The Research On The DSSS/BPSK Signal Detection And Parameters Estimation In Non-cooperative Communication System

Direct sequence spread spectrum (DSSS) signal is widely used in non-cooperation communication domain such as communication reconnaissance and spectrum surveillance. DSSS signal is more difficult to intercept, detect and analyze, for its characteristics of low power spectrum, anti-interference, low probability of intercept (LPI) and low probability of detection (LPD). Therefore, it is a practical and challenging research subject of DSSS signal detection and blind parameters estimation, without a priori information.This dissertation mainly researches on DSSS/BPSK signal detection, recognition and parameters estimation, aims to complete the signal analysis work of communication reconnaissance and supplies technical parameters for communication countermeasure.The main research contents of this dissertation are as follows:(1) The carrier frequency estimation theory of carrier doubler method (CDM) and cyclic spectrum algorithm are analyzed. In order to obtain lower SNR, an improved algorithm of CDM: autocorrelation-CDM is proposed.(2) Comparative analysis of autocorrelation accumulation, cepstrum, power spectrum reprocessing and wavelet transform is taken. The algorithms of cepstrum and wavelet transform are improved by combining autocorrelation accumulation with them, what has improved the signal to noise ratio tolerance and accuracy of estimation. An adaptive method of detection threshold is given.(3) Algorithms of PN sequence reconstruction including subspace eigenvalues decomposition and neural network are studied. In view of the defect of present NN algorithm in BER and running time, a new NN model APEX is adopted, which has great improvement in estimation accuracy and algorithm running time.(4) An algorithm of autocorrelation-cyclic spectrum, based on the cyclo-stationarity characteristic of DSSS and the periodicity and autocorrelation characteristics of spread sequence, is proposed to solve the problem of detection and recognition of DSSS/BPSK and narrowband BPSK in lower SNR condition. The algorithm can detect DSSS/BPSK signal in lower SNR, and gain higher accuracy of signal recognition and parameters estimation.(5) The system of DSSS/BPSK signal detection and parameters estimation is designed, the simulation system in Simulink and the Matlab GUI are also given. Finally, mass simulation experiments and analysis for factors that influence the estimated performance are carried out, and comparative analysis of different algorithms is taken, which has practicability and reference value.