Research On Fast Acquisition Algorithm For L1/L2C Dual-frequency In-car GPS Software Receiver

Now GPS has been widely applied to car navigation, however, due to the higher demand of positioning accuracy for outdoor AGV and Intelligent Vehicle, traditional single frequency GPS receiver can not meet the positioning accuracy. The above problem can be effectively solved by the double-frequency In-car software receiver which appeared in America’s GPS modernization program. As the first step of baseband signal processing, acquisition performance is of great importance for following signal tracking and navigation data calculation. It’s significant to study on fast acquisition algorithms with superior performance for improving the positioning performance of L1/L2 C double-frequency software receiver and solving the exiting problem in vehicle navigation. This paper takes fast acquisition of L1 signal as starting-point and focuses on the studies of aided fast acquisition and direct rapid acquisition for L2 C signal.For fast acquisition of L1 signal, split-radix FFT instead of radix-2 FFT is introduced to do correlation in the traditional parallel code phase acquisition algorithm, which reduces calculated amount. For fast acquisition of L2 C signal, two perspectives is explored, namely, the aided fast acquisition under successful L1 acquisition and the direct fast acquisition under failed L1 acquisition in the shade environment.For the aided fast acquisition of L2 C signal, aiming at the lower code phase acquisition accuracy by using traditional aided acquisition method, a modified time-domain-correlation-based aided acquisition method and a frequency-domain-correlation-based aided acquisition method are present, both of which can improve the acquisition precision. The former one can keep basic unchanged computation cost and the later one can reduce the implementation complexity because of its same FFT-based implementation way with L1 signal fast acquisition.For the direct fast acquisition of L2 C signal, CM and CL codes are respectively studied. For CM code, two algorithms separately based on segmented FFT and based on segment and folding are proposed. The former can reduce the number of one-time FFT point and the later can decrease overall calculation amount, which are both helpful for the implementation of algorithms on embedded system. For CL code, a modified CL code acquisition algorithm combining Hyper code method and averaging correlation method is put forward, which reduces both the number of one-time FFT point and the overall calculation amount without affecting the acquisition sensitivity.The proposed algorithms are validated by both simulation signal source and real satellite signal, simulation and experiment results are consistent with theoretical analysis.