Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
PERFORMANCE IMPROVEMENTS TO THE EXISTING GPS CONFIGURATION 99 of the GPS satellites as well as the clock errors for the monitor site receivers. Updated orbit and clock corrections are uploaded to each satellite once a day. With the current GPS constellation, the clock and ephemeris errors contribute approximately 3.4 and 1.4 meters (1Ï), respectively, to the SPS and PPS error budget, for a combined error of 3.6 meters (1Ï)32, as shown in Table 3-1.33 Once SA, the atmospheric, receiver noise, and multipath errors have been eliminated or reduced, ephemeris and clock errors become the largest contributors to the UERE. As shown below, several methods can be used to reduce combined clock and ephemeris errors to increase accuracy and improve overall performance. Accuracy Improvements Planned Experiments Involving Correction Updates and Additional Monitoring Stations. An innovative, near-term strategy for improving PPS accuracy and integrity has been investigated by the Air Force, and an experiment to test the strategy is expected to begin in the spring of 1995. The experiment involves uploading pseudorange corrections for all satellites with each scheduled, individual satellite upload.34 These corrections would be made available to PPS users in the navigation message. A PPS receiver can decode the messages from all satellites it is tracking and apply the most recent correction set. The Air Force expects that this will improve the combined error contribution of clock and ephemeris for PPS users by half, to approximately 2 meters (1Ï). If SA is turned to zero as previously recommended, SPS users will not receive the same benefit from this experiment as PPS users unless current security classification policies are changed to allow the most recent clock and ephemeris parameters to be broadcast from each satellite unencrypted.35 In conjunction with the above experiment, the Air Force is investigating another enhancement that could provide further reduction in the combined PPS clock and ephemeris error. This enhancement involves the integration of data from five Defense Mapping Agency (DMA) GPS monitoring sites with the existing Air Force operational control segment in a simulated Kalman Filter. By including additional data from the DMA sites, which are located at higher latitudes than the Air Force sites, an additional 15 percent improvement in combined clock and ephemeris accuracy can be anticipated, based on tests previously 32 The error of 3.6 meters (1Ï) was obtained by taking the square root of the sum of the squares of 3.4 and 1.4 meters (1Ï). 33 J. F. Zumberge and W. I. Bertiger, ''Ephemeris and Clock Navigation Message Accuracy in the Global Positioning System," Volume I, Chapter 16. Edited by B. W. Parkinson, J. J. Spilker, P. Axelrad, and P. Enge. To be published by AIAA, in press, 1995. 34 Satellites are normally uploaded once per day. 35 Currently the most recent clock and ephemeris updates are broadcast in an encrypted portion of the navigation message. Clock and ephemeris parameters less than 48 hours old are classified.
PERFORMANCE IMPROVEMENTS TO THE EXISTING GPS CONFIGURATION 100 conducted by the DMA.36 It should be emphasized that this experiment will be conducted as a software simulation only, so PPS users will not actually observe the additional 15 percent simulated improvement. Recommended Implementation of More Frequent Updates and Additional Monitoring Stations. Full operational implementation of the first experiment, which involves uploading of clock and ephemeris corrections for all satellites with each scheduled, individual satellite upload, should not be difficult to accomplish and would appear to reduce the combined clock and ephemeris error to half of its current value. Operational implementation of the second planned experiment, which involves the incorporation of DMA monitor site data, is more difficult to achieve. While well-distributed geometrically, DMA GPS monitor stations do not have secure communications data links to the master control station. Existing Air Force sites, which are used for other purposes, have secure data links to Air Force Space Command (co-located with the GPS master control station), but are not well distributed in latitude for GPS monitoring and do not have GPS receivers. Additional GPS monitoring sites are expected to improve stand-alone GPS accuracy. More importantly, a well- distributed set of monitor sites would allow continuous tracking of each satellite, enabling the prompt detection of satellite failures. An estimated cost of $9 million for using DMA data in real-time and an estimated cost of co- locating Air Force monitor stations at DMA sites was provided to the committee.37 The DOD's more frequent satellite navigation correction update strategy should be fully implemented as soon as possible following the successful test demonstration of its effectiveness. In addition, the current security classification policy should be examined to determine the feasibility of relaxing the 48-hour embargo on the clock and ephemeris parameters to civilian users. Additional GPS monitoring stations should be added to the existing operational control segment. Comparison studies between cost and location should be completed to determine if Defense Mapping Agency or Air Force sites should be used. Recommended Use of a Non-Partitioned Kalman Filter with Improved Dynamic Models. The original computer hardware used for the OCS was not capable of processing all satellites in a single Kalman Filter. The existing software was written with this limitation as well. The hardware has since been upgraded, leaving only the software to restrict full processing of all satellite clock and ephemeris data simultaneously. Unfortunately, there currently are no definite plans to upgrade the Kalman Filter software, including the dynamic 36 Stephen Malys, DMA, Viewgraphs from presentation at the PAWG 1993, Colorado Springs, Colorado. 37 Information provided by the Aerospace Corporation, 21 February 1995.
PERFORMANCE IMPROVEMENTS TO THE EXISTING GPS CONFIGURATION 101 model. It is possible that the winning contractor of the 1995 contract may choose to eliminate the partitions, but there is not a specified requirement to do so. Based on recent improvements to the DMA's Kalman Filter, which originally had a configuration similar to the GPS Kalman Filter, use of an updated, non-partitioned GPS Kalman Filter is expected to reduce the combined clock and ephemeris error by 15 percent.38 Furthermore, an additional 5 percent improvement can be achieved by using improved dynamic models in the Kalman Filter, which would allow better predictions of satellite behavior 1 day ahead.39 An estimated cost of $7.5 for upgrading the Kalman Filter and improving its dynamic models was provided to the committee.40 The operational control segment Kalman Filter should be improved to solve for all GPS satellites' clock and ephemeris errors simultaneously through the elimination of partitioning and the inclusion of more accurate dynamic models. These changes should be implemented in the 1995 OCS upgrade request for proposal. The combined clock and ephemeris improvement obtained with each of the above upgrades is shown in Table 3-9. If all three of the recommendations above are implemented, the combined clock and ephemeris error is expected to be approximately 1.2 meters (1s). As shown in Table 3-10 and Figure 3-5, if: (1) SA is turned to zero; (2) an additional GPS L-band signal is added; (3) more advanced receivers are utilized; and (4) each of the clock and ephemeris accuracy improvements are implemented, then a stand-alone GPS SPS accuracy of 5.4 meters (2 drms) with a narrow, L-band signal should be obtainable, and a stand-alone GPS SPS accuracy of 4.9 meters (2 drms) with a wide-band signal should be obtainable.41 In addition, as shown in Table 3-11, a PPS accuracy of 4.2 meters (2 drms) (1.8 meters CEP) also would be obtainable. With stand-alone accuracies at this level, many civilian and military accuracy requirements, such as the following will be met: â¢ Aviation â Category I approach and landing. â¢ Maritime â Recreational boating, vessel-tracking services, and harbor/harbor approach requirements. 38 Stephen Malys, DMA, Viewgraphs from presentation at the PAWG 1993 meeting, Colorado Springs, Colorado. 39 Stephen Malys, DMA, Viewgraphs from presentation at the PAWG 1993 meeting, Colorado Springs, Colorado. 40 Information provided by the Aerospace Corporation, 21 February 1995. 41 Civil users would have access to this level of accuracy only if the 48-hour embargo on clock and ephemeris parameters is lifted.
PERFORMANCE IMPROVEMENTS TO THE EXISTING GPS CONFIGURATION 102 â¢ ITS â Infrastructure management, highway navigation and guidance, mayday incident and alert, automated bus/railstop annunciation, collision avoidance (hazardous situation), and vehicle or cargo location (hazardous material transport). â¢ Earth Science â Oceanographic navigation and real-time positioning. â¢ Spacecraft â Real-time satellite orbit determination. â¢ Military â Precision-guided munitions. Table 3-9 Reduction of Combined Clock and Ephemeris Errors Enhancement Anticipated Combined Clock and Ephemeris Error Improvement over Existing Combined Error of 3.6 meters (1Ï) Correction Updates (50% reduction) 1.8 meters Additional Monitor Stations 1.5 meters (additional 15% reduction) Non-partitioned Kalman Filter 1.3 meters (additional 15% reduction) Improved Dynamic Model 1.2 meters (additional 5% reduction)
PERFORMANCE IMPROVEMENTS TO THE EXISTING GPS CONFIGURATION Figure 3-5 103 Approximate stand-alone horizontal SPS accuracy, 2 drms, resulting from recommended improvements and enhancements.
Table 3-10 Impact of Reduced Clock and Ephemeris Error on SPS Stand-Alone Accuracy Error Source Typical Range Error Magnitude (meters, 1Ï) SPS With II/IIA Satellites SPS Improved (no SA, additional narrow L-band SPS Improved (no SA, additional wide L-band signal) signal) 1237.83 1258.29 1841.40 1258.29 1841.40 Narrow-band, C/A- Narrow-band, C/A-type Narrow-band, C/A-type Wide-band, Wide-band, type code code code P-type code P-type cc Selective Availability 24.0 0.0 0.0 0.0 0.0 0 Atmospheric Error Ionospheric 7.0 0.01 0.01 0.01 0.01 0.01 Tropospheric 0.7 0.2 0.2 0.2 0.2 0.2 Clock and Ephemeris Error 3.6 1.2 1.2 1.2 1.2 1.2 Receiver Noise 1.5 0.6 0.7 0.9 0.5 0.8 Multipath 1.2 1.2 1.6 2.3 1.0 1.9 Total User Equivalent Range 253 1.8 2.1 2.8 1.7 2.4 Error (UERE) Typical Horizontal DOP 1.5 1.5 1.5 1.5 1.5 1.5 (HDOP) Total Stand-Alone 76.0 5.4 6.4 8.3 4.9 7.1 PERFORMANCE IMPROVEMENTS TO THE EXISTING GPS CONFIGURATION Horizontal Accuracy (2 drms) 104
PERFORMANCE IMPROVEMENTS TO THE EXISTING GPS CONFIGURATION 105 Table 3-11 Impact of Reduced Clock and Ephemeris Error on PPS Stand-Alone Accuracy Error Source Typical Range Error Magnitude (meters, 1Ï) PPS with II/IIA satellites PPS Improved Selective Availability 0.0 0.0 Atmospheric Error Ionospheric 0.01 0.01 Tropospheric 0.2 0.2 Clock and Ephemeris Error 3.6 1.2 Receiver Noise 0.3 0.3 Multipath 0.6 0.6 Total User Equivalent Range Error (UERE) 3.7 1.4 Typical Horizontal DOP (HDOP) 1.5 1.5 Total Stand-Alone Horizontal Accuracy,2 11.1 4.2 As with the previous performance improvements, the increased positioning accuracy achieved by reducing clock and ephemeris errors also enhances availability. For example, for a stand-alone horizontal accuracy of 100 meters, the availability of four satellites would increase from the previous value of 99.96 percent to 99.97 percent. The improved RAIM availability is shown in Table 3-12.42 Table 3-12 Effect of SA Removal, Dual-Frequency Capability and Reduced Clock and Ephemeris Errors on RAIM Availability for Aviation Applications a Aviation Application Availability With SA Turned to Zero Availability With SA Turned to Zero, and L4 Added L4 Added, and Reduced Clock and Ephemeris Error Phase of Flight Protection Limit 21 Satellitesb 24 Satellitesc 21 Satellites 24 Satellites En Route 2.0 nautical miles 96.80% 100.00% 97.08% 100.00% Terminal Area 1.0 nautical miles 95.19% 99.98% 95.70% 100.00% Non-Precision 03 nautical 93.12% 100.00%d 94.36% 100.00%d 42 Based on analysis conducted by the MITRE Corporation for the NRC committee, 7 February 1995. For more details, see footnote 1 earlier in this chapter.