Research On Key Techniques Of DSSS Signal Reconnaissance

Spread spectrum communication has many advantages,such as widening frequency band,low detection probability and strong anti-interference ability,so spread spectrum communication has been widely used in military and civilian applications.Among them,direct sequence spread spectrum communication is a common spread spectrum communication mode in spread spectrum communication system.For DSSS signal reconnaissance,first to detect the existence of DSSS signal,and then estimate the important parameters of DSSS signal frequency,PN code period,code rate and code sequence,so as to get the original information code or a variety of interference.In the signal detection and parameter estimation of DSSS communication,this thesis mainly studies the square frequency method,delay multiplication detection,cyclic spectrum slice method,and two order moments and two power spectrum method to detect the DSSS signal and estimate the presence of DSSS signal carrier frequency(square frequency method,cyclic spectrum slice method),PN code rate(delay multiplication algorithm,cyclic spectrum slice method),pseudo code period(two order moment,two power spectrum method),according to the different algorithms,in the algorithm to verify validity based on parameter estimation and detection,and then the performance analysis.In the estimation of PN sequence,two algorithms,maximum likelihood estimation and wavelet neural network,are adopted.In the case of maximum likelihood estimation,the computational procedure of maximum likelihood estimation is greatly simplified by using the idea of iteration.In the estimation of PN sequence using neural network,to improve the traditional neural network using wavelet neural network,strengthen the anti noise performance of the neural network,so that in the condition of low SNR,the estimation performance of PN code has been improved.Finally,for periodic long code DSSS signals,we use a piecewise maximum likelihood estimation algorithm to estimate pseudo code sequences,and verify the performance of the algorithm.