Design Of High Performance GNSS Receiver Based On Rake Architecture

The GNSS (Global Navigation Satellite System) system is more and more widely used in various fields with the development of the system and continuous improvement of the receiver's performance. China is also actively developing its own Beidou satellite navigation system. However, due to the lack of receiver technology accumulation, the core technology of high performance receiver has increasingly become the bottleneck of the navigation industry chain in China. Therefore, improving the performance of navigation receiver has become an important and urgent task. This thesis researches in-depth on GNSS receiver. It proposes a new anti-multipath architecture named MRake (Modified Rake), and researches in-depth on the impact of MRake architecture on performance of the receiver's main aspects. It has made certain achievements in improving the anti-multipath performance, acquisition speed, tracking sensitivity and position calculation precision. The achievements provide a theoretical and practical guidance for high performance receiver research in China. The main contributions of this thesis are summarized as follows:First, to solve the conflict between the performance and resource consumption of existing anti-multipath techniques, MRake architecture is used to reduce the multipath error in GPS receiver. It tracks direct-path and multipath components of the received signal separately, and uses the time-delay of direct-path component to calculate the user's position. The proposed MRake structure has good anti-multipath performance without consumption of many hardware resources or complex calculation.Secondly, since MRake architecture can track the direct and multipath signals separately, it also can play an important role in capturing signals quickly and efficiently. Using the parallel pseudo-code searching and serial carrier searching method in time domain can make full use of hareware resources in MRake structure to improve the acquisition speed. On this basis, use extended multiple correlator (XMC) method to further improve the acquisition speed without increasing the hardware resources. Treat the acquired satellite signal as direct-path signal, and then capture and separate the multipath signals by MRake structure. Thirdly, in order to meet the demand of navigation in complex environments such as indoor environment, the technology to improve the tracking sensitivity of the receiver is studied in depth. Based on the quantitative analysis of the common techniques and summarization of their problems, an MCC (Modified the Classic of Costas) phase detector algorithm is proposed. It modifies the CC (Classic Costas) phase detector, and can significantly improve the tracking sensitivity without increaseing the calculation.Finally, to solve the problems of two most commonly used position calculation algorithms, least square method and Kalman filtering method, two new algorithms are proposed. One is MKF (Modified Kalman Filtering) that can improve the positioning accuracy and avoid the complexity calculation of the noise covariance matrix. Another one is PF (Particle Filtering) that can further improve the positioning accuracy, and overcome the drawbacks of traditional Kalman filter that requires prior knowledge of receiver's dynamic model and noise properties.Based on the above theoretical study and program design, experiments with GPS signals generated by Spirent simulator have achieved the anticipated results. The research has important practical value and referential significance on the high performance navigation receiver design.