Advanced GPS signal tracking applications under high dynamic conditions

The Global Positioning System (GPS) is widely used for guided missiles and rockets to track their targets and to navigate on the desired route. However, the fact that these platforms have high dynamics such as speed and acceleration significantly affects GPS performance. The most critical operation on a GPS receiver is to track GPS signals continuously. A receiver processes the signals received from the satellite to determine from which satellite the signals arrive, and after that, tracks each satellite signal separately with the signal tracking circuits. If the receiver fails to track the satellite signals, positioning cannot be performed. For this reason, reliable signal tracking mechanisms are needed in GPS receivers. In order for the receiver to be able to determine the position, the signal tracking circuit in the receiver must lock to the carrier and code phase / frequency of the signals received from the GPS satellites. In GPS receivers, carrier phase / frequency and code phase are tracked by Phase Locked Loop (PLL) circuits. PLL performance is affected by factors such as low signal-to-noise ratio, Doppler frequency shift, thermal noise and fluctuation in the accuracy of the sensitive oscillator in the receiver due to vibration and acceleration. The most dominant problem among the above factors is the Doppler frequency shift due to high relative velocity between the satellite and the receiver. In order to be able to track the signal under high dynamic conditions properly, the PLL must be able to tolerate the carrier frequency deviations caused by the Doppler Effect and be locked to the correct carrier frequency. Nonetheless, simple signal tracking circuits cannot operate under high dynamic conditions. Therefore, it is necessary to use advanced signal tracking circuits. Inertial navigation systems can also be used to help the existing conventional structures for continuous signal tracking, but these systems, which are directly proportional to the cost of reliability, are a huge financial burden for special applications. Thus, it is preferable to use advanced signal tracking schemes that can be performed in a software or hardware environment. In this thesis, the characteristics of the GPS signal, the basic operations performed by a GPS receiver and the GPS signal tracking mechanism in detail are discussed. The 2nd degree FLL assisted 3rd degree PLL, Kalman filter PLL and Fuzzy controlled PLL structures that can perform continuous signal tracking under high dynamic conditions have been investigated. 2nd Degree FLL assisted 3rd Degree PLL is one of the most widely used signal tracking structures and it can reduce phase and frequency errors caused by high velocity and acceleration to minimum with time thanks to the high order loop filter. Kalman filter PLL estimates the carrier phase and frequency using the carrier phase dynamic model of conventional PLL structures, phase and frequency error generated in the discriminators and Doppler frequency shift rate. Fuzzy controlled PLL finds the carrier frequency by passing the phase and frequency errors calculated in the discriminators through a fuzzy logic system to calculate the different loop noise bandwidths according to the error rate and to determine the carrier frequency in the best way. In the application part of the thesis study, the algorithms belonging to the three different signal tracking structures mentioned above were created and these algorithms were implemented into a software based GPS receiver project in the MATLAB environment. The GPS signals required for verification of the algorithms are provided from a GPS satellite simulator device in the laboratory environment. The performances of the conventional tracking structure and proposed three advanced signal tracking structures were tested and compared with each other under a very high dynamic scenario. According to the test results, it has been shown that above mentioned structures can track more accurate carrier frequency under high dynamic conditions than the conventional structure. 2nd Degree FLL assisted 3rd Degree PLL has tracked Doppler frequency change without losing lock by minimizing phase and frequency errors which caused by high dynamics. Kalman filter PLL has tracked signals thanks to its carrier phase dynamic model even noisy errors are measured by the discriminators. Also, fuzzy controlled PLL have managed to track dynamic signals through its adaptive bandwidth feature. Under very high dynamic parts of scenario, it has increased the loop bandwidth to keep in locked state. Besides, under normal dynamic parts of scenario, it has decreased the loop bandwidth to track Doppler frequency change more accureately. Performances of proposed enhanced signal tracking structures were compared according to crucial parameters related to signal tracking and superiorities of these structures were specified. In order to specify these superiorities, the performances of these structures were evaluated with different integration periods, loop bandwidths and applied signal power levels. When the integration time is increased by 5 ms, it has been shown that all the advanced structures can track the signals of the dynamic scenario, however the loop locks are lost and the signal tracking can not be done since the beginning of accelerated movement with the integration time of 10 ms. In the 5 ms integration period, it has been shown that thermal noise jitter occured less according to 1 and 2 ms cases, and more accurate Doppler frequency estimation has been achieved. Among the three advanced signal tracking structures with the same integration period, it has been shown that Kalman filter and fuzzy controlled PLL structures have tracked Doppler frequency with less noise than 2nd degree FLL assisted 3rd degree PLL structure due to their adaptive gain or bandwidth characteristics. The relation between loop bandwidth and tolerance to high dynamics is indicated by using 2nd degree FLL assisted 3rd degree PLL and fuzzy controlled PLL structures. Accordingly, it has been shown that the loop bandwidths must be kept high at high dynamic moments to prevent the loss of lock of the PLLs. Finally, the performances of the proposed signal tracking structures were compared when the signals from the satellites reach the receiver with low power. Test results have shown that all advanced signal tracking structures have successfully managed to track signals at a signal strength level of at least -135 dBm.