Detection And Parameters Estimation Of Direct Sequence Spread Spectrum Communication Signals In Non-cooperative Context

Direct Sequence Spread Spectrum (DSSS or DS) communication as the important part of Spread Spectrum communication has been widely used both in military and commercial telecommunication area, owing to working at low signal to noise ratio(SNR) , the ability of strong anti-jamming, lower probability of being intercepted and captured. Meanwhile, in the field of communication antagonism, the methods of detection and parameter estimation of spread spectrum signals in non-cooperation communication have been important research.The dissertation refers to theory and flow of the spread spectrum blind detection and processing system, focuses on the detection, estimation and PN sequence re-construction methods of DS signals in non-cooperative context. The dissertation is organized as follows:1,Basing on the background of research, the dissertation presents situation of domestic and foreign and the DS signals involved in this dissertation, analyses the spread spectrum blind detection and processing system in functional and signal context.2,The dissertation analyses the basic theory of time-domain correlation and cyclic spectrum which are used to calculate DS signals, simulation shows that time-domain correlation and cyclic spectrum algorithm have great identification. While, the dissertation uses time-domain correlation algorithm to estimate the PN sequence period, simulation shows the SNR range and performance of the algorithm, the dissertation uses cyclic spectrum algorithm to distill carrier frequency and rate of the PN codes, simulation shows the performance of the algorithm.3,Several usual blind code synchronization methods, including delay multiply, maximum norm and average noise reduction approaches are summed up and analyzed. On this basis, the blind synchronization algorithm of direct sequence spread spectrum signals based on search of autocorrelation pulse peak is proposed, to get blind code synchronization, simulations show that the algorithm has greater accuracy, needs fewer data, and is easy to realize but with strong anti-jamming capability.4,On the basic of code synchronization information, the dissertation comes to focus on the PN sequence blind estimation method of DS signals, include matrix eigenvalue decomposition and neural networks methods. First, the theory of matrix eigenvalue decomposition for DS signals is analyzed, simulations show that the algorithm can get PN sequence accurately; second, basing upon the thoery of the matrix eigenvalue decomposition, we introduce the neural networks(N.N.) approach to estimate the PN sequence, formula derivation certificates the principal component analysis is equal to matrix eigenvalue decomposition, in addition, the neural network PN sequence method based on the Hebbian learning rule is presented, simulations show the convergence performance , SNR range of algorithm and estimation of PN sequence. With the blind synchronization algorithm of direct sequence spread spectrum signals based on search of autocorrelation pulse peak, we can get that matrix eigenvalue decomposition and neural networks methods.can reconstruct PN sequence in lower SNR, and overcome the phase alternation by asynchronous.