT in 2006, he developed GPS receivers for the SERVIS-1, USERS, ALOS, and other satellites and the H2A rocket, and subsystems for QZSS in the Japanese space industry.

Sandra Verhagen is an assistant professor at Delft University of Technology in Delft, The Netherlands. Together with Peter Buist, she is working on the Australian Space Research Program GARADA project on synthetic aperture radar formation flying.

Tatsuaki Hashimoto received his Ph.D. in electrical engineering from the University of Tokyo in 1990. He is a professor of the Institute of Space and Astronautical Science (ISAS), Japan Aerospace Exploration Agency (JAXA).

Shujiro Sawai received his Ph.D. in engineering from the University
of Tokyo in 1994. He is an associate professor at ISAS/JAXA.

Shin-Ichiro Sakai received his Ph.D. degree from the University of Tokyo in 2000. He joined ISAS/JAXA in 2001 and became associate professor in 2005.

Nobutaka Bando received a Ph.D. in electrical engineering from the University of Tokyo in 2005. He is an assistant professor at ISAS/JAXA.

Shigehito Shimizu received a master’s degree in engineering from Tohoku University in Sendai, Japan, in 2007. He is an engineer in the Navigation, Guidance and Control Group at JAXA.

Innovation: Multipath Minimization Method

GPS World staff — Fri, 01 Jul 2011 21:36:35 +0000

INNOVATION INSIGHTS with Richard Langley

By Luis Serrano, Don Kim, and Richard B. Langley

Multipath is real and omnipresent, a detriment when GPS is used for positioning, navigation, and timing. The authors look at a technique to reduce multipath by using a pair of antennas on a moving vehicle together with a sophisticated mathematical model. This reduces the level of multipath on carrier-phase observations and thereby improves the accuracy of the vehicle’s position.

“OUT, DAMNED MULTIPATH! OUT, I SAY!” Many a GPS user has wished for their positioning results to be free of the effect of multipath. And unlike Lady Macbeth’s imaginary blood spot, multipath is real and omnipresent. Although it may be considered beneficial when GPS is used as a remote sensing tool, it is a detriment when GPS is used for positioning, navigation, and timing — reducing the achievable accuracy of results.

Clearly, the best way to reduce the effects of multipath is to try avoiding it in the first place by siting the receiver’s antenna as low as possible and far away from potential reflectors. But that’s not always feasible. The next best approach is to reduce the level of the multipath signal entering the receiver by attenuating it with a suitably designed antenna. A large metallic ground plane placed beneath an antenna will modify the shape of the antenna’s reception pattern giving it reduced sensitivity to signals arriving at low elevation angles and from below the antenna’s horizon. So-called choke-ring antennas also significantly attenuate multipath signals. And microwave-absorbing materials appropriately placed in an antenna’s vicinity can also be beneficial.

Multipath can also be mitigated by special receiver correlator designs. These designs target the effect of multipath on code-phase measurements and the resulting pseudorange observations. Several different proprietary implementations in commercial receivers significantly reduce the level of multipath in the pseudoranges and hence in pseudorange-based position and time estimates. Some degree of multipath attenuation can be had by using the low-noise carrier-phase measurements to smooth the pseudoranges before they are processed. The effect of multipath on carrier phases is much smaller than that on pseudoranges. In fact, it is limited to only one-quarter of the carrier wavelength when the reflected signal’s amplitude is less than that of the direct signal. This means that at the GPS L1 frequency, the multipath contamination in a carrier-phase measurement is at most about 5 centimeters. Nevertheless, this is still unacceptably large for some high-accuracy applications.

At a static site, with an unchanging multipath environment, the signal reflection geometry repeats day to day and the effect of multipath can be reduced by sidereal filtering or “stacking” of coordinate or carrier-phase-residual time series. However, this approach is not viable for scenarios where the receiver and antenna are moving such as in machine control applications. Here an alternative approach is needed.

In this month’s column, I am joined by two of my UNB colleagues as we look at a technique that uses a pair of antennas on a moving vehicle together with a sophisticated mathematical model, to reduce the level of multipath on carrier-phase observations and thereby improve the accuracy of the vehicle’s position.

Real-time-kinematic (RTK) GNSS-based machine automation systems are starting to appear in the construction and mining industries for the guidance of dozers, motor graders, excavators, and scrapers and in precision agriculture for the guidance of tractors and harvesters. Not only is the precise and accurate position of the vehicle needed but its attitude is frequently required as well.

Previous work in GNSS-based attitude systems, using short baselines (less than a couple of meters) between three or four antennas, has provided results with high accuracies, most of the time to the sub-degree level in the attitude angles. If the relative position of these multiple antennas can be determined with real-time centimeter-level accuracy using the carrier-phase observables (thus in RTK-mode), the three attitude parameters (the heading, pitch, and roll angles) of the platform can be estimated.

However, with only two GNSS antennas it is still possible to determine yaw and pitch angles, which is sufficient for some applications in precision agriculture and construction. Depending on the placement of the antennas on the platform body, the determination of these two angles can be quite robust and efficient.

Nevertheless, even a small separation between the antennas results in different and decorrelated phase-multipath errors, which are not removed by simply differencing measurements between the antennas.

The mitigation of carrier-phase multipath in real time remains, to a large extent, very limited (unlike the mitigation of code multipath through receiver improvements) and it is commonly considered the major source of error in GNSS-RTK applications. This is due to the very nature of multipath spectra, which depends mainly on the location of the antenna and the characteristics of the reflector(s) in its vicinity. Any change in this binomial (antenna/reflectors), regardless of how small it is, will cause an unknown multipath effect.

Using typical choke-ring antennas to reduce multipath is typically not practical (not to mention cost prohibitive) when employing multiple antennas on dynamic platforms. Extended flat ground planes are also impractical. Furthermore, such antenna configurations typically only reduce the effects of low angle reflections and those coming from below the antenna horizon.

One promising approach to mitigating the effects of carrier-phase multipath is to filter the raw measurements provided by the receiver. But, unlike the scenario at a fixed site, the multipath and its effects are not repeatable. In machine automation applications, the machinery is expected to perform complex and unpredictable maneuvers; therefore the removal of carrier-phase multipath should rely on smart digital filtering techniques that adapt not only to the background multipath (coming mostly from the machine’s reflecting surfaces), but also to the changing multipath environment along the machine’s path.

In this article, we describe how a typical GPS-based machine automation application using a dual-antenna system is used to calibrate, in a first step, and then remove carrier-phase multipath afterwards. The intricate dynamical relationship between the platform’s two “rover” antennas and the changing multipath from nearby reflectors is explored and modeled through several stochastic and dynamical models. These models have been implemented in an extended Kalman filter (EKF).

MIMICS Strategy

Any change in the relative position between a pair of GNSS antennas most likely will affect, at a small scale, the amplitude and polarization of the reflected signals sensed by the antennas (depending on their spacing). However, the phase will definitely change significantly along the ray trajectories of the plane waves passing through each of the antennas.

This can be seen in the equation that describes the single-difference multipath between two close-by antennas (one called the “master” and the other the “slave”):

(1)

where the angle is the relative multipath phase delay between the antennas and a nearby effective reflector (α₀ is the multipath signal amplitude in the master and slave antennas, and is dependent on the reflector characteristics, reflection coefficient, and receiver tracking loop).

As our study has the objective to mimic as much as possible the multipath effect from effective reflectors in kinematic scenarios with variable dynamics, we decided to name the strategy MIMICS, a slightly contrived abbreviation for “Multipath profile from between receIvers dynaMICS.”

The MIMICS algorithm for a dual-antenna system is based on a specular reflector ray-tracing multipath model (see Figure 1).

Figure 1. 3D ray-tracing modeling of phase multipath for a GNSS dual-antenna system. 0 designates the “master” antenna; 1, the “slave” antenna; Elev and Az, the elevation angle and the azimuth of the satellite, respectively. The other symbols are explained in the text.

After a first step of data synchronization and data-snooping on the data provided by the two receiver antennas, the second step requires the calculation of an approximate position for both antennas, relaxed to a few meters using a standard code solution.

A precise estimation of both antennas’ velocity and acceleration (in real time) is carried out using the carrier-phase observable. Not only should the antenna velocity and acceleration estimates be precisely determined (on the order of a few millimeters per second and a few millimeters per second squared, respectively) but they should also be immune to low-frequency multipath signatures. This is important in our approach, as we use the antennas’ multipath-free dynamic information to separate the multipath in the raw data.

We will start from the basic equations used to derive the single-difference multipath observables.

The observation equation for a single-difference between receivers, using a common external clock (oscillator), is given (in distance units) by:

(2)

where m indicates the master antenna; s, the slave antenna; prn, the satellite number; Δ, the operator for single differencing between receivers; Φ, the carrier-phase observation; ρ, the slant range between the satellite and receiver antennas; N, the carrier-phase ambiguity; M, the multipath; and ε, the system noise.

By sequentially differencing Equation (2) in time to remove the single-difference ambiguity from the observation equation, we obtain (as long as there is no loss of lock or cycle slips):

(3)

where

(4)

One of the key ideas in deriving the multipath observable from Equation (3) is to estimate given by Equation (4). We will outline our approach in a later section.

From Equation (3), at the second epoch, for example, we will have:

(5)

If we continue this process up to epoch n, we will obtain an ensemble of differential multipath observations.

If we then take the numerical summation of these, we will have

(6)

Note that n samples of differential multipath observations are used in Equation (6). Therefore, we need n + 1 observations.

Assume that we perform this process taking n = 1, then n = 2, and so on until we obtain r numerical summations of Equation (6) and then take a second numerical summation of them, we will end up with the following equation:

(7)

where E is the expectation operator.

Another key idea in our approach is associated with Equation (7). To isolate the initial epoch multipath, , from the differential multipath observations, the first term on the right-hand side of Equation (7), , should be removed.

This can be accomplished by mechanical calibration and/or numerical randomization. To summarize the idea, we have to create random multipath physically (or numerically) at the initialization step. When the isolation of the initial multipath epoch is completed, we can recover multipath at every epoch using Equation (5).

Digital Differentiators. We introduce digital differentiators in our approach to derive higher order range dynamics (that is, range rate, range-rate change, and so on) using the single-difference (between receivers connected to a common external oscillator) carrier-phase observations. These higher order range dynamics are used in Equation (4).

There are important classes of finite-impulse-response differentiators, which are highly accurate at low to medium frequencies. In central-difference approximations, both the backward and the forward values of the function are used to approximate the current value of the derivative.

Researchers have demonstrated that the coefficients of the maximally linear digital differentiator of order 2N + 1 are the same as the coefficients of the easily computed central-difference approximation of order N.

Another advantage of this class is that within a certain maximum allowable ripple on the amplitude response of the resultant differentiator, its pass band can be dramatically increased. In our approach, this is something fundamental as the multipath in kinematic scenarios is conceptually treated as high-frequency correlated multipath, depending on the platform dynamics and the distance to the reflector(s).

Adaptive Estimation. To derive single-difference multipath at the initial epoch, , a numerical randomization (or mechanical calibration) of the single-difference multipath observations is performed in our approach. A time series of the single-difference multipath observations to be randomized is given as

(8)

Then our goal is to achieve the following condition:

(9)

It is obvious that the condition will only hold if multipath truly behaves as a stochastic or random process. The estimation of multipath in a kinematic scenario has to be understood as the estimation of time-correlated random errors. However, there is no straightforward way to find the correlation periods and model the errors.

Our idea is to decorrelate the between-antenna relative multipath through the introduction of a pseudorandom motion. As one cannot completely rely only on a decorrelation through the platform calibration motion, one also has to do it through the mathematical “whitening” of the time series.

Nevertheless, the ensemble of data depicted in the above formulation can be modeled as an oscillatory random process, for which second or higher order autoregressive (AR) models can provide more realistic modeling in kinematic scenarios. (An autoregressive process is simply another name for a linear difference equation model where the input or forcing function is white Gaussian noise.) We can estimate the parameters of this model in real time, in a block-by-block analysis using the familiar Yule-Walker equations. A whitening filter can then be formed from the estimation parameters.

We obtain the AR coefficients using the autocorrelation coefficient vector of the random sequences. Since the order of the coefficient estimation depends on the multipath spectra (in turn dependent on the platform dynamics and reflector distance), MIMICS uses a cost function to estimate adaptively, in real time, the appropriate order. An order too low results in a poor whitener of the background colored noise, while an order too large might affect the embedded original signal that we are interested in detecting.

The cost function uses the residual sum of squared error. The order estimate that gives the lowest error is the one chosen, and this task is done iteratively until it reaches a minimum threshold value. Once this stage is fulfilled, the multipath observable can be easily obtained.

Testing

The main test that we have performed so far (using a pair of high performance dual-frequency receivers fed by compact antennas and a rubidium frequency standard, all installed in a vehicle) was designed to evaluate the amount of data necessary to perform the decorrelation, and to determine if the system was observable (in terms of estimating, at every epoch, several multipath parameters from just two-antenna observations). Receiver data was collected and post-processed (so-called RTK-style processing) although, with sufficient computing power, data processing could take place in real, or near real, time.

In a real-life scenario, the platform pseudorandom motions have the advantage that carrier-phase embedded dynamics are typically changing faster and in a three-dimensional manner (antennas sense different pitch and yaw angles). Thus a faster and more robust decorrelation is possible.

One can see from the bottom picture in Figure 2 the façade of the building behaving as the effective reflector. The vehicle performed several motions, depicted in the bottom panel of Figure 3, always in the visible parking lot, hence the building constantly blocked the view to some satellites. We used only the L1 data from the receivers recorded at a rate of 10 Hz.

Figure 2A. Kinematic test setup. The offset third antenna was not used for the test discussed in this article.

Figure 2B. Kinematic test setup. The offset third antenna was not used for the test discussed in this article.

In the bottom panel of Figure 3, one can also see the kind of motion performed by the platform. Accelerations, jerk, idling, and several stops were performed on purpose to see the resultant multipath spectra differences between the antennas. The reference station (using a receiver with capabilities similar to those in the vehicle) was located on a roof-top no more than 110 meters away from the vehicle antennas during the test. As such, most of the usual biases where removed from the solution in the differencing process and the only remaining bias can be attributed to multipath. The data from the reference receiver was only used to obtain the varying baseline with respect to the vehicle master antenna.

In the top panel of Figure 3, one can see the geometric distance calculated from the integer-ambiguity-fixed solutions of both antenna/receiver combinations. Since the distance between the mounting points on the antenna-support bar was accurately measured before the test (84 centimeters), we had an easy way to evaluate the solution quality. The “outliers” seen in the figure come from code solutions because the building mentioned before blocked most of the satellites towards the southeast. As a result, many times fewer than five satellites were available.

Figure 3. Correlation between vehicle dynamics (heading angle) and the multipath spectra.

Looking at the first nine minutes of results in Figure 4, one can see that when the vehicle is still stationary, the multipath has a very clear quasi-sinusoidal behavior with a period of a few minutes. Also, one can see that it is zero-mean as expected (unlike code multipath). When the vehicle starts moving (at about the four-minute mark), the noise figure is amplified (depending on the platform velocity), but one can still see a mixture of low-frequency components coming from multipath (although with shorter periods).

These results indicate, firstly, that regardless of the distance between two antennas, multipath will not be eliminated after differencing, unlike some other biases. Secondly, when the platform has multiple dynamics, multipath spectra will change accordingly starting from the low-frequency components (due to nearby reflectors) towards the high-frequency ones (including diffraction coming from the building edges and corners). As such, our approach to adaptively model multipath in real time as a quasi-random process makes sense.

Figure 4. Position results from the kinematic test, showing the estimated distance between the two vehicle antennas (upper plot) and the distance between the master antenna and the reference antenna.

Multipath Observables. The multipath observables are obtained through the MIMICS algorithm. It is quite flexible in terms of latency and filter order when it comes to deriving the observables. Basically, it is dependent on the platform dynamics and the amplitude of the residuals of the whitened time series (meaning that if they exceed a certain threshold, then the filtering order doesn’t fit the data).

When comparing the observations delivered every half second for PRN 5 with the ones delivered every second, it is clear that the larger the interval between observations, the better we are able to recover the true biased sinusoidal behavior of multipath. However, in machine control, some applications require a very low latency. Therefore, there must be a compromise between the multipath observable accuracy and the rate at which it is generated.

Multipath Parameter Estimation. Once the multipath observables are derived, on a satellite-by-satellite basis, it is possible to estimate the parameters (a₀, the reflection coefficient; γ₀, the phase delay; φ₀, the azimuth of reflected signal; and θ₀, the elevation angle of reflected signal) of the multipath observable described in Equation (1) for each satellite. As mentioned earlier, an EKF is used for the estimation procedure.

When the platform experiences higher dynamics, such as rapid rotations, acceleration is no longer constant and jerk is present. Therefore, a Gauss-Markov model may be more suitable than other stochastic models, such as random walk, and can be implemented through a position-velocity-acceleration dynamic model.

As an example, the results from the multipath parameter estimation are given for satellite PRN 5 in Figure 5. One can see that it takes roughly 40 seconds for the filter to converge. This is especially seen in the phase delay.

Converted to meters, the multipath phase delay gives an approximate value of 10 meters, which is consistent with the distance from the moving platform to the dominant specular reflector (the building’s façade).

Figure 5. PRN 5 multipath parameter estimation.

Multipath Mitigation. After going through all the MIMICS steps,
from the initial data tracking and synchronization between the dual-antenna system up to the multipath parameter estimation for each continuously observed satellite, we can now generate the multipath corrections and thus correct each raw carrier-phase observation.

One can see in Figure 6 three different plots from the solution domain depicting the original raw (multipath-contaminated) GPS-RTK baseline up-component (top), the estimated carrier-phase multipath signal (middle), and the difference between the two above time series; that is, the GPS-RTK multipath-ameliorated solution (bottom). A clear improvement is visible. In terms of numbers, and only considering the results “cleaned” from outliers and differential-code solutions (provided by the RTK post-processing software, when carrier-phase ambiguities cannot be fixed), the up-component root-mean-square value before was 2.5 centimeters, and after applying MIMICS it stood at 1.8 centimeters.

Figure 6. MIMICS algorithm results for the vehicle baseline from the first 9 minutes of the test.

Concluding Remarks

Our novel strategy seems to work well in adaptively detecting and estimating multipath profiles in simulated real time (or near real time as there is a small latency to obtain multipath corrections from the MIMICS algorithm). The approach is designed to be applied in specular-rich and varying multipath environments, quite common at construction sites, harbors, airports, and other environments where GNSS-based heading systems are becoming standard. The equipment setup can be simplified, compared to that used in our test, if a single receiver with dual-antenna inputs is employed.

Despite its success, there are some limitations to our approach. From the plots, it’s clear that not all multipath patterns were removed, even though the improvements are notable. Moreover, estimating multipath adaptively in real time can be a problem from a computational point of view when using high update rates. And when the platform is static and no previous calibration exists, the estimation of multipath parameters is impossible as the system is not observable. Nevertheless, the approach shows promise and real-world tests are in the planning stages.

Acknowledgments

The work described in this article was supported by the Natural Sciences and Engineering Research Council of Canada. The article is based on a paper given at the Institute of Electrical and Electronics Engineers / Institute of Navigation Position Location and Navigation Symposium 2010, held in Indian Wells, California, May 6–8, 2010.

Manufacturers

The test of the MIMICS approach used two NovAtel OEM4 receivers in the vehicle each fed by a separate NovAtel GPS-600 “pinweel” antenna on the roof. A Temex Time (now Spectratime) LPFRS-01/5M rubidium frequency standard supplied a common oscillator frequency to both receivers. The reference receiver was a Trimble 5700, fed by a Trimble Zephyr geodetic antenna.

Luis Serrano is a senior navigation engineer at EADS Astrium U.K., in the Ground Segment Group, based in Portsmouth, where he leads studies and research in GNSS high precision applications and GNSS anti-jamming/spoofing software and patents. He is also a completing his Ph.D. degree at the University of New Brunwick (UNB), Fredericton, Canada.

Don Kim is an adjunct professor and a senior research associate in the Department of Geodesy and Geomatics Engineering at UNB where he has been doing research and teaching since 1998. He has a bachelor’s degree in urban engineering and an M.Sc.E. and Ph.D. in geomatics from Seoul National University. Dr. Kim has been involved in GNSS research since 1991 and his research centers on high-precision positioning and navigation sensor technologies for practical solutions in scientific and industrial applications that require real-time processing, high data rates, and high accuracy over long ranges with possible high platform dynamics.

Innovation: Doppler-Aided Positioning: Improving Single-Frequency RTK in the Urban Enviornment

Richard Langley — Sun, 01 May 2011 21:09:47 +0000

INNOVATION INSIGHTS with Richard Langley

By Mojtaba Bahrami and Marek Ziebart

A look at how Doppler measurements can be used to smooth noisy code-based pseudoranges to improve the precision of autonomous positioning as well as to improve the availability of single-frequency real-time kinematic positioning, especially in urban environments.

WHAT DO A GPS RECEIVER, a policeman’s speed gun, a weather radar, and some medical diagnostic equipment have in common? Give up? They all make use of the Doppler effect. First proposed in 1842 by the Austrian mathematician and physicist, Christian Doppler, it is the change in the perceived frequency of a wave when the transmitter and the receiver are in relative motion.

Doppler introduced the concept in an attempt to explain the shift in the color of light from certain binary stars. Three years later, the effect was tested for sound waves by the Dutch scientist Christophorus Buys Ballot. We have all heard the Doppler shift of a train whistle or a siren with their descending tones as the train or emergency vehicle passes us. Doctors use Doppler sonography — also known as Doppler ultrasound — to provide information about the flow of blood and the movement of inner areas of the body with the moving reflectors changing the received ultrasound frequencies. Similarly, some speed guns use the Doppler effect to measure the speed of vehicles or baseballs and Doppler weather radar measures the relative velocity of particles in the air.

The beginning of the space age heralded a new application of the Doppler effect. By measuring the shift in the received frequency of the radio beacon signals transmitted by Sputnik I from a known location, scientists were able to determine the orbit of the satellite. And shortly thereafter, they determined that if the orbit of a satellite was known, then the position of a receiver could be determined from the shift. That realization led to the development of the United States Navy Navigation Satellite System, commonly known as Transit, with the first satellite being launched in 1961. Initially classified, the system was made available to civilians in 1967 and was widely used for navigation and precise positioning until it was shut down in 1996. The Soviet Union developed a similar system called Tsikada and a special military version called Parus. These systems are also assumed to be no longer in use — at least for navigation.

GPS and other global navigation satellite systems use the Doppler shift of the received carrier frequencies to determine the velocity of a moving receiver. Doppler-derived velocity is far more accurate than that obtained by simply differencing two position estimates. But GPS Doppler measurements can be used in other ways, too. In this month’s column, we look at how Doppler measurements can be used to smooth noisy code-based pseudoranges to improve the precision of autonomous positioning as well as to improve the availability of single-frequency real-time kinematic positioning, especially in urban environments.

Correction and Further Details

The first experimental Transit satellite was launched in 1959. A brief summary of subsequent launches follows:

Transit 1A launched 17 September 1959 failed to reach orbit
Transit 1B launched 13 April 1960 successfully
Transit 2A launched 22 June 1960 successfully
Transit 3A launched 30 November 1960 failed to reach orbit
Transit 3B launched 22 February 1961 failed to deploy in correct orbit
Transit 4A launched 29 June 1961 successfully
Transit 4B launched 15 November 1961 successfully
Transits 4A and 4B used the 150/400 MHz pair of frequencies and provided geodetically useful results.
A series of Transit prototype and research satellites was launched between 1962 and 1964 with the first fully operational satellite, Transit 5-BN-2, launched on 5 December 1963.
The first operational or Oscar-class Transit satellite, NNS O-1, was launched on 6 October 1964.
The last pair of Transit satellites, NNS O-25 and O-31, was launched on 25 August 1988.

“Innovation” is a regular column that features discussions about recent advances in GPS technology and its applications as well as the fundamentals of GPS positioning. The column is coordinated by Richard Langley of the Department of Geodesy and Geomatics Engineering at the University of New Brunswick, who welcomes your comments and topic ideas. To contact him, email lang @ unb.ca.

Real-time kinematic (RTK) techniques enable centimeter-level, relative positioning. The technology requires expensive, dedicated, dual-frequency, geodetic-quality receivers. However, myriad industrial and engineering applications would benefit from small-size, cost-effective, single-frequency, low-power, and high-accuracy RTK satellite positioning. Can such a sensor be developed and will it deliver? If feasible, such an instrument would find many applications within urban environments — but here the barriers to success are higher. In this article, we show how some of the problems can be overcome.

Single-Frequency RTK

Low-cost single-frequency (L1) GPS receivers have attained mass-market status in the consumer industry. Notwithstanding current levels of maturity in GPS hardware and algorithms, these receivers still suffer from large positioning errors. Any positioning accuracy improvement for mass-market receivers is of great practical importance, especially for many applications demanding small size, cost-effectiveness, low power consumption, and highly accurate GPS positioning and navigation. Examples include mobile mapping technology; machine control; agriculture fertilization and yield monitoring; forestry; utility services; intelligent transportation systems; civil engineering projects; unmanned aerial vehicles; automated continuous monitoring of landslides, avalanches, ground subsidence, and river level; and monitoring deformation of built structures. Moreover, today an ever-increasing number of smartphones and handsets come equipped with a GPS receiver. In those devices, the increasing sophistication of end-user applications and refinement of map databases are continually tightening the accuracy requirements for GPS positioning.

For single-frequency users, the RTK method does appear to offer the promise of highly precise position estimates for stationary and moving receivers and can even be considered a candidate for integration within mobile handhelds. Moreover, the RTK approach is attractive because the potential of the existing national infrastructures such as Great Britain’s Ordnance Survey National GNSS Network-RTK (OSNet) infrastructure, as well as enabling technologies such as the Internet and the cellular networks, can be exploited to deliver RTK corrections and provide high-precision positioning and navigation.

The basic premise of relative (differential) positioning techniques such as RTK is that many of the sources of GNSS measurement errors including the frequency-dependent error (the ionospheric delay) are spatially correlated. By performing relative positioning between receivers, the correlated measurement errors are completely cancelled or greatly reduced, resulting in a significant increase in the positioning accuracy and precision.

Single-Frequency Challenges. Although RTK positioning is a well-established and routine technology, its effective implementation for low-cost, single-frequency L1 receivers poses many serious challenges, especially in difficult and degraded signal environments for GNSS such as urban canyons. The most serious challenge is the use of only the L1 frequency for carrier-phase integer ambiguity resolution and validation. Unfortunately, users with single-frequency capability do not have frequency diversity and many options in forming useful functions and combinations for pseudorange and carrier-phase observables. Moreover, observations from a single-frequency, low-cost receiver are typically “biased” due to the high level of multipath and/or receiver signal-tracking anomalies and also the low-cost patch antenna design that is typically used. In addition, in those receivers, measurements are typically contaminated with high levels of noise due to the low-cost hardware design compared to the high-end receivers. This makes the reliable fixing of the phase ambiguities to their correct integer values, for single-frequency users, a non-trivial problem. As a consequence, the reliability of single-frequency observations to resolve ambiguities on the fly in an operational environ
ment is relatively low compared to the use of dual-frequency observations from geodetic-quality receivers. Improving performance will be difficult, unless high-level noise and multipath can be dealt with effectively or unless ambiguity resolution techniques can be devised that are more robust and are less sensitive to the presence of biases and/or high levels of noise in the observations.

Traditionally, single-frequency RTK positioning requires long uninterrupted initialization times to obtain reliable results, and hence have a time-to-fix ambiguities constraint. Times of 10 to 25 minutes are common. Observations made at tens of continuous epochs are used to determine reliable estimates of the integer phase ambiguities. In addition, these continuous epochs must be free from cycle slips, loss of lock, and interruptions to the carrier-phase signals for enough satellites in view during the ambiguity fixing procedure. Otherwise, the ambiguity resolution will fail to fix the phase ambiguities to correct integer values. To overcome these drawbacks and be able to determine the integer phase ambiguities and thus the precise relative positions, observations made at only one epoch (single-epoch) can be used in resolving the integer phase ambiguities. This allows instantaneous RTK positioning and navigation for single-frequency users such that the problem of cycle slips, discontinuities, and loss of lock is eliminated. However, for single-frequency users, the fixing of the phase ambiguities to their correct integer values using a single epoch of observations is a non-trivial problem; indeed, it is considered the most challenging scenario for ambiguity resolution at the present time.

Instantaneous RTK positioning relies fundamentally upon the inversion of both carrier-phase measurements and code measurements (pseudoranges) and successful instantaneous ambiguity resolution. However, in this approach, the probability of fixing ambiguities to correct integer values is dominated by the relatively imprecise pseudorange measurements. This is more severe in urban areas and difficult environments where the level of noise and multipath on pseudoranges is high. This problem may be overcome partially by carrier smoothing of pseudoranges in the range/measurement domain using, for example, the Hatch filter. While carrier-phase tracking is continuous and free from cycle slips, the carrier smoothing of pseudoranges with an optimal smoothing filter window-width can effectively suppress receiver noise and short-term multipath noise on pseudoranges. However, the effectiveness of the conventional range-domain carrier-smoothing filters is limited in urban areas and difficult GNSS environments because carrier-phase measurements deteriorate easily and substantially due to blockages and foliage and suffer from phase discontinuities, cycle-slip contamination, and other measurement anomalies. This is illustrated in Figure 1. The figure shows that in a kinematic urban environment, frequent carrier-phase outages and anomalies occur, which cause frequent resets of the carrier-smoothing filter and hence carrier smoothing of pseudoranges suffers in robustness and effective continuous smoothing.

Figure 1. Satellite tracking and carrier-phase anomaly summary during the observation time-span. These data were collected in a dense urban environment in both static and kinematic mode. The superimposed red-points show epochs where carrier-phase observables are either missing or contaminated with cycle slips, loss of locks, and/or other measurement anomalies.

Doppler Frequency Shift. While carrier-phase tracking can be discontinuous in the presence of continuous pseudoranges, a receiver generates continuous Doppler-frequency-shift measurements. The Doppler measurements are immune to cycle slips. Moreover, the precision of the Doppler measurements is better than the precision of pseudoranges because the absolute multipath error of the Doppler observable is only a few centimeters. Thus, devising methods that utilize the precision of raw Doppler measurements to reduce the receiver noise and high-frequency multipath on pseudoranges may prove valuable especially in GNSS-challenged environments. Figure 2 shows an example of the availability and the precision of the receiver-generated Doppler measurements alongside the delta-range values derived from the C/A-code pseudoranges and from the L1 carrier-phase measurements. This figure also shows that frequent carrier-phase outages and anomalies occur while for every C/A-code pseudorange measurement there is a corresponding Doppler measurement available.

Figure 2. Plots of C/A-code-pseudorange-derived delta-ranges (top), L1 carrier-phase-derived delta-ranges (middle), and L1 raw receiver-generated Doppler shifts that are transformed into delta-ranges for the satellite PRN G18 during the observation time-span when it was tracked by the receiver (bottom).

Smoothing. A rich body of literature has been published exploring aspects of carrier smoothing of pseudoranges. One factor that has not received sufficient study in the literature is utilization of Doppler measurements to smooth pseudoranges and to investigate the influence of improved pseudorange accuracy on both positioning and the integer-ambiguity resolution. Utilizing the Doppler measurements to smooth pseudoranges could be a good example of an algorithm that maximally utilizes the information redundancy and diversity provided by a GPS/GNSS receiver to improve positioning accuracy. Moreover, utilizing the Doppler measurements does not require any hardware modifications to the receiver. In fact, receivers measure Doppler frequency shifts all the time as a by-product of satellite tracking.

GNSS Doppler Measurement Overview

The Doppler effect is the apparent change in the transmission frequency of the received signal and is experienced whenever there is any relative motion between the emitter and receiver of wave signals. Theoretically, the observed Doppler frequency shift, under Einstein’s Special Theory of Relativity, is approximately equal to the difference between the received and transmitted signal frequencies, which is approximately proportional to the receiver-satellite topocentric range rate.

Beat Frequency. However, the transmitted frequency is replicated locally in a GNSS receiver. Therefore, strictly speaking, the difference of the received frequency and the receiver locally generated replica of the transmitted frequency is the Doppler frequency shift that is also termed the beat frequency. If the receiver oscillator frequency is the same as the satellite oscillator frequency, the beat frequency represents the Doppler frequency shift due to the relative, line-of-sight motion between the satellite and the receiver. However, the receiver internal oscillator is far from being perfect and therefore, the receiver Doppler measurement output is the apparent Doppler frequency shift (that includes local oscillator effects). The Doppler frequency shift is also subject to satellite-oscillator frequency bias and other disturbing effects such as atmospheric effects on the signal propagation.

To estimate the range rate, a receiver typically forms an average of the delta-range by simply integrating the Doppler over a very short period of time (for example, 0.1 second) and then dividing it by the duration of the integration interval. Since the integration of frequency over time gives the phase of the signal over that time interval, the procedure continuously forms the carrier-phase observable that is the integrated Doppler over time. Therefore, Doppler frequency shift can also be estimated by time differencing carrier-phase measurements. The carrier-phase-derived Doppler is com
puted over a longer time span, leading to smoother Doppler measurements, whereas direct loop filter output is an instantaneous measure produced over a short time interval.

Doppler frequency shift is routinely used to determine the satellite or user velocity vector. Apart from velocity determination, it is worth mentioning that Doppler frequency shifts are also exploited for coarse GPS positioning. Moreover, the user velocity vector obtained from the raw Doppler frequency shift can be and has been applied by a number of researchers to instantaneous RTK applications to constrain the float solution and hence improve the integer-ambiguity-resolution success rates in kinematic surveying. In this article, a simple combination procedure of the noisy pseudorange measurements and the receiver-generated Doppler measurements is suggested and its benefits are examined.

Doppler-Smoothing Algorithm Description

Motivated both by the continual availability and the centimeter-level precision of receiver-generated (raw) Doppler measurements, even in urban canyons, a method has been introduced by the authors that utilizes the precision of raw Doppler measurements to reduce the receiver noise and high-frequency multipath on code pseudoranges. For more detail on the Doppler-smoothing technique, see Further Reading. The objective is to smooth the pseudoranges and push the accuracy of the code-based or both code- and carrier-based positioning applications in GNSS-challenged environments.

Previous work on Doppler-aided velocity/position algorithms is mainly in the position domain. In those approaches, the improvement in the quality of positioning is gained mainly by integrating the kinematic velocities and accelerations derived from the Doppler measurement in a loosely coupled extended Kalman filter or its variations such as the complementary Kalman filter. Essentially, these techniques utilize the well-known ability of the Kalman filter to use independent velocity estimates to reduce the noise of positioning solutions and improve positioning accuracy. The main difference among these position-domain filters is that different receiver dynamic models are used.

The proposed method combines centimeter-level precision receiver-generated Doppler measurements with pseudorange measurements in a combined pseudorange measurement that retains the significant information content of each.

Two-Stage Process. The proposed Doppler-smoothing process has two stages: (1) the prediction or initialization stage and (2) the filtering stage. In the prediction stage, a new estimated smoothed value of the pseudorange measurement for the Doppler-smoothing starting epoch is obtained. In this stage, for a fixed number of epochs, a set of estimated pseudoranges for the starting epoch is obtained from the subsequent pseudorange and Doppler measurements. The estimated pseudoranges are then averaged to obtain a good estimated starting point for the smoothing process. The number of epochs used in the prediction stage is the averaging window-width or Doppler-smoothing-filter length. In the filtering stage, the smoothed pseudorange profile is constructed using the estimated smoothed starting pseudorange and the integrated Doppler measurements over time. The Doppler-smoothing procedures outlined here can be performed successively epoch-by-epoch (that is, in a moving filter), where the estimated initial pseudorange (the averaged pseudorange) is updated from epoch to epoch. Alternatively, an efficient and elegant implementation of the measurement-domain Doppler-smoothing method is in terms of a Kalman filter, where it can run as a continuous process in the receiver from the first epoch (or in post-processing software, but then without the real-time advantage). This filter allows real-time operation of the Doppler-smoothing approach.

In the experiments described in this article, a short filter window-width is used. The larger the window width used in the averaging filter process, the more precise the averaged pseudorange becomes. However, this filter is also susceptible to the ionospheric divergence phenomenon because of the opposite signs of the ionospheric contribution in the pseudorange and Doppler observables. Therefore, the ionospheric divergence effect between pseudoranges and Doppler observables increases with averaging window-width and the introduced bias in the averaged pseudoranges become apparent for longer filter lengths.

Using the propagation of variance law, it can be shown that the precision of the delta-range calculated with the integrated Doppler measurements over time depends on both the Doppler-measurement epoch interval and the precision of the Doppler measurements, assuming that noise/errors on the measurements are uncorrelated.

Experimental Results

To validate the improvement in the performance and availability of single-frequency instantaneous RTK in urban areas, the proposed Doppler-aided instantaneous RTK technique has been investigated using actual GPS data collected in both static and kinematic pedestrian trials in central London. In this article, we only focus on the static results and the kinematic trial results are omitted. It is remarked, however, that the data collected in the static mode were post-processed in an epoch-by-epoch approach to simulate RTK processing.

In the static testing, GPS test data were collected with a measurement rate of 1 Hz. At the rover station, a consumer-grade receiver with a patch antenna was used. This is a single-frequency 16-channel receiver that, in addition to the C/A-code pseudoranges, is capable of logging carrier-phase measurements and raw Doppler measurements. Reference station data were obtained from the Ordnance Survey continuously operating GNSS network. Three nearby reference stations were selected that give different baseline lengths: Amersham (AMER) ≈ 38.3 kilometers away, Teddington (TEDD) ≈ 20.8 kilometers away, and Stratford (STRA) ≈ 7.1 kilometers away. In addition, a virtual reference station (VRS) was also generated in the vicinity (60 meters away) of the rover receiver.

Doppler-Smoothing. Before we present the improvement in the performance of instantaneous RTK positioning, the effect of the Doppler-smoothing of the pseudoranges in the measurement domain and comparison with carrier-phase smoothing of pseudoranges is given. To do this, we computed the C/A-code measurement errors or observed range deviations (the differences between the expected and measured pseudoranges) in the static mode (with surveyed known coordinates) using raw, Doppler-smoothed and carrier-smoothed pseudoranges. FIGURE 3a illustrates the effect of 100-second Hatch-filter carrier smoothing and FIGURE 3b shows a 100-second Doppler-smoothing of the pseudoranges for satellite PRN G28 (RINEX satellite designator) with medium-to-high elevation angle. The raw observed pseudorange deviations (in blue) are also given as reference. The quasi-sinusoidal oscillations are characteristic of multipath. Comparing the Doppler-smoothing in Figure 3b to the Hatch carrier-smoothing in Figure 3a, it can be seen that Doppler-smoothing of pseudoranges offers a modest improvement and is more robust and effective than that of the traditional Hatch filter in difficult environments.

Figure 3. Smoothed pseudorange errors (observed range deviations) using the traditional Hatch carrier-smoothing filter. Smoothing filter length in the experiments for both filters was set to 100 seconds. Satellite PRN G28 was chosen to represent a satellite at medium-to-high elevation angle.

Figure 3. Smoothed pseudorange errors (observed range deviations) using the Doppler-smoothing filter. Smoothing filter length in the experiments for both filters was set to 100 seconds. Satellite PRN G28 was chosen to represent a satellite at medium-to-high elevation angle.

Figure 4a illustrates carrier-phase Hatch-filter smoothing for low-elevation angle satellite PRN G18. In this figure, the Hatch carrier-smoothing filter reset is indicated. It can be seen that due to the frequent carrier-phase discontinuities and cycle slips, the smoothing has to be reset and restarted from the beginning and hardly reaches its full potential. In contrast, Doppler smoothing for PRN G18 shown in FIGURE 4b had few filter resets and managed effectively to smooth the very noisy pseudorange in some sections of the data.

Figure 4. Smoothed pseudorange errors (observed range deviations) and filter resets and filter length (window width) using the traditional Hatch carrier-smoothing filter. Smoothing filter length in the experiments for both filters was set to 100 seconds. Satellite PRN G18 was chosen to represent a satellite at low elevation angle as it rises from 10 to 30 degrees.

Figure 4. Smoothed pseudorange errors (observed range deviations) and filter resets and filter length (window width) using the Doppler-smoothing. Smoothing filter length in the experiments for both filters was set to 100 seconds. Satellite PRN G18 was chosen to represent a satellite at low elevation angle as it rises from 10 to 30 degrees.

Considering RTK in this analysis, we can demonstrate the increase in the success rate of the Doppler-aided integer ambiguity resolution (and hence the RTK availability) by comparison of the obtained integer ambiguity vectors from the conventional LAMBDA (Least-squares AMBiguity Decorrelation Adjustment) ambiguity resolution method using Doppler-smoothed pseudoranges with those obtained without Doppler-aiding in post-processed mode. The performance of ambiguity resolution was evaluated based on the number of epochs where the ambiguity validation passed the discrimination/ratio test. The ambiguity validation ratio test was set to the fixed critical threshold of 2.5 in all the experiments. In addition to the ratio test, the fixed solutions obtained using the fixed integer ambiguity vectors that passed the ratio test were compared against the true position of the surveyed point to make sure that indeed the correct set of integer ambiguities were estimated.

The overall performance of the single-epoch single-frequency integer ambiguity resolution obtained by the conventional LAMBDA ambiguity resolution method without Doppler-aiding is shown in Figure 5 for baselines from 60 meters up to 38 kilometers in length. In comparison, the performance of the single-epoch single-frequency integer ambiguity resolution from the LAMBDA method using Doppler-smoothed pseudoranges are shown in Figure 6 for those baselines and they are compared with integer ambiguity resolution success rates of the conventional LAMBDA ambiguity resolution method without Doppler-aiding. Figure 6 shows that using Doppler-smoothed pseudoranges enhances the probability of identifying the correct set of integer ambiguities and hence increases the success rate of the integer ambiguity resolution process in instantaneous RTK, providing higher availability. This is more evident for shorter baselines. For long baselines, the residual of satellite-ephemeris error and atmospheric-delay residuals that do not cancel in double differencing potentially limits the effectiveness of the Doppler-smoothing approach. It is well understood that those residuals for long baselines strongly degrade the performance of ambiguity resolution. Relative kinematic positioning with single frequency mass-market receivers in urban areas using VRS has also shown improvement.

Figure 5. Single-epoch single-frequency integer ambiguity resolution success rate obtained by the conventional LAMBDA ambiguity resolution method without Doppler-aiding.

Figure 6. Plots of integer ambiguity resolution success rates: single-epoch single-frequency integer ambiguity resolution success rate obtained by the conventional LAMBDA ambiguity resolution method without Doppler-aiding (in blue) and using Doppler-smoothed pseudoranges (in green).

Conclusion

In urban areas, the proposed Doppler-smoothing technique is more robust and effective than traditional carrier smoothing of pseudoranges. Static and kinematic trials confirm this technique improves the accuracy of the pseudorange-based absolute and relative positioning in urban areas characteristically by the order of 40 to 50 percent.

Doppler-smoothed pseudoranges are then used to aid the integer ambiguity resolution process to enhance the probability of identifying the correct set of integer ambiguities. This approach shows modest improvement in the ambiguity resolution success rate in instantaneous RTK where the probability of fixing ambiguities to correct integer values is dominated by the relatively imprecise pseudorange measurements.

The importance of resolving the integer ambiguities correctly must be emphasized. Therefore, devising innovative and robust methods to maximize the success rate and hence reliability and availability of single-frequency, single-epoch integer ambiguity resolution in the presence of biased and noisy observations is of great practical importance especially in GNSS-challenged environments.

Acknowledgments

The study reported in this article was funded through a United Kingdom Engineering and Physical Sciences Research Council Engineering Doctorate studentship in collaboration with the Ordnance Survey. M. Bahrami would like to thank his industrial supervisor Chris Phillips from the Ordnance Survey for his continuous encouragement and support. Professor Paul Cross is acknowledged for his valuable comments. The Ordnance Survey is acknowledged for sponsoring the project and providing detailed GIS data.

Manufacturer

The data for the trial discussed in this article were obtained from a u-blox AG AEK-4T receiver with a u-blox ANN-MS-0-005 patch antenna.

Mojtaba Bahrami is a research fellow in the Space Geodesy and Navigation Laboratory (SGNL) at University College London (UCL). He holds an engineering doctorate in space geodesy and navigation from UCL.

Marek Ziebart is a professor of space geodesy at UCL. He is the director of SGNL and vice dean for research in the Faculty of Engineering Sciences at UCL.

GPS World » Algorithms & Methods

GNSS PPP Workshop Early Registration Extended to May 3

Innovation: Getting at the Truth: A Civilian GPS Position Authentication System

Using White Noise as a Watermark

An Intrinsic GPS Watermark

System Architecture

C/A Residual Filter

Position Calculation

Hardware and Software

Performance Validation

Conclusion

Acknowledgments

Manufacturers

FURTHER READING

Innovation: Getting Along: Collaborative Navigation in Transitional Environments

The Concept

Field Experiments and Methodology

Image-Based Navigation

Data Analysis

Conclusions

Acknowledgments

Manufacturers

FURTHER READING

The Kinematic GPS Challenge: First Gravity Comparison Results

Innovation: Filling in the Gaps: Improving Navigation Continuity Using Parallel Cascade Identification

RISS/GPS Integration

Parallel Cascade Identification

Augmented Kalman Filter

Experimental Setup

Experimental Results

Conclusion

Acknowledgments

Manufacturers

FURTHER READING

Innovation: The Right Attitude: Experimenting with GPS on Board High-Altitude Balloons

In this month’s column, we look at how a team of Dutch and Japanese researchers is using GPS to determine the attitude of a payload launched from a high-altitude balloon.

Balloon Experiments

Theory of Attitude Determination

Flight Experiments

Future Work

Acknowledgments

Manufacturers

FURTHER READING

Innovation: Multipath Minimization Method

MIMICS Strategy

Testing

Concluding Remarks

Acknowledgments

Manufacturers

FURTHER READING

Innovation: Doppler-Aided Positioning: Improving Single-Frequency RTK in the Urban Enviornment

A look at how Doppler measurements can be used to smooth noisy code-based pseudoranges to improve the precision of autonomous positioning as well as to improve the availability of single-frequency real-time kinematic positioning, especially in urban environments.

Correction and Further Details

Single-Frequency RTK

GNSS Doppler Measurement Overview

Doppler-Smoothing Algorithm Description

Experimental Results

Conclusion

Acknowledgments

Manufacturer

FURTHER READING