Category: Applications

So, You’ve Been Hearing About L5

It seems in the world of high-precision GPS, there is always something new to consider. A new satellite launched … a new signal being broadcast … a new whiz-bang product introduced.

In fact, engineers are designing GNSS receivers to be “compatible” with signals that aren’t even being broadcast yet. Even more interesting is that salespeople are selling GNSS receivers that are “capable” of receiving these signals that aren’t being broadcast yet, and satellite navigation systems that don’t even exist yet, such as Galileo.

The third civil GPS signal, L5, has been in the news lately, so hearing about it has probably got your antenna up. First of all, let me clarify that a useful L5 signal is still years away – several years.

The L5 news you’ve been hearing about lately is a test of the L5 frequency from a modified Block IIR-M satellite built by Lockheed-Martin last year. For ~$6 million, Lockheed modified one of the Block IIR-M satellites to broadcast some test data on the L5 frequency (1176Mhz). The modified Block IIR-M satellite is scheduled to launch next month.

However, the L5 test data being broadcast won’t be usable by your GNSS receiver that is capable of receiving GPS L5. In fact, I’m almost certain that even if you purchased an L5-capable GNSS receiver today, a serious software update (if not hardware) will be required even when usable data is eventually broadcast on L5.

For a quick review, let’s summarize which satellite models are/will broadcast the various signals:

• Block I/II/IIA (26 are broadcasting) – L1 C/A, L2 P/Y
• Block IIR-M (six are broadcasting, two remain to be launched) – L1 C/A, L2 P/Y, L2C
• Block IIF (12 are planned, first launch planned in 2009) – L1 C/A, L2 P/Y, L2C, L5
• Block IIIA (eight or possibly 10, first launch planned for 2014) – L1 C/A, L2 P/Y, L2C, L5, L1C

The first Block IIF is scheduled for launch in 2009. At an aggressive launch rate of three per year, all twelve wouldn’t be in orbit until the end of 2012. Even then, we’ll only have twelve satellites broadcasting on L5. Users will derive some benefit from those twelve, but nothing like you would with a full constellation.

Since there are only twelve Block IIF satellites planned, having more than twelve satellites broadcasting L5 means we’ll have to wait for the Block III satellites to be launched. This could go well into the next decade as the U.S. Air Force awarded the contract to Block IIIA satellites to Lockheed just last week, with a first launch optimistically scheduled for 2014.

Besides launching satellites, there are many other L5 technical issues that need to be addressed, such as the control segment infrastructure (software and hardware). There are also significant non-technical issues, money being a big one. Launching satellites is an expensive endeavor; up to $100 million per launch is the number commonly referred to. This is one major reason why GPS satellite launch schedules are a moving target. As much as I’d like to see more satellites in orbit, I understand why the bean counters might want to push out GPS satellite launches for a year when there are 31 operational satellites today, the most ever in history. For that reason, I wouldn’t care to place a bet on where we’ll be with L5 four years from now anymore than I’d bet on where Garmin stock will be four years from now.

What’s so cool about L5 anyway?

Setting launch schedules, infrastructure development, and funding aside, L5 offers some really significant benefits to the high-precision user. First and most significant is that the days of worrying about ionospheric activity will be over. At 1176 Mhz, the L5 frequency separation from L1 (1575Mhz) is significant enough that the ability of your receiver to directly mitigate the ionospheric refraction will render the iono beast essentially harmless.

If you’re relatively new to GPS (in the last six years or so), you will experience the iono beast in the coming years before L5 is ready to help. The current solar cycle has just started and will continue for eleven years. It will peak around 2011 to 2012. The next solar cycle’s affect on GPS is worthy of a newsletter column by itself, and I will have one in the coming months.

Another benefit of L5 is that its broadcast strength is about four times stronger than L2C. A stronger signal combined with a superior code structure means that you’ll get more robust performance in tough GPS conditions. That’s great news for high-precision users working in marginal GPS conditions.

Breaking News on GPS modernization

On Friday of last week, the Office of Space Commercialization published a notice in the U.S Federal Register requesting public comments from within the United States and beyond on the U.S. government’s plan to phase out codeless and semi-codeless access to GPS by December 31, 2020.

This is an historic step. It will be the first time in history that GPS equipment will be rendered obsolete. Specifically, the units affected are high-precision, dual frequency GPS receivers that weren’t designed to use L2C or L5. For example, I have a set of dual-frequency RTK receivers that are about 12 years old – designed well before L2C and L5 specifications were finalized. They work great now, but would be affected by this change.

Granted, we’re talking about 2020, twelve years from now. But there are a great number of you (and me, to a certain extent) that subscribe to the “if it ain’t broke, don’t fix it” philosophy, and even if we did upgrade, we’d hang on to the old equipment as a back-up.

It’s important to recognize that the equipment will not stop working on December 31, 2020. The idea is that the DoD will no longer guarantee the availability of the Y-code on Block IIR-M, Block IIF and Block IIIA satellites that would used by codeless and semi-codeless receivers. Legacy Block II satellites would continue to broadcast Y-code until those satellites are decommissioned. But by 2020, not many legacy Block II satellites will still be operational. Legacy Block II satellites comprise 25 of the current 31 operational satellites today.

After I digest the Office of Space Commercialization’s notice a bit more and talk to a few people, I’ll publish my thoughts in this newsletter in the coming weeks.

What’s up with Galileo

Given some recent events, it’s about time to check in with you regarding Galileo, Europe’s attempt at creating a satellite navigation system similar to GPS. There’s also been some activity on the GLONASS front that is worth writing about too.

In my Directions 2007 article in the December 2006 issue of GPS World magazine, I wrote that Galileo has huge potential benefits for the survey/construction user. But I also wrote that the key new items to watch for would be significant delays. And the delays did come. The problem has been that Galileo is becoming known as the most talked about satellite navigation system that has never been … vaporware, if you will. The program has been belabored with delays and hiccups of both the political and financial sort for many years. One would surmise that after so many miscues that people would stop talking about it. The reason people don’t stop talking about it is that the benefits of Galileo to the satellite navigation user are so significant.

One of the major challenges that Galileo faces is funding. Galileo is to be the first global, civilian satellite navigation system. While GPS and Russia’s GLONASS are both funded by their respective defense departments, Galileo was going to be funded by a partnership between civilian government and commercial entities referred to as the Private Public Partnership (PPP). Exactly how much was private funding and how much was public (government) funding was the question that was a moving target for years.

Until May 2007, the PPP ball was in the hands of a consortium of private companies with the likes of Thales, Alcatel and Inmarsat among other members. Their job was to formulate a plan on how to operate Galileo; essentially, they were trying to figure out how to make it a viable commercial venture. They couldn’t come up with a plan, so in May 2007 Europe abandoned the PPP approach and conceded that the only way Galileo was going to get off the ground was via public funds, to the tune of €3.4 billion. In late 2007, the Galileo program was near death when the European Union (EU), comprising 27 countries, voted in November 2007 to keep Galileo alive, although funding was still an issue.

Just last month, the European Union transport ministers approved the Galileo Implementation Regulation, which allocates that €3.4 billion taken from the EU agriculture and administration budgets. With the money flowing and all 27 countries of the EU in favor, the path to launching Galileo seems as clear as it’s ever been. The next step is releasing contracts for the ground infrastructure and satellites. The European Parliament subsequently cleared the last bureaucratic hurdle later in the month. It approved legislative regulations that lay down security requirements for Galileo and the European Geostationary Satellite Navigation Service (EGNOS), both of which will be managed by the European Commission and the European Global Navigation Satellite System Supervisory Authority.

With the funding seemingly taken care of, however, the implementation schedule is less clear. The EU says the system will be ready by 2013. Folks, that’s only five years from now, not very long at all in the space systems business. They need to build the ground infrastructure, launch 30 satellites and write the software that needs to control and monitor all of the systems.

As far-fetched as the schedule may seem, Europe is dangling the Galileo carrot and I’m still chasing after it.

There is even a carrot dangling in orbit; the second Galileo satellite, GIOVE-B, lifted off late last month. Only 29 more to go — 28, if you count GIOVE-A, but it’s getting a bit long in the tooth.

I have a lot of patience because I’ve bet all along that Galileo was going to materialize. The reason is because of one key application that no other country is going to rely on a U.S. military system for: aviation navigation. It’s abundantly clear that the future of aviation navigation will be based on satellite navigation. How comfortable would the U.S. Federal Aviation Administration be with basing their entire navigation system on a foreign, military-run navigation system? They wouldn’t be and they wouldn’t do it. I think that’s about as comfortable as the EU would be with banking its aviation navigation future on GPS. GPS was a great proof-of-concept tool for them, but I don’t see how it’s a long-term solution for them. Galileo is.

We, as precision users of GNSS, would benefit greatly from Galileo. Frankly, if I had to choose between GPS modernization (L2C/L5/GPS III) and Galileo, I’d take Galileo. Even with 31 healthy GPS satellites broadcasting today, I still find myself losing productivity in the field because of the lack of satellite signals, and I hear similar complaints from others. If you’re trying to do elevation work, it’s even worse. Having ~30 GPS satellites and ~30 Galileo satellites available would mean your average number of satellites in view would be 20 plus. Your PDOP values would be ridiculously low throughout the day. Your redundancy and reliability would be ridiculously high. You wouldn’t have to choose a particular time of day to do your elevation work. Machine control wouldn’t be only usable in wide-open sky environments. These are days to dream about.

One could argue that Russia’s GLONASS system is closer to achieving this than Galileo. With GLONASS having 14 operational satellites, it’s a valid argument. But GPS and GLONASS were designed when the US and Russia were Cold War enemies. You can bet that United States and Russian scientists weren’t cooperating at that time. On the other hand, once the U.S. government decided to support Galileo (they didn’t initially), United States and European scientists worked closely together to ensure that GPS/Galileo receivers would work efficiently.

GLONASS/GPS receivers aren’t efficient by any stretch of the imagination. GPS is based on CDMA technology whereas GLONASS is based on FDMA technology. It’s kind of like designing a video cassette recorder (VCR) that accepts both VHS and Beta tapes in the same slot. You can design it, but there are consequences. One example in GPS/GLONASS receivers is that the receiver must use one GLONASS satellite to sync the two systems together. In real terms, if your GPS/GLONASS receiver is tracking three GLONASS satellites, only two are used. Power consumption is another issue.

These are some of the reasons you’ll never see a GLONASS/GPS receiver in the consumer market from Garmin, TomTom, Magellan, etc.

In response to pressure to be more compatible with GPS and Galileo, Russia announced earlier this year that they are near final approval in adding CDMA architecture to their GLONASS-K generation of satellites that are scheduled for launch in 2010. It’s a smart move on Russia’s part, as any system using anything other than CMDA in the next decade is going to be largely ignored by the manufacturers.

MSAS: SBAS in the Land of the Rising Sun

Quietly, the Japanese Civil Aviation Bureau (JCAB) has been developing that country’s MTSAT Satellite-Based Augmentation System (MSAS) over the years. MSAS is Japan’s satellite-based augmentation system (SBAS) that is designed to enhance GPS for aviation navigation. It’s similar and compatible with the United States’ Wide Area Augmentation System WAAS, as well as Europe’s European Geostationary Navigation Overlay System (EGNOS). It became operational in September 2007.

So, why am I writing about an aviation navigation system in Japan when this is a survey and construction column? Well, for the same reasons I’ve written about WAAS and EGNOS in the past. The use of SBAS by high-precision, non-aviation users is growing by leaps and bounds. I’m not talking about consumer GPS users. Autonomous GPS is so accurate these days that the average consumer doesn’t feel the benefit of SBAS vs. autonomous GPS. I’m talking about the professional users in mapping and surveying who require GPS equipment that will consistently deliver positioning at the meter-level and also at the centimeter-level. MSAS can help achieve both.

MSAS Coverage Area
MSAS was designed to be compatible with the United States WAAS program. Some of the same contractors, like Raytheon, for example, that helped develop WAAS were also involved in the development of MSAS. A GPS receiver designed to use WAAS, or EGNOS for that matter, can also be used to receive MSAS corrections.

Like WAAS and EGNOS, MSAS is a regional SBAS. It serves the region around Japan in support of aviation navigation. Also like WAAS and EGNOS, MSAS is a powerful tool for non-aviation users who require a highly accurate source of GPS corrections in applications like mapping and surveying.

If you are interested in using MSAS, the first question to answer is whether MSAS “covers” the area you are working in. Following is an MSAS coverage map:

It should be noted that this is the minimum coverage area. Some manufacturers have developed schemes to extrapolate the ionospheric grid when working outside of the published coverage area as well. Using this scheme, the grid map can be expanded by approximately 10 degrees in latitude and longitude.

For example, Thailand is slightly outside of the published MSAS coverage map. A typical GPS receiver designed for SBAS won’t be able to use MSAS corrections in Thailand. However, a GPS receiver designed to operate on the fringes of SBAS coverage areas may well be able to operate in areas like Thailand where there is no central SBAS for that country. Granted, this wouldn’t be useful for aviation users, but for ground users looking for meter-level accuracy, it may work just fine.

Essentially, the official, published coverage area is determined when a GPS receiver is in a position where it can see GPS satellites that are in view of at least two MSAS ground reference stations. This is the same for all SBAS. Following is a map of the MSAS ground infrastructure as presented by a representative from the Japan Aerospace Exploration Agency at the Institute of Navigation conference last September in Forth Worth, Texas. The full presentation can be viewed here.

Another factor in determining ability to use MSAS corrections is the visibility of the geostationary broadcasting satellites (GEOs). Like all SBAS, MSAS corrections are broadcast via geostationary satellites. To use MSAS, a GPS receiver must be able to receive corrections from one of those satellites. There are two satellites broadcasting MSAS corrections located at 140.0 degrees east longitude (PRN 129) and 145.0 degrees east longitude (PRN 137). Following is a map of the “footprint” that is covered by the two MSAS broadcasting satellites. Please not that just because a receiver located within the broadcasting satellite footprint does not mean it will be able to use MSAS corrections; it must also be within or on the fringe of the MSAS grid map discussed above.

Finally, not only does a receiver need to be within the MSAS broadcasting footprint of one of the two broadcasting satellites (GEOs), the signal from the GEOs is line-of-sight. This means that buildings, terrain, trees/vegetation, etc. can block the signal. Some manufacturers have developed various schemes to overcome this, but others have not, so just because a particular receiver is not able to use MSAS corrections, doesn’t mean that MSAS is unusable in that area.

MSAS Accuracy
While the JCAB doesn’t make available test bed results like the FAA does for WAAS, I’ve seen results of data collected using high-performance GPS receivers and MSAS as the correction source. The results are on the same level as WAAS, being well under a meter horizontally. Of course, there are a million caveats when discussing GPS accuracy, so I’ll leave it at that for now, but it does demonstrate the potential accuracy that non-aviation users could realize when using MSAS as a source of GPS corrections.

MSAS for Survey-grade GPS Receivers?
Up until this point in the article, the discussion has been for those interested in sub-meter or meter-level positioning for mapping.

Development efforts by some companies in the last couple of years have resulted in survey-grade GPS receivers with centimeter-level accuracy, benefiting from SBAS signals. One example is the Magellan PM3 RTK. It was introduced last year as one of the first L1-only real-time kinematic (RTK) systems. A key innovation in the PM3 RTK technology is its ability to use SBAS GEOs as a source of ranging measurements. Essentially, the two MSAS GEOs are treated as another constellation of satellites to augment GPS satellite measurements.

According to Magellan, the addition of SBAS makes a significant difference in resolving ambiguities for centimeter-level positioning. Following is a chart from a Magellan white paper.

SBAS for Non-aviation Users Continues to Strengthen
The utility of SBAS in the mapping/surveying/construction industries continues to grow. One indicator of a successful technology is that it spurs the development of other technologies around it. MSAS and SBAS in general have certainly done that.

Japan’s MSAS is the latest SBAS to become operational. Look for future reports on other SBAS, as India develops and introduces its system called GPS and Geo Augmented Navigation (GAGAN), which is referenced on the first graphic above.

RTK Crops Up in Precision Agriculture

Take a look at the major manufacturers of multi-frequency GNSS equipment and the markets they serve. Obviously, surveying is an important market; construction is a major one too. Maybe you’ve noticed that agriculture is also making its way up the food chain, so to speak. Trimble, Topcon, Leica/Novatel, Deere/Navcomm, and Hemisphere GPS are all pursuing the precision agricultural market with RTK-like precision.

The terms “precision agriculture” and “precision farming” have been around for a long time. There are conferences dedicated solely to sharing information on this topic, such as the 9th International Conference on Precision Agriculture. The classic precision ag market has been relatively unchanged for the past decade or so. Yes, in the past few years WAAS (Wide Area Augmentation System) has made a substantial impact with respect to a reliable and convenient source of corrections for one-meter accuracy using GPS, but all in all, one-meter accuracy is what most precision ag users have adapted to.

Stepping back, GPS/GNSS is only one component of the precision agriculture puzzle. Remote sensing (aerial photography/LIDAR), various sensors, and GIS (geographic information systems) software all play a key role in enabling people to record, analyze, and apply data used to make decisions that optimize the output of a particular agricultural area.

There are a few keys areas where GPS/GNSS is used in agriculture, though. These aren’t all-inclusive, but cover significant tasks where GPS has been very beneficial.

1. Field mapping: there have always been field maps of some sort, whether they were hand-sketched or derived from an aerial photo, but GPS has improved the precision substantially. Not only can one map the perimeter of the field, but also infrastructure such as roads, drain tile, outlets, wells, buildings, etc. The farmer doesn’t necessarily do this, but fertilizer distributors have become GPS-savvy and offer this service.
2. Yield mapping: as crops are harvested, a GPS receiver connected to a yield monitor sensor records a coordinate along with the yield data. This data is combined and analyzed to create a map of how well different areas of the field are producing.
3. Guidance: when spreading fertilizer or planting, equipment operators have traditionally used markers such as foam or other visual aids to mark where they’ve been to try and avoid overlap. The assistance of GPS and onboard guidance systems, such as a light bar, can further reduce overlap.

As I mentioned before, for many years precision ag users have settled for one-meter precision. Companies like Hemisphere GPS (formerly CSI Wireless) did very well designing single-frequency GPS receivers for the precision ag market. Hemisphere is also a leading designer of radio beacon (Coast Guard) receivers. Radio beacons, in addition to WAAS, are a free source of corrections for one-meter accuracy. Trimble was also an early supplier of precision ag GPS receivers and related equipment, typically offering single-frequency products like the AG-132.

While the real-time kinematic (RTK) technique has been around since the early 90s, it didn’t gain wide acceptance in the precision ag industry. The accuracy was great, down to approximately 2 cm at the time, but the equipment was clunky. The user had to set up a reference station near the field he was working on. The communication link was complicated, and some types needed Federal Communications Commission (FCC) licensing. Consequently, there were several potential points of failure. Lastly, the cost for a complete RTK system (base, rover, and radios) was high, upwards of $50,000. It just wasn’t cost-effective.

What is an RTK Network?

It’s an ambiguous term, because it means different things depending on the industry, but essentially the hardware setup is the same no matter which industry we are discussing. An RTK network is a series of dual-frequency reference stations spaced optimally within a region so as to provide RTK corrections to subscribers within that region. Following is an illustration of an RTK network for agriculture in the Ohio area.

All the reference stations are surveyed (more on that below) with respect to one another.

The network subscriber is assigned a primary reference station to use. RTK networks for agriculture are single-baseline solutions; the subscriber can only use one reference station at a time. There is no “network solution” per se, or redundancy like there is in RTK networks used in the surveying and construction industries. Therefore, when a single reference station “goes down,” the subscribers in that area are down also. There is software available from RTK network vendors that allows the administrator to monitor all reference stations via the Internet, but some networks don’t have that feature implemented. In that case the administrator hears about problems when subscribers have problems.

Another major difference between RTK networks for agriculture and RTK networks for surveying and construction is the communication method. The latter primarily use data plans on mobile phones to receive corrections. Either the mobile phone is linked via Bluetooth to the receiver or a cellular modem is built inside the receiver.

RTK networks for agriculture, on the other hand, primarily use spread spectrum radios (900 Mhz band) to transmit corrections to the receiver. The benefit of a spread spectrum radio is that they are free to use and don’t require a license from the FCC to operate. They are limited in their broadcast range, however, which is typically two to three miles. To solve this problem, radio repeaters are used to extend the distance. Depending on the topography, as many as four repeaters may need to be permanently mounted to effectively cover the broadcast range needed, based on the spacing of the reference stations. Even then, there may still be some areas that are not fully covered. In that case, temporary mobile repeaters may be used to provide coverage to that area.

The Wild, Wild West of RTK Networks

Bill Henning, real-time specialist with the National Geodetic Survey (NGS), said it best: the recent explosion of RTK networks is like the Wild, Wild West. They are proliferating so quickly that it’s hard to keep track of them. One of his tasks is to help develop guidelines for RTK network operators, and I think NGS is making inroads into the survey/construction industry with their initiative. People are looking for that sort of guidance with respect to RTK network setup, as well as monitoring for the networks once they become operational.

RTK networks for agriculture seem less structured than in other disciplines, though, and administrators rely more heavily on vendor recommendations. For example, some are based on the ITRF reference frame, while others are based on some version of NAD83. Some networks hire land surveyors to establish their reference station locations, while others do it themselves using NGS’s OPUS program or other methods. Very few, I think, realize the resources available from the NGS, such as the Cooperative CORS program.

Separate Industries, Separate RTK Networks

Even this early in the RTK network game, the duplicity of networks between agriculture and survey/construction is interesting. For example, an RTK network for agriculture can cover the same area as an RTK network for survey/construction. In the state of Georgia, there are several RTK networks for agriculture and several for survey/construction, some of which overlap. In fact, one would think that these folks (ag and survey/construction) would consolidate their efforts. The RTK network equipment is virtually the same. But alas, the manufacturers don’t want this. Why not? To answer that question, just employ the old adage: follow the money.

The fact is that a farmer isn’t going to pay the same RTK network subscription rate that a surveyor or construction company will. The numbers are vastly different. The typical subscription rate for access to an agricultural RTK Network is $1,300 to $1,500 per unit per year. Subscription rates for access to a survey/construction RTK network are as high as $4,500 per unit per year.

Some industry folks say that aggressive subscription pricing is the reason RTK networks in the agriculture market have expanded rapidly in the past few years. A farm is very hesitant to pay $4,500 annually when they can select a service like OmniSTAR and pay $1,500 annually.

Again, there are differences between the networks used in agriculture and those in survey/construction; most, if not all, are software-related. RTK networks for survey/construction offer a true networked solution, where several reference stations are used to compute a correction, whereas RTK networks for ag are single-baseline solutions, like users would normally set up as a base rover for their own use.

Others at the Party

Of course, OmniSTAR (HP/XP),Deere (Starfire), and Novariant (AutoFarm) offer a GPS-based solution in the precision ag industry. They are not pure-play RTK solutions like RTK networks are, although they do have RTK capability. True RTK networks are capable of constantly delivering ~2 cm accuracy day-in and day-out. These folks going after the precision ag market offer decimeter-level services primarily (1 decimeter being the equivalent of 10 cm), and then RTK solutions when needed.

It will be interesting to see how pure-play RTK players respond as RTK networks for agriculture continue to expand … which they most certainly will.