Engineering topics
Publications and source records attributed to Wu, S. C..
Report summarizes method for synchronizing clocks at intercontinental distances employing satellites of Global Positioning System (GPS) in high Earth orbit and transit satellite in orbit at relatively low altitude of about 1,300 km. When fully implemented, method expected to supply precise time measurements for world-wide communication and navigation.
The first field test of NASA's Global Positioning System (GPS) Geodetic Program took place in March of 1985. The principal objective of this test was the demonstration of the feasibility of the fiducial station approach to precise GPS-based geodesy and orbit determination. Other objectives included an assessment of the performance of the several GPS receiver types involved in these field tests and the testing of the GIPSY software for GPS data analysis. In this article, the GIPSY (GPS Inferred Positioning System) software system is described and baseline solutions are examined for consistency with independent measurements made using very long baseline interferometry.
As part of an international campaign to develop precise geodetic applications of the Global Positioning System (GPS), the Jet Propulsion Laboratory is conducting a demonstration of differential GPS tracking using Landsat-5. Two strategies have been investigated: one in which only the Landsat-5 orbit is estimated, and one in which both the Landsat-5 and GPS orbits are estimated together. Error studies show that under the limited conditions of the experiment, three-dimensional Landsat-5 position accuracies of about 5 m with the first strategy and 2 m with the second strategy can be achieved over a 20-min period of good observing geometry. Orbit determination results using a version of the first strategy appear to achieve the 5 m goal. This is supported by various formal error measures and independent comparisons. The more powerful strategy has yet to be carried out.
A technique is described for employing the Global Positioning System (GPS) to determine the position of a low earth orbiter with decimeter accuracy without the need for user dynamic models. A differential observing strategy is used requiring a GPS receiver on the user vehicle and a network of six ground receivers. The technique uses the continuous record of position change obtained from GPS carrier phase to smooth position measurements made with pseudo-range. The result is a computationally efficient technique that can deliver decimeter accuracy down to the lowest altitude orbits.
Report describes accurate method of determining orbit of low-altitude Earth satellite. Method insensitive to clock errors and provides continuous determination of orbit. Report examines effects of employing fewer ground stations, applying different solution strategies, and introducing fictitious thrust parameters to minimize geopotential modeling error - one of major error sources.
Time synchronization between two sites using differential GPS has been investigated by a number of researchers. When the two sites are widely separated, the common view period of any GPS satellite becomes shorter; low elevation observations are inevitable. This increase the corrupting effects of the atmospheric delay and, at the same time, narrows the window for such time synchronization. This difficulty can be alleviated by synchronization. This difficulty can be alleviated by using a transit site located midway between the two main sites. The main sites can now look at different GPS satellites which are also in view at the transit site. However, a ground transit site may not always be conveniently available, especially across the Pacific Ocean; also, the inclusion of a ground transit site introduce additional errors due to its location error and local atmospheric delay. An alternative is to use a low Earth orbiter (LEO) as the transit site. A LEO is superior to a ground transit site in three ways: (1) It covers a large part of the Earth in a short period of time and, hence, a single LEO provides worldwide transit services; (2) it is above the troposphere and thus its inclusion does not introduce additional tropospheric delay error; and (3) it provides strong dynamics needed to improve GPS satellite positions which are of importance to ultraprecise time synchronization.
Two techniques mathematically estimate effects of flicker noise in pulse trains used as clock signals.
An MLS (microwave landing system) certification system based on the Global Positioning System (GPS) is described. To determine the position history of the flight inspection aircraft during runway approach, signals from the GPS satellites, together with on-board radar altimetry, are used. It is shown that the aircraft position relative to a fixed point on the runway at threshold can be determined to about 30 cm vertically and 1 m horizontally. A requirement of the system is that the GPS receivers be placed on each flight inspection aircraft and at selected ground sites. The effects of different error sources on the determination of aircraft instantaneous position and its dynamics are analyzed.
Results of covariance analyses are presented for the determination of baseline vectors of length from 150 m to 4000 km. The data are doubly differenced ranges derived from the signals transmitted by the satellites of the Global Positioning System. The modeling of various error sources is described. The results indicate the low cost baseline determination system is capable of subdecimeter accuracy for baselines up to 1000 km in length.
Baselines between a ground station and a geosynchronous orbiter provide high resolution Delta VLBI data which is beyond the capability of ground-based interferometry. The effects of possible error sources on such Delta VLBI data for the determination of spacecraft angle and angle rate are investigated. For comparison, the effects on spacecraft-only VLBI are also studied.
Four approaches to radio interferometric tracking of geosynchronous satellites are analyzed and compared. Quasar-based differential very-long-baseline interferometry, which requires a very sensitive receiver, can achieve meter-level position accuracy with a two-baseline system. Satellite-based differential VLBI gives somewhat lower accuracy with a compact, inexpensive receiver. Nondifferential VLBI, using less precise media and clock calibrations obtained by observing the GPS satellites, still gives 5-10 m position accuracy with two baselines. For a sufficiently inclined orbit, all interferometric approaches can yield six-component satellite state from a single baseline.
This paper investigates the use of one type of GPS-signal derived measurement, the doubly differenced GPS RF Doppler, in the orbit determination of low-altitude earth satellites. This measurement is simple to make, is insensitive to clock errors and provides a continuous determination of the user satellite orbit. Analysis shows that, with a constellation of 18 GPS satellites and 13 ground stations, the 1300 km altitude of a user satellite (TOPEX) can be determined to 5 cm in 2 hours of tracking. Examination is made of the effects of employing fewer ground stations, of different solution strategies, and of introducing fictitious thrust parameters to alleviate the geopotential modeling error which is one of the major error sources.
A scheme for flicker noise generation is given. The second approach is that of successive segmentation: A clock fluctuation is represented by 2N piecewise linear segments and then converted into a summation of N+1 triangular pulse train functions. The statistics of the clock instability are then formulated in terms of two sample variances at N+1 specified averaging times. The summation converges very rapidly that a value of N 6 is seldom necessary. An application to radio interferometric geodesy shows excellent agreement between the two approaches. Limitations to and the relative merits of the two approaches are discussed.
Different types of media effects on ARIES baseline determination are compared. The effectiveness of simple ionospheric calibration models are studied. To perform the covariance analysis, an ARIES observation sequence needs to be assumed. For the current purposes, the observation sequence is selected to be that of experiment 80D over the JPL/Goldstone baseline (approximately 180 km). This experiment consisted of 96 observations over a period of approximately 25 hours on March 25 to 26, 1980. It is found through covariance analyses that the component most sensitive to media depends heavily upon the correlation, between the two stations, of the media effects. It is also found that relying on the cancellation of ionospheric delays between the two ray paths of VLBI observations at S band results in a large error in baseline length determination. High degree removal of ionospheric effects is possible with a crude model, providing correct diurnal peak and minimum ionospheric levels are input.
An overview of the design of the ORION mobile system is presented. System capability and performance characteristics are outlined. Functional requirements and key performance parameters are stated for each of the nine subsystems. A master design and implementation schedule is given.
A set of experiments in the use of Differential Very Long Baseline Interferometry (delta VLBI) for spacecraft navigation were completed. Data using both Voyager spacecraft and a single quasar were acquired during the Jupiter encounter time period. The data were processed and analyzed to assess the navigation accuracy of delta VLBI. The data reduction and techniques for assessing data quality and consistency are discussed.
Error sources in the Operational Radio Interferometry Observing Network (ORION) are addressed. Specifically, the performance of the S-band receiver and the optimum allocation of data volume between S- and X-band observations is studied. It is found that the requirements on the S-band receiver are not very stringent. The system temperature of the ORION S-band receiver has a very small effect on baseline determination accuracy. It is unwise to pay a high cost for an S-band receiver with a low system temperature; a system temperature as high as 240 K is tolerable. Only 20 to 40 percent of data volume is to be allocated to S-band observations for minimum baseline error. The error remains low over a rather wide range of data volume ratio. Hence precise allocation of the 14 pairs of the Mark 3 very long base interferometer data channels between S- and X-band observations is not critical.
A set of experiments in the use of Differential Very Long Baseline Interferometry for spacecraft navigation have been completed. Data using both Voyager spacecraft and a single quasar were acquired during the Jupiter encounter time period. The data were processed and analyzed in order to assess the navigation accuracy of Differential Very Long Baseline Interferometry. The paper focuses on the data reduction and techniques for assessing data quality and consistency.