Survey Perspectives – Early June 2008

June 5, 2008  - By
Image: GPS World
Image: GPS World

Is Dual-Frequency GPS — As We Know It — Becoming Obsolete?

On Friday, May 16, 2008, the Office of Space Commercialization issued a Notice for Public Comment. In it, the U.S. Department of Defense (DoD) proposes to discontinue supporting P(Y) codeless/semi-codeless on both GPS L1 and L2 frequencies on modernized satellites (Block IIR-M, Block IIF and Block IIIA/B/C) beginning December 31, 2020. After 2020, legacy dual frequency receivers may still work, but the DoD would no longer assure that P(Y) power levels and the navigation message would remain the same. Therefore, the civil GPS community has no assurance that legacy dual frequency receivers will operate as before.

Essentially, this means that every dual frequency receiver designed in the 1980’s, 1990’s and many in the early 2000’s would become virtually obsolete. In the interest of disclosure, that includes my own legacy real-time kinematic (RTK) system.

I caution you … it is very easy to rush to judgment regarding this proposal. When I first read it, my first response was “Whoa, dude, no way!” However, it’s important to take a deep breath and work your way through the logic. Your conclusion may be the same as your initial response, but at least you’ve thought it through. That being said, you must also realize that this is the first action, in the history of GPS, which will render a massive amount of GPS equipment obsolete.

What is Codeless/Semi-codeless Processing?

On L1, there is C/A code and P(Y) code. C/A is for civilian use, P(Y) for military use. On L2, originally there was no civilian code, only P(Y) for military use.

Back in the 1980’s, engineers in the commercial sector were trying to figure out a way to utilize L2 because it would significantly increase the receiver’s performance. Some really smart ones figured out how to track the encrypted P(Y) code on L1 and L2. Then they figured out how to cross correlate the measurements on L1 and L2 and voilà, the modern dual frequency receiver was born. This technique is used in virtually all dual frequency equipment sold today in the commercial (non-military) market.

All post-processing and RTK algorithms are based on using codeless/semi-codeless techniques of one sort or another.

Codeless/semi-codeless processing would not have been needed if L2C had been around on the original GPS satellites. In fact, even today there are only six satellites broadcasting L2C. Every satellite launched since 2005 (six of them to date) and each launched in the future will broadcast L2C, so eventually every satellite will.

What’s Being Proposed?

After December 31, 2020, the DoD proposes to discontinue supporting P(Y) on L1 and L2 for the commercial market on all modernized satellites (Block IIR-M, Block IIF, Block IIIA/B/C). Block IIA/IIR satellites will continue to operate as they do today. However, unlike today where Block IIA/IIR account for 25 of 31 operational satellites, the youngest Block IIR satellite will be 16 years old in 2020, if any still exist at all.

The DoD’s proposal assumes that most organizations will have upgraded their GPS equipment by 2020 and will be utilizing L2C (and other modernized signals), so that L1/L2 P(Y) codeless/semi-codeless processing won’t be needed any longer.

In their synopsis, the DoD states that GPS manufacturers have indicated that the user community needs approximately ten years to replace legacy GPS equipment with equipment capable of utilizing modernized GPS signals. You can read the DoD’s full proposal here. It contains a lot of pertinent background information.

Who’s Affected?

Unfortunately for those of us in the survey/engineering/construction/deformation monitoring/high-precision GIS industries, we would be the ones affected the most.

In real terms, this means that any dual frequency receiver not designed to use L2C will essentially become a paperweight after December 31, 2020. The receiver may still work after that date, but there is no assurance it will continue operating properly. The list of receivers affected is quite long and includes models from all major manufacturers, such as Trimble, Leica, Topcon, Magellan (Ashtech/Thales), and NovAtel among others. Check with the manufacturer of your equipment to determine if it is capable of utilizing L2C. If it’s not, then it’s considered a legacy receiver and would become obsolete.

Also, one should be careful and not assume that all receivers sold today are capable of utilizing L2C. Ask your dealer or the manufacturer of the equipment before your purchase.

No L1-only receivers, such as hand-held GPS units, car navigation systems, various tracking devices, GPS-enabled mobile phones, L1-only GPS mapping systems, and timing receivers are affected by this proposal. L1-only RTK receivers are not affected either. None of these use codeless/semi-codeless techniques. The exception is some of the newer GPS receivers designed for GIS data collection at the decimeter (or sub-foot) level. Although not actively marketed as such, these are dual frequency receivers and might be affected if this proposal is carried out.

It doesn’t take long for one to think about the thousands and perhaps tens of thousands of reference stations worldwide that will need to be replaced. Just the United States CORS (Continually Operating Reference Stations) network alone comprises more than 1,000 receivers. Granted, some are modernized receivers that may only need a minor update, but many others are legacy receivers that will need to be replaced or risk obsolescence. Those 1,000+ CORS receivers service thousands of users monthly. In April 2008 alone, the National Geodetic Survey reported that more than one million FTP requests were made for CORS data.

The DoD says that December 31, 2020 isn’t a “hard” date. In other words, GPS equipment using codeless/semi-codeless techniques may work just fine after December 31, 2020. What they are saying is that after December 31, 2020, they won’t guarantee they will not do something that will impact P(Y) code and subsequently prevent your receiver from performing like you’d expect.

Timing Is Everything

I think I understand the DoD’s logic. They are developing these modernized signals (L2C, L5, L1C) that should be commonplace by the time 2020 rolls around. Continued support of semi-codeless would interfere with some new features they want to play with on the military side of GPS. Why support the legacy stuff when the new stuff is better anyway?

The first issue I thought of is what the status of the GPS constellation will be in 2020. Today, GPS users have 31 satellites to work with. As high-precision users, we need every one of those. Just last week, I was stuck in the middle of a GPS fieldwork day waiting for a sixth satellite to come into view so I could continue my RTK work.

The DoD is still only committed to a 24-satellite constellation, but they’ve been spoiling us with 30 or more for quite awhile now. It would be hard to go back.

So, of course, I started doing the math to guesstimate how many satellites will be operational in 2020, based on information provided in the DoD’s proposal and other sources. The DoD’s proposal states that they expect 24 satellites to be broadcasting L2C by 2016 and 24 satellites will be broadcasting L5 by 2018. We know that eight Block IIR-M satellites were built and twelve Block IIF satellites will be built. We also know that, as announced last month, eight Block IIIA, eight Block IIIB and eight Block IIIC satellites will be built. From this information, one can deduce that in 2016 the constellation will look something like this:

  • 8 ea. Block IIR-M satellites broadcasting L1 C/A, L2C
  • 12 ea. Block IIF satellites broadcasting L1 C/A, L2C, L5
  • 4 ea. Block IIIA satellites broadcasting L1 C/A, L2C, L5, L1C

The above lists and the ones following the paragraph below are the civil signals. Of course we can assume that each satellite is still broadcasting military P(Y) code and M-code on L1/L2.

Based on the exceptional life span of legacy Block IIA/IIR GPS satellites, there would still be approximately six to eleven of them still broadcasting L1 C/A code. By 2018 we can deduce that the remaining four Block IIIA satellites and four new Block IIIB satellites will have been launched, giving a total of 24 satellites broadcasting L5. The constellation would look something like this:

  • 8 ea. Block IIR-M satellites broadcasting L1 C/A, L2C
  • 12 ea. Block IIF satellites broadcasting L1 C/A, L2C, L5
  • 8 ea. Block IIIA satellites broadcasting L1 C/A, L2C, L5, L1C
  • 4 ea. Block IIIB satellites broadcasting L1 C/A, L2C, L5, L1C

There should also be a handful of remaining Block IIR satellites available for service that are still broadcasting L1 C/A code.

If I’ve done the math right and the DoD keeps this schedule, that’s not bad; not bad at all. In 2016, there would be somewhere between 30 and 35 operational satellites. In 2018, there would be somewhere around 37 operational satellites. In terms of sheer numbers, that’s equal to or better than where we are today.

After working through this, I think it’s obvious that we will be better off than we are today with respect to the satellite constellation. As I’ve written before, triple frequency receivers (L1, L2, and L5) will be far superior to today’s dual frequency receivers that utilize codeless/semi-codeless techniques. If you add Galileo on top of that, it’s a no-brainer. I look forward to the day that I’m in the field and have 20 or more GPS/Galileo satellites in view when just last week I was struggling to find six.

Lastly, in case you missed it ,the DoD stated that if the new satellite schedule were delayed, they would reassess the codeless/semi-codeless sunset date.

It’s All About the $$$

Alas, at the end of the day, this is where it’s going to hurt the user community the most.

I think nearly everyone’s heard of the Spring 2009 sunset date for analog television in the US. On that date, full-power television stations will stop broadcasting on analog channels, rendering analog television sets obsolete. Congress was so concerned about consumer backlash that they are subsidizing analog-digital conversion boxes to the tune of $890 million, based on a price of $50 to $70 each. To put it in perspective, that doesn’t even cover the cost of 3-meter L1/L2 antenna cable.

We are going to get hit in the wallet … hard.

An argument in support of all this states that triple frequency GPS equipment will be much cheaper at that time. I agree it will be cheaper, but we are still talking about tens of thousands of dollars. The survey/engineering/deformation monitoring/high-precision GIS market is relatively limited in size, is highly technical, and requires complex software, training, and technical support. It’s not like spending $150 at WalMart for a Garmin receiver that you can figure out without reading a manual.

Another argument in support of the DoD’s proposal is that 12 years gives us plenty of time to enjoy a solid return on investment (ROI) on our current equipment. While I follow that logic, I’ve seen a lot of GPS equipment in the field that is 15 years to 20 years old. The stuff just keeps working.

I’ve been amazed that my RTK system, based on 12-year-old technology, still cranks up like it did the first time I used it. Maybe it’s more of an emotional feeling than anything else, but as much as I work through the logic, it’s hard to swallow that my $40,000 system has a date with the trash bin.

I know codeless/semi-codeless dual frequency GPS is the core technology for thousands of small to medium sized businesses around the world. Outside of vehicles, GPS equipment may have been the largest capital investment for them. For those who made that purchase in the last couple of years, the codeless/semi-codeless obsolescence is not something they want to hear about even if it is 12 years away.

The U.S. Department of Commerce has done a quick survey and prediction, to get a rough idea of the dollar-value of equipment that will need to be upgraded sometime toward the 2020 time frame. Its figure for the economic impact is $1.3 billion to $1.7 billion dollars per year if semi-codeless were taken away today. That’s the estimated value of at least 200,000 semi-codeless receivers out in the field today, a figure that it
acknowledges to be conservative, by the way.

According to the DoC analyst, if semi-codeless were taken away in five years, in the year 2012, using some growth rates and extrapolating, the estimate would grow to between 373,000 and half a million users worldwide, and the economic loss on a worldwide basis would be between $3.6 billion and $4.8 billion; within the U.S. alone, that portion would be between $1.1 billion and 1.9 billion.

These figures formed the rationale for a proposed decision to push the discontinue date out to 2020, to give manufacturers and the user base adequate time to re-equip for using L2C and L5. Incidentally, a full-length interview on this topic with a senior DoC analyst and advisor to the National Coordination Office for Space-Based Positioning, Navigation, and Timing will appear in the July print edition of GPS World magazine.

I don’t have an answer on the money issue. For the manufacturers and dealers, it’s going to a salesman’s dream, not unlike Y2K and GPS Week Rollover were. For the user community, it’s going to taste sour no matter how it goes down.

You Have Your Chance: the DoD Is Listening

It’s important to note that the DoD is seeking comments from everyone around the globe. The potato farmer in Argentina, the land surveyor in Australia, the geodetic surveyor in the United States, and the engineer in Denmark are all encouraged to comment. GPS is a tool that knows no boundaries.

Col. Mark Crews, the U.S. Air Force GPS Chief Engineer, says the GPS Wing is keenly interested in public comment on the proposal. The Air Force estimates there are approximately 250,000 worldwide users of dual frequency receivers that use P(Y) codeless/semi-codeless.

“We are trying to do everything absolutely the right way in pre-notifying everybody in the world. If anybody has any concerns, please notify us,” said Crews. “We are taking every precaution to transition semi-codeless users to civil coded signals in a stable, measured, and transparent manner by 2020. That’s why we’re taking action now to pre-notify semi-codeless users worldwide and ask for their input by means of the Federal Register’s request for comments.”

Crews further says that the Air Force recognizes that that dual frequency GPS receivers are a “huge business.” It recognizes that these receivers “play an extremely positive role in survey, agriculture, and all high-accuracy augmentation systems. We are bending over backwards until we have at least two other civil signals, being L2C and L5, on 24 satellites in time for people to transition,” he said.

The US Department of Commerce (DoC), on behalf of the DoD, is seeking public comments on the codeless/semi-codeless sunset proposal. Time is short though. You have until June 16, 2008 to submit your comments. I think that’s a mistake; it’s not enough time. They should allow at least 90 days so the word has a chance to spread.

All comments submitted are a matter of public record and can be viewed by anyone at http://www.space.commerce.gov/gps/semicodeless/. As of June 1, 2008, there have been no comments posted and we are half way through the 30-day comment period already.

That concerns me.

Clarifications/Corrections to The Last Column Regarding L5

In my last column I included a listing of satellite models and signals they broadcast. An astute reader was quick to point out two omissions, and I discovered an error as well. First, I neglected to list P(Y) on L1, which is especially important to note, given the subject of this newsletter.

Second, I listed Block I/II/IIA as one group. There are no Block I/II operational satellites any longer. There is only Block IIA/IIR. For complete clarification and for no other reason than I’ve intended to do this for awhile now, I’ve provided a comprehensive table of operational GPS satellites below.

Lastly, I stated last time that there are 26 Block IIA/IIR satellites broadcasting. There are actually 25. Below is a complete list of operational satellites.

PRN

MODEL

OPERATIONAL

PLANE/
SLOT

CIVIL
SIGNALS

MILITARY SIGNALS

9

Block IIA

July 20, 1993

A1

L1 C/A

L1 P(Y), L2 P(Y)

31

Block IIR-M

Oct. 13, 2006

A2

L1 C/A, L2C

L1 P(Y), L1M, L2 P(Y), L2M

8

Block IIA

Dec. 18, 1997

A3

L1 C/A

L1 P(Y), L2 P(Y)

7

Block IIR-M

Mar. 15, 2008

A4

L1 C/A, L2C

L1 P(Y), L1M, L2 P(Y), L2M

<

/td>

25

Block IIA

Mar. 24, 1992

A5

L1 C/A

L1 P(Y), L2 P(Y)

27

Block IIA

Sept. 30, 1992

A6

L1 C/A

L1 P(Y), L2 P(Y)

16

Block IIR

Feb. 18, 2003

B1

L1 C/A

L1 P(Y), L2 P(Y)

30

Block IIA

Oct. 1, 1996

B2

L1 C/A

L1 P(Y), L2 P(Y)

28

Block IIR

Aug. 17, 2000

B3

L1 C/A

L1 P(Y), L2 P(Y)

12

Block IIR-M

Dec. 13, 2006

B4

L1 C/A, L2C

L1 P(Y), L1M, L2 P(Y), L2M

5

Block IIA

Sept. 28, 1993

B5

L1 C/A

L1 P(Y), L2 P(Y)

None

None

None

B6

None

None

6

Block IIA

Mar. 28, 1994

C1

L1 C/A

L1 P(Y), L2 P(Y)

3

Block IIA

April 9, 1996

C2

L1 C/A

L1 P(Y), L2 P(Y)

19

Block IIR

April 5, 2004

C3

L1 C/A

L1 P(Y), L2 P(Y)

17

Block IIR-M

Nov. 13, 2005

C4

L1 C/A, L2C

L1 P(Y), L1M, L2 P(Y), L2M

None

None

None

C5

None

None

29

Block IIR-M

Jan. 2, 2008

C6

L1 C/A, L2C

L1 P(Y), L1M, L2 P(Y), L2M

2

Block IIR

Nov. 22, 2004

D1

L1 C/A

L1 P(Y), L2 P(Y)

11

Block IIR

Jan. 3, 2000

D2

L1 C/A

L1 P(Y), L2 P(Y)

21

Block IIR

April 12, 2003

D3

L1 C/A

L1 P(Y), L2 P(Y)

4

Block IIA

Nov. 22, 1993

D4

L1 C/A

L1 P(Y), L2 P(Y)

24

Block IIA

Aug. 30, 1991

D5

L1 C/A

L1 P(Y), L2 P(Y)

None

None

None

D6

None

None

20

Block IIR

June 1, 2000

E1

L1 C/A

L1 P(Y), L2 P(Y)

22

Block IIR

Jan. 12, 2004

E2

L1 C/A

L1 P(Y), L2 P(Y)

10

Block IIA

Aug. 15, 1996

E3

L1 C/A

L1 P(Y), L2 P(Y)

18

Block IIR

Feb. 15, 2001

E4

L1 C/A

L1 P(Y), L2 P(Y)

32

Block IIA

Dec. 12, 1990

E5

L1 C/A

L1 P(Y), L2 P(Y)

None

None

None

E6

None

None

14

Block IIR

Dec. 10, 2000

F1

L1 C/A

L1 P(Y), L2 P(Y)

15

Block IIR-M

Oct. 31, 2007

F2

L1 C/A, L2C

L1 P(Y), L1M, L2 P(Y), L2M

13

Block IIR

Jan. 31, 1998

F3

L1 C/A

L1 P(Y), L2 P(Y)

23

Block IIR

July 9, 2004

F4

L1 C/A

L1 P(Y), L2 P(Y)

26

Block IIA

July 23, 1992

F5

L1 C/A

L1 P(Y), L2 P(Y)

None

None

None

F6

None

None

This article is tagged with and posted in Opinions, Survey