Research On Passive Detection Technology Of Underwater Target Tone Based On Unmanned Underwater Vehicle

The tonal signals radiated from target are of great significance to detect the low-noise,quiet underwater targets.In this paper,the tonal signals radiated from underwater unknown target are autonomously detected by using the sensor array which is mounted on an underwater unmanned vehicle(abbreviated as UUV).The difficulty of detecting the target tones is increased due to the autonomy and flexibility of the UUV platform.Aiming at the above difficulties,the in-depth analysis and research are conducted on the broadband beamforming technology and the time-domain tone detection algorithm.Then,the autonomous detection methods of target tone based on the UUV platform are proposed,which lays a foundation for the passive detection of underwater target based on the UUV platform.Considering the broadband beamforming in the absence of direction-of-arrival(DOA)information of a tonal target,a self-steering broadband beamforming method is proposed.Firstly,we introduce the self-adjusting characteristics of adaptive line enhancer to a time domain broadband beamformer.Then the beam self-steering is achieved.Secondly,the selfsteering beamforming is performed on decimated sub-band signals in order to alleviate the contradiction between the convergence speed of the self-steering beamformer and the steadystate weight noise.A space-time projection matrix is used to suppress the strong tonal interferences.The proposed method can adaptively steer the main beam towards the DOA of the tonal target,and the complexity of the autonomous detection of tone is reduced.The proposed method is suitable for the tonal detection based on UUV.Simulation experiments and sea trials further verify the effectiveness of the proposed beamformer.Discrete Fourier Transform(abbreviated as DFT)is a classic tonal detection method.The target tonal frequency is unknown and there is usually Doppler shift.Thus,it is inevitable that the DFT frequency domain discrete point deviates from the real tonal frequency,which causes the energy loss of tone,namely the roughness loss.Aiming at the scalloping loss problem caused by DFT discrete frequency deviation from the tonal frequency,a DFT-based tonal detector that is robust against the scalloping loss is proposed.Firstly,a hypothesis testing problem is built utilising the complex sequences on adjacent DFT frequency bins.Then,a generalized likelihood ratio detector is deduced.The proposed detector can gather the leaked energy of signal on the adjacent frequency bins.The influence of scalloping loss on the performance of the DFT-based detector is reduced.The proposed method is suitable for the autonomous detection of the tonal signals based on UUV.The effectiveness of the proposed algorithm is verified by simulation experiments and lake experiments.The ambient noise as well as the noise radiated from UUV platform is usually color noise.The color noise background fluctuations are not conducive to the autonomous extraction of target tones.In addition,UUV or target maneuver causes the Doppler shift and the rapid change in the direction of arrival of target,which limits the effective integral length of the system.The shortening of integration time reduces the physical resolution of the frequency.Aiming at the issue that tone is autonomously extracted at high resolution under color noise background,an autonomous extraction tone method based on the sparse reconstruction under color noise background is proposed.In this method,a sparse reconstruction method is used for tone extraction at high resolution.The subband processing is used to solve the problem that the color noise background fluctuations impair the sparsity of tones in frequency domain.In order to extract the tones from the sparse solution,an adaptive threshold is set using an estimate of the background noise variance.The proposed method achieves autonomous extraction of tone in the background of color noise while obtaining higher frequency resolution.The effectiveness of the proposed method is verified by theoretical analysis,simulation experiments and lake experiments.