GPS receiver sensitivity testing and results

The problem investigated in this thesis was the determination of single channel code tracking accuracy and loss of lock points for commercial hardware and software Global Positioning System (GPS) receivers. The NovAtel OEM4, NordNav R30, and SiRFstar III commercial receivers tested exhibited loss of lock points of 28, 26, and 14 dB-Hz, respectively, when tracking a single GPS satellite. Once this tracking baseline was established algorithms for improving weak signal tracking were researched and implemented in the M&barbelow;ATLAB S&barbelow;oftware D&barbelow;efined R&barbelow;eceiver G&barbelow;PS T&barbelow;racking F&barbelow;ramework (MSGTF). Critical to the success of this research was the development of an experimental setup where the commercial receivers could be tested simultaneously with a single channel GPS signal simulator. Fine time control of the simulator through serial commands was crucial so that an accurate pseudorange truth reference could be developed from the simulator profile. This truth reference was compared against receiver outputs and these experimental jitter results were ultimately compared to theoretical predictions and agreed well. The NovAtel OEM4 code loop bandwidth of 1 Hz was determined from this comparison with theoretical predictions. Intermediate frequency (IF) samples logged during the commercial receiver evaluation were processed by the MSGTF using several tracking algorithm improvements. Weak signal tracking techniques including extended coherent integration time and open loop carrier tracking were successfully implemented in the MSGTF. Signal tracking utilizing a known Doppler profile showed up to a 5 dB improvement in loss of lock power level as compared to conventional carrier loop tracking.