Research On Detection Methods Of Non-cooperative Underwater Acoustic DSSS Signals

With the development of underwater acoustic(UWA)communication countermeasure technology,direct sequence spread spectrum(DSSS)communication systems with good concealment and low probability of interception are becoming more and more widely used in UWA communication.In order to realize the detection and effective jamming of the non-cooperative UWA DSSS signal,it is a prerequisite and priority for the UWA communication countermeasure to accurately detect the low SNR DSSS signal.This thesis focuses on non-cooperative(i.e.,without prior knowledge)UWA DSSS signal,to study a variety of detection methods under low SNR conditions.Mainly studied three detection algorithms of power spectrum detection,cepstrum detection and autocorrelation detection for DSSS signals,and deduces some theoretical derivation and simulation.The Bartlett method and the Welch method are studied on the basis of the periodic graph method.Then the indirect method and the B & T method corresponding to the periodic graph method are studied.In addition,the parameter model method of modern power spectrum estimation is introduced.Cepstrum detection and autocorrelation detection are based on traditional cepstrum and autocorrelation methods,for cepstrum and autocorrelation methods detecting spectrum line are susceptible,overlapping sub-windowed approach and averaging method,and then detecting a certain area of the spectral line,which improves the detection performance and SNR tolerance of the algorithms.According to the difference of the detection results of the three methods,the corresponding detection and judgment methods are proposed respectively,and the false alarm probability of blind detection is reduced to a certain extent.A variety of simulation experiments were carried out to compare and analyze the simulation results of different detection methods under different SNR or different detection parameters: Increasing the acquisition signal window length can significantly improve the detection probability of improved cepstrum and improved autocorrelation and reduce the false alarm probability of Welch method.Increasing the length of the DSSS signal chip,the false alarm probability is not changed,the detection probability of the various methods has been improved,and the improvement degree of the improved autocorrelation method is larger.Increasing the length of M sequence can improve the detection probability to a certain extent and can effectively reduce the false alarm probability variance.In addition,analyzed the detection results of DSSS signals in the ocean and low SNR measured DSSS signals.Finally,a real-time software system for acquisition and detection of underwater DSSS signals is developed by using LabVIEW.A variety of detection and judgment methods are integrated,and the data acquisition mode is improved to enhance the blind detection efficiency.