GLONASS/GPS for everyone: tests for accuracy and accessibility of positioning of a single-chip receiver in difficult operating conditions

Philip Mattos (Philip Mattos)
Translation: Andrey Rusak
support@site
Victoria Bulanova
[email protected]
The single-chip GNSS receiver, which has now entered mass production, was tested in dense urban environments to demonstrate the benefits of multi-system (GLONASS and GPS) operation as a consumer receiver. The use of the combined GLONASS/GPS system began with several tens of thousands of receivers for geodetic surveys; millions of such consumer devices are currently operating. Thanks to the growth in the number of personal satellite navigation devices, the emergence of automotive OEM systems and mobile phones, it was possible to achieve significant market volumes in 2011. Confidence in the prospects for the development of the market for navigation user devices is pushing manufacturers of high-frequency specific components, such as antennas and SAW filters, to increase production volumes and optimize the cost of goods. One of the first Russian companies to market modules based on the STM receiver was NAVIA. NAVIA GLONASS modules have already proven themselves as reliable, convenient modules for the production of ready-made navigation terminals and control of moving objects. Various module tests have shown that ML8088s and GL 8088s meet all the manufacturer’s stated characteristics and can be successfully used in monitoring devices.

Tests of a single-chip GLONASS/GPS receiver in London, Tokyo and Texas were carried out to show that the joint use of all visible GLONASS satellites coupled with GPS provides better positioning availability in dense urban areas, and in the case of poor positioning availability - better positioning. accuracy.

It is obvious that multi-system receivers are in great demand in the consumer market. They can ensure operation over a larger number of satellites in conditions of “urban canyons”, where only part of the celestial hemisphere is visible in the visibility zone and high reliability in filtering out unnecessary signals is required, when the quality of useful signals is greatly degraded due to multiple reflections and attenuations. The following briefly describes the difficulties of integrating the GLONASS system (and subsequently GALILEO), on the basis of which cost-effective devices are produced for the mass consumer. For such a market, on the one hand, cost comes first, and on the other hand, there are high performance requirements associated with low signal levels, limited power consumption, short cold start times and positioning stability.

The goal was to use all available satellites to improve the performance of consumer navigation devices in indoor and urban environments. 2011 passed under the auspices of GLONASS support; the development of this satellite system is approximately three years ahead of GALILEO. When designing receivers, it was important to overcome the problems of incompatibility of hardware support for GLONASS and GPS. That is, the frequency-modulated GLONASS signal required a wider frequency band than the pulse-code modulation signals used by GPS, bandpass filters with different frequency centers and different transmission rates of signal elements. And all this without significantly increasing the cost of the receiver.

Under ideal operating conditions, satellites from additional constellations will be ineffective, since positioning availability I get close to 100 percent using only GPS. The presence in the ionosphere of seven, eight or nine satellites used for positioning in fixation mode minimizes the total error and gives correct coordinates.

In extreme operating conditions, the use of only GPS allows one to determine the position, but the use of only three, four, five satellites concentrated in a narrow part of the celestial hemisphere leads to poor DOP values. Increasing the number of satellites significantly improves accuracy, thereby improving DOP and averaging multipath errors. Limiting the number of positioned satellites leads to the imposition of multipath errors on the determination of the coordinates of the amplified DOPs. Adding a second or third satellite constellation involves expanding the number of visible satellites, and thus more satellites are involved in the coordinate determination process, which leads to a reduction in errors.

Therefore, in extreme conditions, where the use of GPS alone is not enough, the additional use of GLONASS satellites (and subsequently GALILEO) increases the availability of positioning to 100% (with the exception of underground tunnels).

In fact, availability is a self-improving positive feedback loop: since satellites are constantly being tracked, even if they are rejected from participating in the current solution to the positioning problem using the RAIM / fault and FDE algorithms, there is no need to search for them again - they have already become available for use previously. If the positioning process is not interrupted, then it is possible to continue to accurately predict phases for satellites with closed obstacles, which allows them to be used instantly when leaving the shadows, since it does not require receiving additional information to search and fix them.

Additional visible satellites are very important for the consumer, in particular - as an example, with “self-assistance”, when the minimum group is represented by five satellites, rather than three or four, in order to autonomously establish that all satellites are “correct” ,using receiver autonomous integrity monitoring (RAIM) techniques. “Self-service” has even more significant advantages for GLONASS: there is no need for any infrastructure such as assisted servers, which always lead to a delay in service. The GLONASS method of transmitting satellite orbit parameters in the Keplerian format is also very suitable for the “self-service” algorithm.

Test value

Previous attempts to characterize the benefits of multi-system devices in urban environments have been stalled by the need to use professional receivers not designed for such signal levels, and would have to obtain separate results for each group or sacrifice one of the satellite measurements to measure time. These circumstances did not allow us to continue testing the devices that were planned for release on the mass market.

The release of a new multi-system solution is of great importance, since the receiver under test is a truly mass-produced device if it has increased sensitivity and is completely ready for both measurement and calculation. Thus, the author of this article reports for the first time absolutely reliable test results.

Background

Tests were carried out on a single-chip GNSS receiver Teseo-II (STA-8088). Brief history: This is a 2009 product manufactured by STM, based on Cartesio+ with GPS/GALILEO and Digital Signal Processor (DSP) already included, it was ready to be implanted with GLONASS functionality, which led to the creation of the Teseo-II chip (2010 product ). Test results with real satellite signals were obtained on a Baseband chip in FPGA implementation at the end of 2009, and in 2010 using a ready-made chip.

The current design required additional minor circuit modifications. The required DSP hardware and software changes were minor and are included in the next scheduled TeseoII circuit update. The implementation of the RF part circuit required much more attention than the two-channel circuit with an intermediate frequency (IF) stage and an analog-to-digital converter (ADC), with additional frequency conversion and a wider bandwidth IF filter. But, since the area of ​​​​the crystal with the RF part located on it is very small in the total volume, even a 30% increase in the circuit is insignificant for the entire circuit. According to the fact that the chip design is for a common single-chip system (RF and BB, from antenna to positioning, velocity and timing (PVT)), so the total die area for the 65nm process is very small.

From a commercial point of view, the inclusion of all three satellite constellations (GPS/GLONASS andGALILEO) into one chip is new for the consumer. Many of the companies present on the Russian market have settled on a two-system approach, just to satisfy the requirements of the Russian government about the need to work in the GLONASS system. They did not think about the global future, when there will be several positioning groupings in the world and perhaps each of the countries participating in this process will further put forward demands for the predominant use of their own system.

In this regard, the solutionTeseo— II is revolutionary because prepared in advance for such a scenario and can already receive GLONASS systems/ GPS/ GALILEO/ QZSSAndSBAS.

Technically, the inclusion of independent channels for receiving and processing the GLONASS system in a group is also new, while the GPS/GALILEO combination is already standard practice. Achieving such flexibility also required new technical solutions that take into account differing RF hardware delays and differences in signal transmission speeds. In addition to this, there are the now well-known Coordinated Universal Time (UTC) correction and the geoid correction problem.

A direct transition to a single-chip solution (RF + Baseband + CPU) is rare: this is an important technological breakthrough. Confidence in this step is due to the experience of using the RF part and the proven Baseband circuit of the processor. The external RF interface STA5630 and a modified GPS/GALILEO DSP, which were previously used in Cartesio+, were taken as a basis.

The reliability of the STA5630/Cartesio+ was proven in mass production in the form of separate circuits even before the release of 3-in-1 SoC solutions.

Unlike dual-chip solutionsGPS/GLONASS modules present on the Russian market, single-chip solution fromSTMicroelectronics (Teseo— II) S.T.A.8088 FG has much greater reliability, noise immunity, lower power consumption and, of course, smaller dimensions (module M.L.8088 shas dimensions 13 x 15 mm).

Support for GLONASS and GALILEO is a step forward compared to the previous generation of RF hardware. GALILEO is compatible with GPS and therefore the existing scheme could be used, but GLONASS required additional changes. See Figures 1 and 2.

Picture 1.

Figure 2.ChangesBaseband parts to support GLONASS

In the RF part, the LNA, RF amplifier and first mixer were combined into one channel. This allowed us to save on the number of chip pins and minimize power consumption. Moreover, this allowed to maintain external costs for equipment manufacturers. The GLONASS signal, reduced in the first mixer to 30 MHz, enters the secondary processing channel (shown in brown) and, mixed to 8 MHz, is fed to a separate ADC and then to the Baseband part.

The Baseband part provides an additional preliminary processing stage (indicated in brown), which converts the signal to 8 MHz, which is necessary for feeding into the Baseband and passes the resulting signal through an anti-interference filter, and also reduces the sampling frequency to the standard value of 16, suitable for processing in DSP hardware.

Existing acquisition devices and tracking channels can choose where and when to receive GPS/GALILEO or GLONASS signals, which makes the distribution of channels in relation to satellite constellations very flexible.

Less visible, but very important to system performance, is the software that controls these hardware resources, firstly to close the PLL tracking loops and take measurements, and secondly, the Kalman filter, which converts what is measured into PVT data. necessary for the user.

All this has undergone a structural modification to provide support for working with many satellite constellations, and not just GLONASS. In this case, expanding the software to receive future global navigation systems will become a stage of evolutionary development, and will not require major modifications to the crystal itself.

The software had been running on a real chip since 2010, but using signals from any simulator or static roof-mounted antennas, only GPS data was available, which was so good that it did not allow any maneuvers for research to improve the system. In early 2011, pre-production chip samples and development boards with antennas in the package became available, making mobile field testing possible worldwide.

Actual results

Before the birth of the crystal with multi-system reception, the results were already visible from preliminary tests carried out using professional receivers with separate GPS and GLONASS measurements. However, these tests did not provide good data for a consumer receiver because they showed low sensitivity. The receivers required a sufficiently clean signal to drive the PLL, but this could not be done in an urban environment, and most importantly, the receivers created two separate solutions with a constant additional satellite to deal with inter-system timing differences. Uncoupled solutions did not make it possible to predict the position of satellites of one constellation by calculating their position based on coordinates calculated using another, which is one of the main advantages of multi-system GNSS receivers.

The simulation of visible satellites was carried out in 2010 in dense urban conditions in Italy, the center of Milan. The results, averaged every minute for a full 24 hours, are presented in Table 1. The average number of visible satellites increased from 4.4 with GPS only, to 7.8 for GPS+GLONASS, with the number of No Fix points equal to zero . Moreover, in the “GPS only” mode, 380 false points were received, which amounted to about 26% of the total reception time.

Table 1.Accuracy and AvailabilityGPSAndGPS+GLONASS, on average over 24 hours

However, the availability of satellites was not an end in itself. Having more satellites in the same small area of ​​the celestial hemisphere over urban areas may not be sufficient due to the geometric reduction in accuracy. To examine this data, the geometric precision represented by HDOP. When using GLONASS and GPS together, the result was 2.5 times better.

Previous studies have shown that in individual test cities, two to three additional satellites were available, but one of them was used for timing. When using a highly sensitive receiver combined on one chip, we assumed that four or five additional satellites would be involved.

The actual results far exceeded our expectations. First, signals from many other satellites appeared, since all previous tests and simulations excluded reflected signals. Having additional signals, the receiver significantly improved DOP performance. The effect of reflections on accuracy was significantly reduced, firstly due to better positioning geometry, and secondly due to the ability of the FDE/RAIM algorithms to maintain satellite tracking stability. In addition, the number of false signals that can distort coordinate data has decreased.

The results presented here are obtained from a fully integrated high-sensitivity receiver such as the NAVIA ML8088s receiver, based on the STA8088s chip. It is optimized to detect even very low-level signals and obtain results directly from all satellites in view, regardless of constellation. This ensures 100% satellite availability and greatly improves accuracy in difficult urban environments.

Availability

The use of highly sensitive receivers that are independent of phase-locking loops (PLLs) ensures full accessibility in modern cities, even when reflected from glass surfaces in modern buildings. Therefore, some other definitions of availability than “four satellites are available” are now required. For example, tracking satellites at a given level of signal quality, the result of which depends on DOP. Even the DOP can be difficult to estimate because the Kalman filter assigns different weights to each satellite, which are not taken into account when calculating the DOP. And also, in addition to instant measurements, this filter uses the historical position and current speed, which leaves the positioning accuracy unchanged.

Figure 3 shows satellite availability in tracking mode. Testing took place in London's financial district in May 2011.

Tracked Satellites –GPS, GLONASS,GPS+GLONASS

Figure 3.GPS(marked in blue) against GLONASS (marked in red) and all tracked satellitesGNSS(marked in green).

As can be seen in Fig. 3, in total there are 7-8 GLONASS satellites and 8-9 GPS satellites, that is, multi-GNSS - about 16 satellites. There was a period when satellite signals were not picked up: during the passage of the Blackfriars Underpass tunnel, time stamp approximately 156400 seconds. In other areas of the city, at approximately 158,500 and 161,300 seconds, visibility dropped to four satellites, but their total number was never less than eight. It should be noted that the testing took place in the old city, where there are mainly stone buildings, so the reflective signals are weaker than from glass and metal buildings.

Although satellite availability is 100% outside tunnels, it may be limited by DOP or positioning accuracy. As can be seen in Figure 4, from other tests in London, multi-GNSS DOP remains below 1, as it should be with 10-16 visible satellites, while GPS-only DOP is often above 4, with no distortion Due to reflections and weak signals, the DOP is significantly increased to 10 at peak.

GPScompared toGNSS

Figure 4.OnlyGPSagainst combinedGPS/GLONASS accuracy reduction indicators

Since the tests conducted in May 2011 were light enough to create stressful conditions under which GPS would need multi-GNSS support, new testing was conducted in August 2011. As shown in the aerial photograph (Fig. 5), the tests were carried out in the modern high-rise part of the city, Canary Wharf. Additionally, the roads in the city are very narrow, which made the city's challenges even more difficult. Glass and metal buildings in the modern part of the city tend to give better reflection than stone buildings, causing the RAIM and FDE algorithms to go off the charts.

Figure 5. GPS vs GNSS, London, Canary Wharf

Obtaining GPS only results was difficult (shown in green), especially in the closed part of Docklands station, center left, bottom track.

Figure 6 shows the same real test results displayed on a schematic road map.

Figure 6. GPS vs GNSS, London, Canary Wharf, sketch map

Multi-GNSS testing (blue) showed very good results, especially on the northern (eastbound) part of the loop (driving in the UK is on the left, so clockwise creates a one-way loop).

Figure 7. a) Tests in Tokyo: Teseo-I (GPS) versus Teseo-II (GNSS); b) DOP when tested in Tokyo

Further testing was carried out at STMicroelectronics offices around the world. Figure 7a shows tests in Tokyo, where yellow indicates the test results of the previous generation of chips without GLONASS, and red indicates Teseo-II with GPS+GLONASS.

Figure 7b provides some clarification of the accuracy definition by showing the DOP during the test. It can be seen that the Teseo-II DOPs were rarely higher than 2, but the GPS-only (Teseo-I) DOPs were between 6 and 12 in the circled northern compound.

We repeat that the test algorithm is simple for GPS, but the accuracy of the determination is difficult.

Further testing in Tokyo was carried out on narrower city streets under the same testing conditions, shown in Figure 9. Blue - GPS only, red - GPS+GLONASS, significant improvement in results is observed.

Figure 9 uses the same color scheme to display the Dallas test results, this time with a competitor's GPS receiver versus Teseo-II in a GPS+GLONASS configuration, again seeing very good results.

Figure 8. OnlyGPS(blue) vs multi-GNSS(red), Tokyo.

Figure 9. OnlyGPS(blue, competitor manufacturer's receiver) compared toGNSS(red), Dallas.

Other satellite constellations

Although the hardwareTeseo— IIsupports andGALILEO, no satellites available yetGALILEO(as of September 2011), so devices based on this chip in use around the world still do not have the software loaded to service this satellite constellation. However, if the time comes to use GALILEO, there is always the opportunity to update the software.

The Japanese QZSS system has one satellite available, transmitting traditional GPS-compatible signals, SBAS signals and L1C BOC signals. Teseo-II, with the help of the functions of the currently loaded software, can handle the first two of them, and while the use of SBAS is useless in urban environments, since signal reflections and interference are local and undetectable, the purpose of the QZSS system is to provide a satellite with a very wide angle so that this satellite was always available in urban areas.

Figure 10 shows the test in Taipei (Taiwan) using GPS (yellow) versus multi-GNSS (GPS plus one QZSS satellite (red)), and ground truth (purple).

Figure 10. OnlyGPS(yellow) versus multi-GNSS (GPS+ QZSS (1 satellite, red)), true value -lilac, Taipei
Further work

Testing will continue to obtain more accurate quantitative results. Testing will take place in the UK, where there are road maps with vector data to display real travel directions. It is planned to modify the hardware to support the Compass system and GPS-III (L1-C), in addition to the existing GALILEO. Finding and tracking these signals has already been demonstrated using pre-recorded broadcast script samples on GNSS signal simulators.

Compass was not available in 2011. In this regard, work on the silicon implementation of Teseo-II was focused mainly on maximum flexibility in the conditions of different code lengths, for example, BOC or BPSK, which made it possible, with one or another loaded software for configuring the DSP hardware functions, gain compatibility between different satellite constellations.

Compatibility work on the current version of the multi-GNSS CHIP has been weak: Because the Compass system's 1561 MHz center frequency can only be maintained using a voltage-controlled oscillator and PLL, the Compass system cannot operate simultaneously with other satellite constellations. In addition, the code transmission rate in the Compass system is 2 million bps, which is also not supported by Teseo-II and can be brought to standard through the use of external alternative circuits, which means serious signal losses.

So Compass support work is only relevant for research and software development, for a single system solution, or using a separate RF chip.

The worldwide Compass signal, which is in GPS/GALILEO signal format at carrier frequency and at code length and rate, will be fully compatible within a single multi-GNSS circuit, but most likely not before 2020.

Tests in urban conditions will be repeated as the group developsGALILEO. If there are 32 channels, you can use 11/11/10 division (GPS/ GALILEO/GLONASS), in the presence of a full complement of all three groups, but within the framework of modern requirements for navigation services, the combination 14/8/10 is more than sufficient.

Conclusion

A multi-system receiver can include GPS, GLONASS and GALILEO at minimally increased cost. With 32 tracking channels and up to 22 visible satellites, even in the harshest urban environments, 100% availability and acceptable positioning accuracy can be ensured. During testing, 10–16 satellites are typically visible. Multiple measurements make RAIM and FDE algorithms much more effective at eliminating poorly reflected signals, while also minimizing the geometric effects of remaining signal distortion.

Recently, with the development of the Russian GLONASS, the needs of the navigation market for multi-system receivers are only growing. A number of domestic companies use single-chip chips STM to develop your own GLONASS modules and ready-made packaged devices. In particular, in 2011, the NAVIA company released 2 combined GLONASS/ GPS/ Galileomodules, tests of which showed very good results.

Instant or integral availability(English) Availability – represents the % of time during which the PDOP condition is satisfied<=6 при углах места КА >= 5 degrees. A simple example: in the old days, before 2010, GLONASS availability in some areas of the globe was no higher than 70-80%, but now it is 100% everywhere!)

Reduced accuracy or Geometric Accuracy Reduction(English) Dilution of precision, DOP, English Geometric Dilution of Precision (GDOP)

RAIM(English) Receiver Autonomous Integrity Monitoring Autonomous Receiver Integrity Monitoring (ARIC), a technology designed to evaluate and maintain the integrity of the GPS system and GPS receiver. This is especially important in cases where the correct operation of GPS systems is necessary to ensure an adequate level of safety, for example in aviation or maritime navigation.

Modern satellite navigation technologies provide location determination with an accuracy of about 10-15 meters. In most cases, this is enough, however, in some cases more is required: say, an autonomous drone moving quite quickly over the earth’s surface will feel uncomfortable in a cloud of coordinates with meter errors.

To clarify satellite data, differential systems and RTK (real-time kinematics) technologies are used, but until recently, such devices were expensive and cumbersome. The latest advances in digital technology in the form of the Intel Edison microcomputer have helped solve this problem. So, meet: Reach - the first compact high-precision GPS receiver, very affordable, and, moreover, developed in Russia.

First, let's talk a little about the differential technologies that allow Reach to achieve such high results. They are well known and quite widely implemented. Differential navigation systems (DNSS) improve the location and speed accuracy of mobile users by providing measurement data or correction information from one or more base stations.

The coordinates of each base station are known with high accuracy, so the station's measurements serve to calibrate data from nearby receivers. The receiver can calculate the theoretical distance and propagation time of the signal between itself and each satellite. When these theoretical values ​​are compared with observational data, the differences represent errors in the received signals. Corrective information (RTCM data) is obtained from these differences.

Accuracy of coordinate determination using Reach. Pay attention to the scale.

Corrective information can be obtained by the Reach device from two sources. Firstly, from a public network of base stations via the Internet using the NTRIP protocol (Networked Transport of RTCM via Internet Protocol), which implements the idea described above in relation to a global computer network. Secondly, with the help of the second Reach, which occupies a stationary position near the first and is thus a base station in terms of DNSS. The second option is preferable (DNSS accuracy drops significantly with increasing distance between the receiver and the BS) - it is no coincidence that as part of the crowdfunding campaign on the Indiegogo website, the creators of Reach offer the first position to buy a set of two devices.

The device specifications are shown in the table below. As you can see, the hardware consists of 3 parts: an Intel Edison computer running Linux OS and RTK software RTKLIB; U-blox NEO-M8T GPS receiver and Tallysman TW4721 antenna. Please note that the receiver supports all existing satellite systems: GPS, GLONASS, Beidou and QZSS. This entire set of software and hardware components provides impressive coordinate determination accuracy: up to 2 cm!
Who can use such a device? As mentioned above, the creators of various mobile robotics, autonomous and not so; Moreover, given its low cost (pre-order $545 for a double set and $285 for a single set), it will appeal not only to professionals, but also to enthusiasts. Further, compilers of various kinds of maps, again, including amateurs. Well, just nerds who want to know their location down to the centimeter.

The creators of Reach, the company Emlid, performed successfully on the indiegogo website: in less than a month, almost double the requested amount was collected. This means that the project will certainly be implemented. You still have time to pre-order and be among the first to receive a completely new navigation device. Distribution of goods is scheduled for July.

The user of a GPS navigator is always interested in the real accuracy of GPS navigation and the degree of confidence in its readings. How close can you get to any navigational hazard relying solely on your GPS receiver? Unfortunately, there is no clear answer to this question. This is due to the statistical nature of GPS navigation errors. Let's take a closer look at them.

The speed of propagation of radio waves is influenced by the ionosphere and troposphere, ionospheric and tropospheric refraction. This is the main source of errors after turning off SA. The speed of radio waves in vacuum is constant, but changes when the signal enters the atmosphere. The time delay is different for signals from different satellites. Radio wave propagation delays depend on the state of the atmosphere and the satellite's altitude above the horizon. The lower , the longer the path its signal travels through the atmosphere and the greater the distortion. Most receivers exclude signals from satellites with an elevation of less than 7.5 degrees above the horizon.

In addition, atmospheric interference depends on the time of day. After sunset, the density of the ionosphere and its influence on radio signals decreases, a phenomenon well known to shortwave radio operators. Military and civilian GPS receivers can autonomously determine atmospheric signal delay by comparing delays at different frequencies. Single-frequency consumer receivers make an approximate correction based on the forecast transmitted as part of the navigation message. The quality of this information has recently increased, which has further increased the accuracy of GPS navigation.

SA mode.

To maintain the advantage of high accuracy for military GPS navigators, the SA (Selective Availability) access restriction mode was introduced in March 1990, artificially reducing the accuracy of a civilian GPS navigator. When the SA mode is enabled, an error of several tens of meters is added in peacetime. In special cases, errors of hundreds of meters may be introduced. The US government is responsible for the performance of the GPS system to millions of users, and it can be assumed that the re-enablement of SA, much less such a significant reduction in accuracy, will not be introduced without sufficiently serious reasons.

Precision coarsening is achieved by chaotic shifting the transmission time of the pseudo-random code. Errors arising from SA are random and equally probable in each direction. SA also affects the GPS heading and speed accuracy. For this reason, a stationary receiver will often show slightly varying speed and heading. So the degree of impact of SA can be assessed by periodic changes in course and speed according to GPS.

Errors in ephemeris data during GPS navigation.

First of all, these are errors associated with the deviation of the satellite from the calculated orbit, clock inaccuracies, and signal delays in electronic circuits. These data are corrected from the Earth periodically, and errors accumulate in the intervals between communication sessions. Due to its small size, this group of errors is not significant for civilian users.

Extremely rare, larger errors may occur due to sudden information failures in the satellite's memory devices. If such a failure is not detected by self-diagnosis, then until the ground service detects the error and transmits a command about the failure, the satellite may transmit incorrect information for some time. There is a so-called violation of continuity or, as the term integrity is often translated, the integrity of navigation.

The influence of the reflected signal on the accuracy of GPS navigation.

In addition to the direct signal from the satellite, the GPS receiver can also receive signals reflected from rocks, buildings, passing ships - the so-called multipath propagation. If the direct signal is blocked from the receiver by the ship's superstructure or rigging, the reflected signal may be stronger. This signal travels a longer path, and the receiver “thinks” it is further from the satellite than it actually is. As a rule, these errors are much less than 100 meters, since only nearby objects can produce a strong enough echo.

Satellite geometry for GPS navigation.

Depends on the location of the receiver relative to the satellites by which the position is determined. If the receiver has picked up four satellites and they are all in the north, the satellite geometry is bad. The result is an error of up to 50-100 meters or even the inability to determine coordinates.

All four dimensions are from the same direction, and the area where the position lines intersect is too large. But if 4 satellites are located evenly on the sides of the horizon, then the accuracy will increase significantly. Satellite geometry is measured by the geometric factor PDOP (Position Dilution Of Precision). The ideal satellite location corresponds to PDOP = 1. Large values ​​indicate poor satellite geometry.

PDOP values ​​less than 6.0 are considered suitable for navigation. In 2D navigation, HDOP (Horizontal Dilution Of Precision) is used, less than 4.0. A vertical geometric factor VDOP less than 4.5 and a temporal TDOP less than 2.0 are also used. PDOP serves as a multiplier to account for errors from other sources. Each pseudo-range measured by the receiver has its own error, depending on atmospheric interference, errors in ephemeris, SA mode, reflected signal, and so on.

So, if the expected values ​​of the total signal delays for these reasons, URE - User Range Error or UERE - User Equivalent Range Error, in Russian EDP - equivalent rangefinder error, total 20 meters and HDOP = 1.5, then the expected determination error space will be equal to 20 x 1.5 = 30 meters. GPS receivers present information differently to evaluate accuracy using PDOP.

In addition to PDOP or HDOP, GQ (Geometric Quality) is used - the inverse value of HDOP, or a qualitative assessment in points. Many modern receivers display EPE (Estimated Position Error) directly in distance units. EPE takes into account the location of the satellites and the forecast of signal errors for each satellite depending on SA, the state of the atmosphere, and satellite clock errors transmitted as part of the ephemeris information.

Satellite geometry also becomes an issue when using the GPS receiver inside vehicles, in dense forests, mountains, or near tall buildings. When signals from individual satellites are blocked, the position of the remaining satellites will determine how accurate the GPS position will be, and their number will indicate whether the position can be determined at all. A good GPS receiver will show you not only which satellites are in use, but also their location, azimuth and elevation, so you can determine if a given satellite is having difficulty receiving.

Based on materials from the book “All about GPS navigators.”
Naiman V.S., Samoilov A.E., Ilyin N.R., Sheinis A.I.

Satellite positioning and navigation systems, originally developed for military needs, have recently found wide application in the civilian sphere. GPS/GLONASS monitoring of transport, monitoring people in need of care, monitoring the movements of employees, tracking animals, tracking luggage, geodesy and cartography are the main areas of use of satellite technologies.

Currently, there are two global satellite positioning systems created in the USA and the Russian Federation, and two regional ones, covering China, the countries of the European Union and a number of other countries in Europe and Asia. GLONASS monitoring and GPS monitoring are available in Russia.

GPS and GLONASS systems

GPS (Global Position System) is a satellite system whose development began in America in 1977. By 1993, the program was deployed, and by July 1995, the system was fully ready. Currently, the GPS space network consists of 32 satellites: 24 main, 6 backup. They orbit the Earth in a medium-high orbit (20,180 km) in six planes, with four main satellites in each.

On the ground there is a main control station and ten tracking stations, three of which transmit correction data to the latest generation satellites, which distribute them to the entire network.

The development of the GLONASS (Global Navigation Satellite System) system began in the USSR in 1982. The completion of the work was announced in December 2015. GLONASS requires 24 satellites to operate, 18 are sufficient to cover the territory and the Russian Federation, and the total number of satellites currently in orbit (including backups) is 27. They also move in a medium-high orbit, but at a lower altitude (19,140 km), in three planes, with eight main satellites in each.

GLONASS ground stations are located in Russia (14), Antarctica and Brazil (one each), and a number of additional stations are planned to be deployed.

The predecessor to GPS was the Transit system, developed in 1964 to control the launch of missiles from submarines. It could locate exclusively stationary objects with an accuracy of 50 m, and the only satellite was in view for only one hour a day. The GPS program was previously called DNSS and NAVSTAR. In the USSR, the creation of a navigation satellite system began in 1967 as part of the Cyclone program.

The main differences between GLONASS and GPS monitoring systems:

• American satellites move synchronously with the Earth, while Russian satellites move asynchronously;
• different heights and number of orbits;
• their different angles of inclination (about 55° for GPS, 64.8° for GLONASS);
• different signal formats and operating frequencies.

Benefits of GPS

• GPS is the oldest existing positioning system; it was fully operational before the Russian one.
• Reliability comes from using a larger number of redundant satellites.
• Positioning occurs with a smaller error than GLONASS (on average 4 m, and for the latest generation satellites - 60–90 cm).
• Many devices support the system.

Advantages of the GLONASS system

• The position of asynchronous satellites in orbit is more stable, which makes them easier to control. Regular adjustments are not required. This advantage is important for specialists, not consumers.
• The system was created in Russia, therefore it ensures reliable signal reception and positioning accuracy in northern latitudes. This is achieved due to the greater angle of inclination of satellite orbits.
• GLONASS is a domestic system and will remain available to Russians if GPS is turned off.

Disadvantages of the GPS system

• Satellites rotate synchronously with the rotation of the Earth, so accurate positioning requires the operation of corrective stations.
• A low tilt angle does not provide a good signal and accurate positioning in polar regions and high latitudes.
• The right to control the system belongs to the military, and they can distort the signal or completely disable GPS for civilians or for other countries in the event of a conflict with them. Therefore, although GPS for transport is more accurate and convenient, GLONASS is more reliable.

Disadvantages of the GLONASS system

• The development of the system began later and until recently was carried out with a significant lag behind the Americans (crisis, financial abuse, theft).
• Incomplete set of satellites. The service life of Russian satellites is shorter than that of American satellites, they require repair more often, so the accuracy of navigation in a number of areas is reduced.
• GLONASS satellite vehicle monitoring is more expensive than GPS due to the high cost of devices adapted to work with the domestic positioning system.
• Lack of software for smartphones and PDAs. GLONASS modules were designed for navigators. For compact portable devices today, the more common and affordable option is support for GPS-GLONASS or GPS only.

Summary

GPS and GLONASS systems are complementary. The optimal solution is satellite GPS-GLONASS monitoring. Devices with two systems, for example, GPS markers with the M-Plata GLONASS module, provide high positioning accuracy and reliable operation. If for positioning exclusively using GLONASS the error averages 6 m, and for GPS – 4 m, then when using two systems simultaneously it decreases to 1.5 m. But such devices with two microchips are more expensive.

GLONASS was developed specifically for Russian latitudes and is potentially capable of providing high accuracy; due to its understaffing with satellites, the real advantage is still on the side of GPS. The advantages of the American system are the availability and wide selection of GPS-enabled devices.

Purpose

GPS (Global Positioning System) allows you to accurately determine the three-dimensional coordinates of an object equipped with a GPS receiver: latitude, longitude, altitude above sea level, as well as its speed, direction of movement and current time.

Short story

The GPS system was developed by the US Department of Defense. Work on this project, called NAVSTAR (NAVigation System with Timing and Ranging - navigation system for determining time and range), began back in the 70s. The first satellite of the system was launched into orbit in 1974, and the last of the 24 needed to cover the entire Earth only in 1993. Initially, GPS was intended for use by the US military (navigation, missile guidance, etc.), but since 1983, when it was shot down a Korean Airlines plane accidentally intruded into Soviet territory, the use of GPS was allowed for civilians as well. At the same time, the accuracy of the transmitted signal was coarsened using a special algorithm, but in 2000 this limitation was lifted. The US Department of Defense continues to maintain and upgrade the GPS system. It was this complete dependence of the system's performance on the government of one country (for example, during the first Gulf War, the civilian sector of GPS was turned off) that prompted other countries to develop alternative navigation systems (Russian - GLONASS, European - GALILEO, Chinese - Beidou).

Principles of determining coordinates

The principle of determining the coordinates of an object in the GPS system is based on calculating the distance from it to several satellites, the exact coordinates of which are known. Information about the distance to at least 3 satellites allows you to determine the coordinates of an object as the point of intersection of spheres, the center of which is the satellites, and the radius is the measured distance.

In fact, there are two points of intersection of the spheres, but one of them can be discarded because it is either deep inside the Earth or very high above its surface. The distance to each satellite is defined as the time it takes for a radio signal to travel from the satellite to the receiver multiplied by the speed of light. The problem arises of accurately determining the transit time of a radio signal. It is solved by generating and transmitting a signal from the satellite, modulated using a special sequence. Exactly the same signal is generated in the GPS receiver, and analysis of the lag of the received signal from the internal signal makes it possible to determine its travel time.

To accurately determine the signal travel time, the clocks of the GPS receiver and satellite must be synchronized as much as possible; a deviation of even a few microseconds leads to a measurement error of tens of kilometers. The satellite has high-precision atomic clocks for these purposes. It is impossible to install a similar clock in a GPS receiver (regular quartz clocks are used), so additional signals from at least one more satellite are used to synchronize time. It is assumed that if the time in the GPS receiver is precisely synchronized, then a circle with a radius equal to the distance from the fourth satellite will intersect the same point as the circles from the other three satellites. The GPS receiver adjusts its clock until this condition is met. Thus, to accurately determine the position of an object in three-dimensional space (3D), signals from at least 4 satellites are required (from 3 satellites without determining the height above the earth’s surface - 2D). In practice, with good visibility of the sky, GPS receivers receive signals from many satellites at once (up to 10-12), which allows them to synchronize clocks and determine coordinates with fairly high accuracy.

Along with the sequence by which the signal propagation time is determined, each satellite transmits binary information - an almanac and ephemeris. The almanac contains information about the current state and estimated orbit of all satellites (having received information from one satellite, it becomes possible to narrow the search sectors for signals from other satellites). Ephemeris - updated information about the orbit of a specific satellite transmitting a signal (the actual orbit of the satellite may differ from the calculated one). It is the exact data about the current position of the satellites that allows the GPS receiver to calculate its own location relative to them.

GPS Accuracy

The typical accuracy of determining coordinates by GPS receivers in the horizontal plane is approximately 1-2 meters (provided good visibility of the sky). The accuracy of determining altitude above sea level is usually 2-5 times lower than the accuracy of determining coordinates under the same conditions (i.e., in ideal conditions, 2-10 meters).

The level of signal reception from satellites, and as a result the accuracy of determining coordinates, deteriorates under dense foliage of trees or due to very heavy clouds. Also, normal reception of GPS signals can be impaired by interference from many terrestrial radio sources. However, the main factor influencing the decrease in GPS accuracy is incomplete visibility of the sky. This is especially evident when the GPS receiver is located in dense urban areas, when a significant part of the sky is hidden by nearby buildings, canopies and other obstacles. The accuracy of determining coordinates can drop to 20-30 meters, and sometimes more. Obstacles do not allow signals from some satellites potentially available at a given point on the Earth to pass through. This leads to the fact that calculations are carried out using a smaller number of signals from satellites located primarily in one sector of the sky. The displacement usually occurs in a plane perpendicular to the obstacle.

In general, if we talk about the accuracy of GPS in urban conditions, based on accumulated statistical data and our own experience, we can draw the following conclusions. The accuracy of determining coordinates when the vehicle is in an open area (parking lot, square, etc.) and when driving along major highways and multi-lane roads will be 1-2 meters. When driving along narrow streets, especially when there are closely spaced houses along them, the accuracy will be 4-10 meters. When the car is in “yard wells”, very close to high-rise buildings, etc. accuracy can drop down to 20-30 meters.

Of course, the accuracy of determining coordinates greatly depends on the quality of the GPS receiver itself, as well as the antennas used and their correct placement on the vehicle