Survey & Construction Newsletter, Early November 2008

Questions from the Webinar:
Is Dual-Frequency GPS — As We Know It — Becoming Obsolete?

First of all, I’d like to thank everyone who attended my webinar a couple of weeks ago. I received many, many questions during the presentation and answered a few of these during the event, but there was no way I could handle them. But most were very good questions that deserve answers, so I’ve devoted this column to answering those I couldn’t address at the time.

Oh yeah, kudos to you who attended; it was the most well-attended webinar to date for GPS World! I really enjoyed it and look forward to the next one in February or so of 2009. The focus on this last webinar was about the Department of Defense decision to discontinue supporting P(Y) on GPS L1 and L2 for civilian users after December 31, 2020.

You can view the archived presentation here.

Col. Mark Crews (ret.), former GPS Chief Engineer, is now retired but was kind enough to comment on some of the following questions that were submitted during the webinar, along with Don Jewell, GPS World’s Military and Government editor, and Richard Langley, the magazine’s Innovation editor.

I should note that it has come to my attention that it is possible the U.S. government might be able to create a “work-around” before 2021 so that the Civil P(Y) sunset date becomes a non-issue. In other words, your legacy dual frequency GPS receivers may end up operating past December 31, 2020 without any problems. Be assured that I will stay on top of this issue and keep you (the readers) up-to-date on any changes regarding Civil P(Y) sunset.

Question #1: Do you think that there will be a program to convert units to new technology like the television change to digital?

Gakstatter: With respect to a government subsidy of sorts? I don’t believe there will be.

Question #2: How do these codes affect the data gathered? And how is the new code going to differ?

Gakstatter: I assume you’re referring to post-processing. L2C and L5 are completely new codes so more data is being collected. It really won’t affect that way you collect data in terms of the user interface. Length of data collection sessions will probably be shorter.

A couple of years ago, I wrote an article about the benefit of the new codes. Some of the information in the article is dated now, but it is still fundamentally relevant. You can view the article here.

Question #3: Explain how civilian receivers such as the Ashtech Z-12 tracks the P(Y) code.

Gakstatter: I can’t speak specifically as to how Ashtech’s Z-Tracking technology works, but fundamentally I believe it reconstructs the full L2 carrier wavelength in addition to cross-correlating the P-code with L1 and L2.

Question #4: How will this affect L1-only receivers using base corrections and getting RTK accuracy?

Gakstatter: It won’t affect L1-only receivers.

Question #5: Will the new codes be better under heavy canopy or forested areas?

Gakstatter: L2C and L5 should improve operation in and around trees, but that’s not necessarily a good thing. Don’t be fooled into thinking you can operate reliably at the centimeter level under heavy tree canopy. That will never happen with only GPS. As I stated in the webinar, the real solution for that environment is the integration of other technologies such as pseudolites, inertial navigation, gyros, lasers, etc. I think those technologies will eventually be small and cheap enough to integrate into a GPS/GNSS receiver to allow seamless operation in GPS-impossible environments at the centimeter level.

Also, more help than L2C and L5 would be the addition of more satellites via GLONASS and Galileo. In this case, I believe that quantity trumps quality. Yes, L2C and L5 are better, but it doesn’t matter how good they are if the receiver can’t track them because of obstructions.

Question #6: Are these upgradeable or will we have to buy complete new equipment sets?

Gakstatter: It depends on the product you have and the level of manufacturer support. If you recently purchased a GPS L1/L2 receiver, I would suspect the manufacturer will probably offer an upgrade to at least L2C without having to completely replace your receiver.

If your L1/L2 is several years old and the manufacturer has discontinued support of that product, then the best you can hope for is some sort of trade-in credit towards the purchase of a new system.

Question #7: Would a receiver that could track L1, L2, L2C, and L5 obtain a worse fix on just L2C and L5 alone?

Gakstatter: I have no idea. In theory, it should perform better than the L1/L2 receivers of today because of better code structure, increased signal strength and better ability to mitigate ionospheric error. The frequency separation of L2 and L5 is not as great as that between L1 and L2 (or L5), so the dual-frequency iono correction potentially might not be quite as good. However, there won’t be enough satellites broadcasting L2C/L5 for at least another decade, so we won’t know until then!

Question #8: Are receiver providers creating units that can be upgraded, or will we be looking at a complete setup replacement (as soon as we are using the stated three codes)?

Gakstatter: Most major manufacturers of survey receivers have models that can track and use L2C, since that’s a signal that’s being broadcast today. Some say certain models are L5-ready. Likely, those will need to be tweaked with firmware when the actual L5 signal broadcasts.

Question #9: Is there a simple-to-understand one page document that may explain all of this to a procurement agent so they know what to consider when purchasing GPS equipment? Does the same exist to pass onto engineers?

Gakstatter: Hmmm … not that I know of, so I’ll try to create one. Email me in a month or so at egakstatter@questex.com.

Question #10: Will this change have any effect on our equipment before 2010?

Gakstatter: No. The semi-codeless sunset date is set for December 31, 2020. The only changes happening before 2010 are that a couple of more Block IIR-M satellites will be launched. The IIR-M broadcasts L2C in addition to L1 C/A. Also, in 2009 the first Block IIF satellite will be launched. The Block IIF will broadcast L2C and L5 in addition to L1 C/A.

Question #11: Will there be new criteria for CORS?

Gakstatter: The National Geodetic Service (NGS) is updating their GNSS strategy and moving toward supporting all broadcast signals (GPS and others). You may want to take a look at their five year and ten year plans.

Question #12: Is there likely to be a gradual degradation in the L2 signal availability in the years approaching the sunset date as the older satellites are lost from the system and replaced by the newer birds?

Gakstatter: I defer to the GPS World’s military and government editor Don Jewell and retired GPS Chief Engineer Col. Mark Crews (ret.). Both say there will be no degradation of the L2 signal, gradual or otherwise.

Question #13: I had read an article that basically stated more frequencies would yield higher accuracies, especially vertical, over more satellites. Is this an accurate statement?

Gakstatter: I don’t believe just adding more signals to the same number of satellites will significantly improve accuracy. I believe that more satellites (thus improved satellite geometry) is the best way to improve accuracy, especially vertical. With a full constellation of GPS, GLONASS and Galileo, the number of satellites in view and the PDOP numbers would be incredibly good for high precision users.

Comment from Dr. Richard Langley, Editor of GPS World’s Innovation column:

Recall that delta-subscript v = VDOP x delta-subscript p where delta-subscript p is the measurement accuracy (pseudorange or carrier phase), if we can reduce delta-subscript p, then we can also reduce delta-subscript v. So yes, reducing VDOP with more satellites will help more but improvements in signal structure and receiver technology will also help. L5 signals, for example, should have lower multipath contamination and also less noise at low elevation angles.

Question #14: Why does L5 require a new antenna?

Gakstatter: It is broadcast on a significantly different frequency (1176.4 5Mhz) than L1 (1575.42 Mhz) and L2 (1227.60 Mhz).

Question #15: What effect will this have on static processed data?

Gakstatter: None with respect to the user interface. Behind the scenes, more data is collected and the algorithms of the post-processing software will change significantly. You may see a slight improvement in accuracy. .

Question #16: So how do you make yourself ready?

Gakstatter: I’ll be writing more about this in the future. Like I said in my webinar, twelve years from now is a long time. There is no rush to take action.

Question #17: With reference to my last question (Question #13), presumably no more L2 birds are being launched in the future; what is their expected lifespan?

Gakstatter: All planned GPS satellites include L2. It’s important to note that L2 isn’t being rendered obsolete. The government is simply reserving the right to change P(Y) on L2 which was originally designed for the military.

L2C code is the future of L2 for civilian users.

Question #18: Do you expect the government/public/DOT VRS CORS systems to be upgraded and established with L2C and/or L5 by the sunset date for the L1/L2 signal?

Gakstatter: I’d refer you to the NGS 10 year plan with respect to CORS. Otherwise, it is up to each individual public or private network (RTK or otherwise) to ensure their network is prepared to handle the sunset date of December 31, 2020.

Question #19: Let’s say I have my semi-codeless system L1/L2 and everything is fine on January. 1, 2021, but on January 2, 2021 I have an issue. What’s my alternative at that point?

Gakstatter: You might be able to use L1-only for post-processing, but you may even have an issue with that.

Question #20: If not planning to already, let listeners know what to spec now in new purchases to ensure equipment will work past 2020.

Gakstatter: I will work on that.

Question #21: Do antennas (ex. Trimble Zephyr) become obsolete, or just the receiver board itself? Or do both components become obsolete?

Gakstatter: It depends on which Zephyr model you are referring to. The newest Zephyr 2 supports:

• GPS: L1, L2, L5
• GLONASS: L1, L2, L3
• Galileo: E1, E2, E5, E6

This antenna will not be affected. In fact, all GPS dual-frequency antennas of today will be fine in tracking L1 and L2C. Only if you desire to utilize L5 might you need a new antenna.

Question #22: What happens to the old satellites? Do they just burn up as their orbit degrades and they approach earth?

Gakstatter: I deferred to the GPS World’s military and government editor Don Jewell and retired GPS Chief Engineer Col. Mark Crews (ret.):

There are definitely some people who believe that we should de-orbit all our satellites, but unless the satellite is just a few hundred miles above us in a Low Earth Orbit (LEO), it is simply not possible. The GPS satellites orbit in the Medium Earth Orbit (MEO) regime which is 20,200 kilometers (~12,000 miles), on average, above the surface of the Earth. When a satellite becomes too old or fails for some reason, it is boosted into a slightly higher orbit plane which puts it out of the way of any operational GPS satellites.

In recent years, through solid engineering and strategic thinking, some of the GPS satellites that were still functional, but would have normally been boosted up to a higher orbit, have been left in the operational MEO orbit plane and put to sleep or in standby mode for future use. Just recently some of these older GPS satellites have been reactivated. If we were to allow the satellites orbit to naturally decay from MEO, we will have been in our graves for thousands of years by the time they reach the Earth’s atmosphere, where they would burn-up on reentry.

Further comment from Dr. Richard Langley:

There is also the potential for collisions of dead satellites from different constellations (GPS, GLONASS, Galileo, etc.) as a result of the satellites drifting out of their assigned orbit bands over the next 100 years or so. The GLONASS folks are studying this and it was mentioned during Sergey Revnivykh’s presentation at the ION GNSS 2008 CGSIC meeting.

Question #23: Are there receivers on the market now anticipating the change?

Gakstatter: Yes, there are many survey receivers on the market right now that can utilize L2C and are prepared for L5. They are typically the premium-priced receivers offered by the manufacturers.

Question #24: Can you give a brand/model example of a GIS sub-foot receiver that might be affected?

Gakstatter: Trimble GeoXH, ProXH.

Question #25: Will L1 receivers using real time reference networks (e.g. Trimble GNSS VRS) be affected?

Gakstatter: I don’t believe there will be any affect.

Question #26: How will the CORS network be affected?

Gakstatter: Please refer to the NGS 10 year plan. I believe the NGS is preparing well for the transition. However, NGS doesn’t have control over most of the CORS stations. That’s a different story.

Question #27: Can you talk in general about accuracy comparison of the new L2C and L5 capabilities with current legacy RTK accuracy? Will we still need base stations?

Gakstatter: I touched on this in the webinar a bit. I don’t believe you’ll see a substantial increase in accuracy. I think you’ll see a substantial increase in reliability and robustness of the positions.

The way to achieve greater accuracy (vertical in particular) is more observations (eg. GLONASS and Galileo).

Question #28: Would GPS World plan on researching GPS manufacturers and their products that DO NOT support L2C at this time? We sell NovAtel OEMV and they have L2C, but what about the OEM4 we sold two years ago? An article on this would be nice as I can’t find it on their spec sheet.

Gakstatter: I think this is a good idea and the subject has been raised before. I’m not sure when it will happen, but I believe I will do something along these lines.

Question #29: Do you have any recommendations for a small startup land survey company? Wait for the new technology to come out or purchase what’s out there now?

Gakstatter: Well, I think it’s more of a business issue than a technology issue. I’ll make a bunch of assumptions when answering. Personally, I’d try to keep your capital investment as low as possible at this point.

If you only need post-processing, then a pair of L1-only static receivers is a relatively small investment (well under $10,000). Or, one GPS dual frequency receiver will do, and use an online positioning service like OPUS.

If you need RTK (real-time, centimeter-level positioning for staking/topo), then the price tag goes up. Do you have access to an RTK network? If so, then you only need an RTK rover and a data plan from a wireless provider like Sprint, AT&T, T-Mobile, etc.

If you need RTK and no RTK Network access, then you’ll need a RTK reference station also. Again, the price tag goes up. Another piece of equipment to consider (maybe in lieu of GPS) would be a robotic total station.

It all depends on what kind of projects your company will be involved in the majority of the time.

Question #30: Comment: GPS tech changes significantly every 1.5 to 2 years, while the useful lifetime of a receiver is on the order of 6 to 8 years. Proper planning will make this a non-issue.

Gakstatter: In general, I agree there’s plenty of time to plan for the transition. However, the fact is that some GPS equipment purchased in the early 1990’s is still working today and some legacy equipment purchased today will still be operating twelve years from now. This is especially true for survey receivers because the price tag is so high.

Question #31: Will my L1/L2 receivers still be able to collect L1 data for static computations?

Gakstatter: I believe they will, but caveat emptor.

Question #32: If I upgrade to L2C and not L5, what will my limitations be?

Gakstatter: You will be no worse off than you are today.

Question #33: Can we expect an increase in vertical positioning accuracy with the new L2C and L5 frequencies?

Gakstatter: I don’t believe so. For better vertical accuracy, the best bang for your buck will be more satellites (eg. GLONASS, Galileo). However, as stated in the answer to Question #13, there will be some gain in accuracy due to improvements in the code structure of both L2C and L5.

Question #34: Are the new satellites capable of maintaining our legacy signals, or is it totally out of the question?

Gakstatter: It’s not that the government wants to eliminate any signals, but rather they reserve the right to alter the military P(Y) signal on L2. After December 31, 2020, it may behave just like it does today or it may not, so yes, they do have the capability to have the satellites behave as they do today.

Question #35: Comment: I’m like your friend who expected his last purchase to carry him through to retirement. The way my retirement account is growing (negatively) I may still be carrying a range rod when this change occurs.

Gakstatter: I empathize. Maybe I’ll join you.

Question #36: will we still need two receivers to use RTK techniques?

Gakstatter: I also talked about this question during the webinar a bit. Yes, there will still be a need, but I think it will be more ubiquitous than it is today, primarily because of the proliferation of RTK networks and wireless communications technology. Because of this, I think you’ll see the need to operate your own RTK reference station diminishes significantly.

Question #37: How will this affect my processing software, such as Ashtech Office?

Gakstatter: Well, you’ll be okay until December 31, 2020. If there is an upgrade path that supports L2C, that might be a good move if your receiver supports L2C.

Question #38: How much will it cost, upgrading to L5?

Gakstatter: Please check with your local dealer or the manufacturer of your equipment.

Question #39: Does adding L1C into the mix have a great advantage?

Gakstatter: With respect to interoperability with Galileo, yes. With respect to the L1C code itself, it will be superior to its predecessor, L1 C/A, much like L2C/L5 and offer enhanced code and carrier tracking.

Question #40: With L2C only (no L5), is the cross correlation with the L1 C/A or will there be a new code on L1 as well?

Gakstatter: There is no new code on L1 at present. The two civilian codes are L1 C/A (the original) and L2C. On Block III GPS satellites, L1C is planned. Basically, a new and improved L1 C/A.

Question #41: How much is the difference between maintaining the legacy signals and not maintaining the signals?

Gakstatter: Again, I deferred to the GPS World’s military and government editor Don Jewell and retired GPS Chief Engineer Col. Mark Crews (ret.):

I assume you are talking about the difference in costs here, but this is really not a question pertinent to this issue, as we are currently not planning on doing away with any current signals. Both the L1 and the L2 signal structure, coded and codeless, will still be broadcast for the foreseeable future. The issue is that, after December 31, 2020, the flex-power capability may cause temporary problems in codeless and semi-codeless civilian receivers for periods of time while the satellites are in flex-power mode.

However, your question is pertinent in the general sense, as there are new GPS signals and frequencies coming on-board and there are those who believe that some of the old signal structures should be abandoned for the newer more capable signals. So far there have been no decisions made to abandon any of the current signals, only to make them stronger and more robust, with more anti-jam and anti-interference capabilities, which is one of the functions of flex-power that serves the war fighter.

There is also the possibility that flexible power mode will be modified by 2021 in such a way that it will not cause a L2 phase shift and affect civilian receivers that are using semi-codeless techniques.

Question #42: Is there a minimum baseline requirement for differential processing (RTK or otherwise) with dual frequency receivers? Will the change from L1/L2 to L1/L5 alter this?

Gakstatter: The minimum baseline won’t change. There really isn’t one for dual frequency GPS ,although very short baselines sometimes fare better with L1-only rather than L1/L2.

With regard to legacy L1/L2 vs. L1/L2C/L5, I believe you’ll have more robust solutions with the latter, and longer baseline processing will be enhanced.

Question #43: If we need dual frequency receivers for survey quality, wouldn’t three frequencies enhance ambiguity resolution and/or accuracy and precision?

Gakstatter: I believe ambiguity resolution will be enhanced (e.g. quicker and more reliable) due to better ionospheric correction with three frequencies. With regards to accuracy, I don’t see a significant improvement. As I mentioned before, the best way to enhance accuracy/precision (especially vertical) is signals from more satellites (e.g. GLONASS, Galileo, or GPS).

Question #44: After the sunset date will dual frequency not work at all, or just give bad data? How will we know if the signal has changed?

Gakstatter: Your receiver won’t be able to correctly resolve the integer ambiguity because of the phase shift. You won’t know until it happens. It’s not a permanent state either. Legacy receivers may work just fine for periods of time, but then may not for periods of time.

Referring back to the answer to Question #41, it is also possible that the US Government will find a way to resolve this situation where we will not see a phase shift at all.

Question #45: Will this enhance my L1 handheld accuracy at all?

Gakstatter: No. L1 C/A will remain the same.

Question #46: Comment: I know several surveyors who need to hear this conference. It was great. Please publish on the ‘net for access or email to us for distribution. Thank you.

Gakstatter: Thanks for listening and taking the time to comment. Pls feel free to forward this email or the links embedded in this email to those whom you think are interested. You can take in the webinar via the archived version here.

If I didn’t fully answer your questions or if it spawned more, please don’t hesitate to email me more questions and comments.