Leadership Talks - The A-GNSS Backbone

December 1, 2007 By: Alan Cameron GPS World

Jari Syrjarinne (JS), technologist in GNSS systems research at Nokia Technology Platforms, spoke with editor Alan Cameron (AC). Nokia recently offered $8.1 billion to acquire digital mapper Navteq.

AC: How will the outbreak of map-based navigation and location-based services change the way we use mobile terminals?

JS: Mapping and location-based services (LBS) form one of the cornerstones of Nokia’s Internet services strategy. In addition to music, games, videos, and more, maps and location-based services make up a key service area we’re focusing on. In fact, I see that location and context information is a key component of our future Internet services including search and navigation, photos, and videos as well as presence and communities. Our vision is that location information helps to build the next phase in Web 2.0 (or whatever you want to call the next phase) with context-sensitive services.

AC: Is accuracy as important to LBS as availability?

JS: High-accuracy (centimeter/decimeter-level) positioning opens the door to many new and interesting applications and games and offers the best possible location experience. Accuracy and availability are equally important in the attempt to exceed customer expectations.

AC: What can you tell us about your work at Nokia in harnessing carrier-phase positioning, new satellite constellations, and hybrid use with non-GNSS technologies?JS: GNSS and assisted GNSS (A-GNSS) will be the backbone of global location technologies. GNSS and A-GNSS provide a unique opportunity to evolve handset location technology. Nokia considers GNSS as a whole with the aim of ensuring a seamless user experience, that is, not taking GNSS as individual, different satellite systems (AGPS, A-GLONASS, A-Galileo, and so on). This very same approach has also been visible in the A-GNSS standardization where Nokia together with a large number of GNSS industrial players (SiRF, Broadcom [GlobalLocate], Motorola, Alcatel-Lucent, Nokia Siemens Networks) have been able to introduce a generic A-GNSS solution that is ready for the future satellite systems.

Regional systems also play an important part in the GNSS/A-GNSS game, especially in Asia and Pacific regions (GAGAN, MSAS, QZSS). These are considered as an essential part of GNSS and A-GNSS.

Implementation of complementing location technologies is the key to ensuring universal coverage. Wireless local area network (WLAN) location is a very promising technology to provide location inside buildings and support in urban canyon environments. Motion sensors can give additional direction/movement information when GNSS is not available, and they also offer new power-saving opportunities. Hybrid positioning with non-GNSS technologies will be important to offer good location experience everywhere.

AC: How are you addressing “the toughest equation” — delivering positioning performance while reducing cost and power consumption at every step?

JS: GNSS receiver power consumption, size, and cost need to be at the levels that make them feasible for handset integration. This is only possible by guiding the receiver manufacturers in the right direction, increasing competition, and simplifying the GNSS receiver/solution architecture to provide optimal partitioning of functionalities between the GNSS hardware and GNSS software. In order to accomplish this, Nokia has taken the lead and defined a specification between GNSS hardware and GNSS software called the Nokia GNSS Measurement Engine Interface (MEIF).

Nokia MEIF aims to introduce a common low-level GNSS hardware interface for the mobile handset/GNSS industry allowing flexible feature and GNSS support. For the first time it is now possible to converge into a common GNSS hardware message interface which gives all receivers the same baseline upon which to test and compare the performance under the same conditions: for example, to qualify sometimes very unprecedented sensitivity and accuracy figures. A common interface also makes it possible for the handset manufacturers to use multiple chipset vendors and, if required, to easily change from one vendor to another without considerable modification to the handet software and software interfaces.

The initial version of Nokia MEIF specification (v 1.0) was released to Forum Nokia in November 2006. It very quickly gained a lot of popularity and currently (12 months later) there are already more than 10 licensees, and the number is still growing fast.

Nokia MEIF does not strictly dictate any specific receiver architecture, but merely defines the logical message interface between real-time and non-real time functions of the GNSS receiver. However, the recommended partitioning is for a measurement engine (ME) hardware chip and positioning engine (PE) software running on the host processor. This partitioning might change in the future due to pressure to integrate GNSS hardware with other radio technologies.

AC: “In LBS, Google Earth is a fantastic platform for all kinds of services.” Can you expand a bit on your statement at the Civil Global Positioning System Service Interface Committee (CGSIC)?

JS: Google Earth, as well as Nokia Maps, are the platforms to build and launch location-based services and other as yet unseen location-based applications and games. The point is that there need to be versatile and easy-to-use map-based platforms where the end user can add and share content and services. These kinds of platforms are the enablers for the LBS.

AC: You said that your number two GNSS priority is GLONASS, after GPS. Can you explain why? If they stick to FDMA? Or even if they change to or add CDMA signal, this will take some time. Does the GLONASS schedule still look good to you?

JS: GNSS implementation in handsets is dependent upon satellite system deployment. Maturity and availability of the various GNSS satellite systems will drive the adoption of GNSS. The GLONASS schedule at the moment is years ahead of Galileo deployment.GLONASS FDMA from the receiver point of view is only a challenge, not an obstacle. Once GLONASS + GPS co-existence and support in the mass market-grade receivers is solved, addition of any other GNSS will be child’s play.

GLONASS will offer 24 additional satellites by the end of 2010, and GLONASS + GPS receivers will naturally outperform GPS-only receivers in urban areas, which will offer an opportunity for enhanced location experience and product differentiation.

GLONASS CDMA in L5 is far in the future. GNSS L5 will not probably be supported in the handsets until GPS supports L5. True dual-frequency receivers are not really feasible due to the additional cost delta.

AC: You mentioned sub-decimeter accuracy for LBS. How would that be useful or valuable in LBS?

JS: There is always a need for better performance. Once A-GPS becomes mainstream and LBS takes off on a large scale, the appetite for more accuracy will grow. Navigation from door-to-door, locating people in a crowd, augmented reality, and geocaching are just a few examples that will benefit from the high accuracy. There really are not any major technical obstacles to improve the accuracy down to decimeter, even down to sub-decimeter level, to get the last drop of performance out of GNSS.

AC: Nokia recently launched its own personal navigation device (PND). Is this a sign that PNDs and phones are in fact converging? What will the future device look like and be capable of?

JS: Handsets with maps and navigation capability will overlap with PNDs. In some sense, these devices are converging because the “navigator in your pocket” will become “your navigator in car” due to ease of use and portability. It is, however, difficult to predict what the future will be.

AC: How will greater accuracy in the handset make things better? How is the new A-GNSS wireless standard that you proposed, and that has been accepted, facilitating this?

JS: The original Nokia A-GNSS proposal contained the necessary messages and support to enable mobile-based carrier-phase positioning, regional ionosphere and troposphere models, and the new Keplerian orbit models, from GPS CNAV-2, for various high-accuracy methods. Unfortunately, the current A-GNSS approved version, that is, by 3GPP, does not have all the means originally proposed due to stiff resistance and objection by a few companies. The standard does however have placeholders for future inclusion of the “left-out” messages with support for multi-frequency measurements, multi-frequency assistance, and also messages to request and transmit carrier-phase measurements from the phone to the network. The A-GNSS work continues, and hopefully the next release will have the support for the “left-out” features.

AC: Is Nokia making its own GPS chips, or are you buying from another supplier?

JS: Nokia is driving solutions based on Nokia MEIF and working closely with multiple GPS chip suppliers.

Bookmark it: digg propeller del.icio.us technorati yahoo facebook

Add Comment