Interception And Identification Of Non-cooperative Underwater Acoustic Communication Signals

The interception and identification of non-cooperative underwater acoustic communication signal is one of the research hotspots in the field of underwater acoustic communication signal processing.In this paper,the techniques of signal interception,feature and parameter extraction and communication signal pattern determination are studied under the condition of no prior information,i.e.non-cooperative condition.The main research contents and contributions of this paper are as follows:(1)To solve the problem that the conventional energy detection method does not make full use of the characteristics of underwater acoustic communication signal and leads to low signal processing gain,this paper introduces a signal detection method based on the cyclostationary characteristics of underwater acoustic communication signal and the good anti-noise performance of cyclostationary analysis.Simultaneously based on the traditional cyclic spectrum,an optimal construction method of single cycle frequency detection statistics is proposed to improve the performance of interception detection under non-cooperative conditions.The analysis shows that,compared with the conventional energy detection methods,the proposed method is easier to set the detection threshold,and has stronger anti-noise ability and robustness.(2)In view of the difference and aliasing of multi-type non-cooperative underwater acoustic communication signals,a signal identification method based on multi-feature layered processing is firstly proposed in this paper.The method uses the autocorrelation and spectral characteristics of the signals as well as the hierarchical classifier architecture to distinguish the signals layer by layer until all the signals are identified.The analysis shows that this method can effectively identify common underwater acoustic communication signals such as BPSK(Binary Phase Shift Keying)signal,QPSK(Quadrature Phase Shift Keying)signal,2FSK(Binary Frequency Shift Keying)signal,4FSK(Quaternary Frequency Shift Keying)signal,MSK(Minimum Shift Keying)signal,DSSS(Direct Sequence Spread Spectrum Signal)signal,and OFDM(Orthogonal Frequency Division Multiplexing)signal.In order to further improve the discriminability of features,a support vector machine classification method based on time-frequency domain features and cyclic spectral entropy features is also proposed in this paper,which further improves the identification performance of phase-shift keying signals and frequency-shift keying signals in low SNR environments.(3)In this paper,a variety of non-cooperative underwater acoustic communication signal parameter estimation methods are studied and proposed,which can realize the effective estimation of signal parameters such as symbol period,the number of subcarriers,pseudo-code period,and pseudo-code sequence.In this paper,based on the characteristics of the cyclic spectrum of underwater acoustic communication signals,a new method of signal period estimation based on carrier frequency estimation and cyclic spectrum cross-section is proposed to achieve a robust estimation of the signal period under low SNR environments.Then,according to the mechanism of the OFDM signal itself,an estimation method of OFDM signal subcarrier number based on the pre-estimation of prefixed symbol length and symbol period is proposed to solve the problem that the number of OFDM signals is difficult to estimate.Aiming at the problem that the traditional secondary spectrum is easy to produce false spectrum peak,a pseudo code period estimation method based on the secondary spectrum correction is proposed for direct sequence spread spectrum signal,which eliminates the influence of false secondary spectrum peak on pseudo-code period estimation.In order to solve the problem of the slow convergence rate of traditional fixed step size LEAP neural network in estimating pseudo-code sequences of direct sequence spread spectrum signals,a variable step-size method is proposed to improve the convergence speed of the LEAP neural network.The analysis shows that the proposed and improved communication signal parameter estimation method has good performance.(4)The intercepting and identifying system of non-cooperative underwater acoustic communication signals is designed and constructed,the overall architecture design and processing flow of the system are given,and the performance of the system is analyzed in detail by using the actual experimental data.