Research And Implementation Of Blind Separation Algorithm Based On PCMA Signals

As a communication satellite multiple access multiplexing technique, PCMA is gaining increasingly wide range of applications because of its high spectrum efficiency and Anti-interception performance. On the basis of in-depth study of particle filter algorithm, a new algorithm designed to reduce the amount of calculation in the conventional algorithm is proposed in this paper to achieve PCMA mixed signal blind separation. When the signals’ symbol rate is different in communication, it is difficult to separate the symbols. To solve this problem, this thesis presents an effective solution. Details are as follows:(1)Using the slow variation characteristics of the parameters of phase shift and time delay in satellite channel, and the frame form of satellite communication data, this thesis improves the traditional algorithm to reduce the amount of calculation. In the process of running the algorithm, the particles with smaller posterior probability were removed. By dynamically adjusting the number of particles paths, the new algorithm greatly reduces the amount of calculation. Assuming that the channel parameters are not changed in the time of a frame, the improved algorithm processes the observed data by frame and estimates the channel parameters with a small amount of observational data. After that, the improved algorithm separates symbols based on the maximum likelihood criterion because of the channel parameters were known. Different from the improved algorithm, the traditional algorithm estimates the channel parameters and symbols at the same time. Through MATLAB simulation, the results show that the improved algorithm can achieve fast convergence in the estimation of channel parameters, and can correctly separate the two symbols in the PCMA mixed signal.(2) When the mixed signals have different symbol rates, the delay of the two signals will slide. When the sliding time is more than a symbol, the corresponding symbol sequences will be changed. Thus it is unable to separate the symbols in the PCMA mixed signal. By modifying the mixed-signal baseband model and establishing a new state transition equation, this thesis improved the algorithm which updates particles and weights, achieving the separation of the symbols. Through MATLAB simulation, the results show that the improved algorithm can separate the two symbols in the PCMA mixed signal.(3) In this thesis, the C language program for the blind separation of PCMA mixed signal based on particle filter is prepared. This thesis enhanced the portability of the program by modular processing which lays a solid foundation for the DSP implementation.In this thesis, the traditional particle filter algorithm is improved, which greatly reduces the amount of calculation of the algorithm and realizes the blind separation of the PCMA mixed signal. Through amounts of MATLAB program simulation, it shows that the performance of the improved algorithm has realized the expected goal.