Research On Key Technologies For Robust And Adaptive GPS Software Receiver

GPS Software Receiver (GSR) is a Software Radio Technology based new receiver architecture. Because of the abilities of accessible to the inside architectures and reprogramebale to the kernel algorithms, GSR provides a powerful tool for GPS research. The users only need to develop high performance algorithm module to upgrade the whole system. The GSR has become one of the most important directions in GPS modernization. On one hand, the research on GSR provides opportunities to learn and master the key technology of receivers in satellites passive navigation system. These experiences will intensively promote the development of COMPASS satellite positioning system in our country. On the other hand, GPS signals are attenuated by complicated electromagnetic environment in some developed high-density cities, such as Hong Kong. In this area, the stability and continuity of GPS applications are seriously restricted. Therefore, it is necessary to intensively study the inside architecture of GPS receiver, and develop high performance kernel algorithm to enhance the system's robustness and precision.This dissertation studies the following techniques: multiple-measurements GSR research based on normal signals, weak signals acquisition algorithms, Kalman filter based carrier/CA code tracking algorithms and the robust and adaptively GPS signals tracking loop. These techniques have important value for the application of GSR in complicated electromagnetic environment.Firstly, the design and application of multiple-measurements GSR based on normal signals are analyzed. Through presenting the characteristic of GPS signals'construction and the mechanism of Software Radios, the parallel code phase acquisition algorithm and the classic second order based Phase Lock Loop (PLL) and Delay Lock Loop (DLL) are described and analyzed. Then this dissertation proposed the multiple-measurems algorithms. Moreover, experiments demonstrate its validity in acquisition and tracking GPS signals.For weak signals acquisition, two methods are proposed. Without external assistance, an improved coherent/non-coherent block acquisition algorithm is presented. It analyzes the Doppler search grids configuration at different coherent data length, also a practical compensation algorithm for the square loss is proposed. However, there still exist some conflict problems in coherent/non-coherent integration. Facing this challenge, under assisted GPS system, the navigation bit synchronization based long sequence coherent weak signal acquisition method is proposed. It narrows down the Doppler search region using base station's information, and eliminates the bit transfer by synchronizing navigation data bit. This new method brings more Signal to Noise Ratio and less time consuming.Strictly analysis, there are many colored jitter noises exist in traditional second order PLL tracking procedure except the white noise. The simulation experiments show that the colored jitter noise can hardly be reduced by simply narrow down the bandwidth. The estimated precision of PLL is seriously restricted, so high performance GPS signal tracking algorithm need to be developed.Kalman filter based GPS signals PLL and DLL in GSR is studied. A linearisation method for Arctan carrier phase discriminator is presented. Through decorrelating the system noise, a Kalman filter based PLL/DLL is built up to optimal estimate the signal tracking parameters. In order to refine the precision of the closed loop system, an advanced PLL state feed back control law and a parallel correlation feed back loop constructure are proposed. Two sets of real sampled data are adopted to demonstrate this new GPS signal tracking loop, the result shows that Kalman filter based PLL+DLL provide higher precision in carrier phase estimation, Doppler frequency estimation and CA code phase estimation.Precisely and continuously tracking GPS signals under complicated electromagnetic environment is important for many applications. At the end of this dissertation, an adaptive and robust Kalman filter based PLL+DLL solution is studied. This filter is able to improve phase tracking precision through automatically adjusting the mis-modelling of the dynamic process and reducing the effects of larger measurement errors under low signal-to-ratio environments. Compared with the traditional second order and standard Kalman filter based tracking loops, the robustness and efficiency of this new filter is demonstrated by real data testing experiments.The research work in this dissertation lays a firm foundation in the further development of GSR,and will improve the application of GSR more effectively in engineering under various environment.