Research On MPSK Signal Blind Demodulation For Non-cooperative Digital Communication System

Non-cooperative digital communication plays an important role in the national economy and defense security,and the research about non-cooperative communication is still in the preliminary stage and involves a series of technical challenges.In noncooperative digital communication system,the receiver almost does not know communication information about the received signal.So,the receiver is facing great difficulties in demodulating.Blind demodulation is the core technology of noncooperative communication systems and research on blind demodulation technology in non-cooperative communication is challenging and necessary.This dissertation is based on non-cooperative communications and we assume the modulation mode of the received signal is MPSK(M is known in priori)in advance,this dissertation deeply investigates the blind demodulation techniques,including the parameter blind estimation,the symbol timing synchronization and the channel blind equalization.Firstly,this dissertation mainly studies the parameter blind estimation in the blind demodulation.Specifically,carrier frequency and symbol rate are extremely important parameters in demodulation,which have a direct effect on the performance of the demodulation.In order to obtain these two parameters accurately,a carrier frequency estimation algorithm based on chaotic Duffing oscillator and a symbol rate estimation method based on second-order cyclic spectrum are proposed respectively.Compared with the existing high-order cyclic accumulation frequency estimation and wavelet transform symbol rate estimation algorithm,the proposed parameter estimation algorithms have better estimation effect in low SNR.Then,we mainly investigate the symbol-timing synchronization to overcome desynchronization due to estimation bias.An improved Gardner timing synchronization algorithm is studied in this dissertation.To be specific,this algorithm improved the synchronization loop interpolation filter,which is based on base polynomial interpolation by optimizing loop filter coefficients.Then we can get higher interpolation accuracy.Simulation results show that the synchronization algorithm has higher synchronization accuracy and faster synchronization speed.At last,this paper studies the blind equalization technique,which is an important approch to inhibit inter-symbol interference and multipath fading.According to the defect of CMA blind equalization algorithm,a algorithm based on probability density function fiting(PDFF)is investigated in this part,and gives a new cost function and tap coefficient iteration principle by this algorithm.With the help of this method,simulation results indicate that the PDFF blind equalization algorithm has faster convergence speed and smaller convergence errors than CMA blind equalization algorithm.