The detection signal-to-noise ratio determines the underwater acoustic reconnaissance ability of AUV in complex marine environments. In this paper, the singular value decomposition method and adaptive noise cancellation method are used to improve the signal-to-noise ratio of dual channel sonar signal. The experimental data processing results show that the singular value decomposition method has a stronger noise reduction ability, but the noise reduction effect is greatly affected by the singular value threshold, and the performance of the algorithm is unstable; Although the noise reduction ability of adaptive noise cancellation method is limited, it has strong environmental adaptability and more stable algorithm performance.

Based on the standard large eddy simulation and k-epsilon model, the steady flow field and navigation flow noise of the underwater vehicle shell are simulated. In this paper, the underwater vehicle model is established by referring to classic SUBOFF model. The standard k-epsilon method is used to study the pressure distribution characteristic, velocity distribution characteristic, and velocity vector distribution characteristic on and around the shell of the underwater vehicle at different speeds. The results indicate that the pressure increases with the increase of the speed and in the bow of the shell is the largest. A large eddy simulation was used to analyze the noise characteristics of sailing flow at different speeds. The results show that the sound pressure level of navigation flow noise is higher in the low-frequency band, and will increase with the increase in speed.

Based on the theory of split beamforming algorithm and the experimental data collected by circular array, this paper analyzes the characteristics of DOA and target size estimation. In this paper, the signal receiving model of planar circular array and the split beamforming algorithm are studied. Then, based on the data collected from the lake experiment, the DOA estimation results of conventional beamforming algorithm, multiple signal classification algorithm and split beamforming algorithm are analyzed, and the target scale estimation results of split beamforming algorithm is analyzed. The experimental data analysis shows that the split beamforming algorithm has good direction finding stability, and the target scale estimation error based on the split beamforming algorithm is small.

Considering the limitation of FFT frequency estimation method in resolution, this paper briefly introduces the complex modulation ZOOM-FFT and Chirp-Z transform spectrum zooming methods. The simulation results show that the ZOOM-FFT method has better anti spectrum leakage and interference effect. Furthermore, the two methods are applied to the frequency estimation of underwater high-speed maneuvering target, and it is found that Chirp-Z transform method can obtain smoother and stable frequency history, and has more flexible frequency resolution.