| This dissertation investigates the potential performance of CPS time transfer in near real-time between a pair of co-orbiting satellites. Simulated spaceborne CPS data sets are produced with a sophisticated hardware-in-the-loop test bed. Three different filters are developed to estimate relative clock states based upon CPS pseudorange and carrier phase measurements. The performance of the filters is evaluated by direct comparison to measurements of the clock using a hydrogen maser and laboratory test equipment.; The test bed includes a spaceborne CPS receiver, CPS signal simulators, and a universal time interval counter. Simulated CPS signal errors include ionosphere, broadcast ephemeris errors, and multipath. The spacecraft are simulated in a TOPEX-like orbit with along-track separations of 50 and 100 km.; An existing CPS orbit determination utility processes the measurements and provides spacecraft position information to the relative clock filters. The Raw Single Difference Only (SDO) clock filter processes only raw single difference pseudoranges. The Smoothed SDO filter smooths the single difference pseudoranges with single difference carrier phase in a complementary filtering scheme. The Time Difference carrier Phase (TDCP) filter processes the current single difference pseudorange along with the difference between the current and previous single difference accumulated carrier phase. Each filter performs state updates every 10 sec.; Each of the three clock filtering algorithms produces relative bias and frequency RMS errors of less than 1 nsec and 0.014 nsec/sec, respectively. The TDCP filter performs the best, with average relative clock bias and frequency RMS errors of less than 0.5 nsec and 0.009 nsec/sec, respectively. The estimation results show excellent potential for near real-time clock comparisons on-orbit. Furthermore, the test bed and testing methodology developed here can be used for evaluation of alternate measurement processing techniques and comparison of receiver performance. |
