A Low Complexity Algorithm for Code Doppler Compensation Using FFT-Based Parallel Acquisition Architecture

Abstract

With the increasing competition in Global Navigation Satellite System (GNSS), the development of satellite navigation industry has become a big challenge for our country in recent years. Focusing on the development trend and effective demand of BeiDou system (BDs), a low complexity algorithm for code Doppler compensation using FFT-based parallel code searching architecture is proposed to correct Pseudo-Random Noise (PRN) code phase migration, which is caused by the code Doppler in the acquisition process of navigation signal in high dynamic and low Carrier-Power-to-Noise-Density Ratio (C/N0) environment. The new algorithm that is named as “Modified Estimation and Circular Shift of Code Phase (MECS-CP)” could cut down the influence of code Doppler on accumulation peak so as to increase the acquisition performance effectively. The MECS-CP algorithm is presented on the foundation of FFT-based full parallel acquisition algorithm. Firstly, the direction and magnitude of PRN code phase migration are estimated by the linear relationship between carrier Doppler and code Doppler in the process of correlation and coarse carrier Doppler compensation. Then, the PRN code correlation results are circularly shifted in phase dimension according to the estimation results, and then all corrected correlation outputs are kept in a three-dimensional array for later use. Finally, fine carrier Doppler compensation and coherent accumulation are carried out. This paper verifies the accuracy of code Doppler compensation, acquisition performance and computational complexity of the proposed MECS-CP algorithm by theoretical analyses and Monte-Carlo simulations. Both results show that the new algorithm could efficiently compensate the code Doppler and sharpen the accumulation peak of weak navigation signals in high dynamic environment, thus improving the peak-to-average power ratio (PAPR) and detection rate. Numerically, it is found that the proposed algorithm can achieve a detection rate of 0.9 and a false alarm rate of 10–5 at C/N0 = 31dBHz during a coherent accumulation time of 32 ms in simulation.