Study On Ship Line Spectrum Adaptive Enhancement

At present,the passive sonar detection and recognition of underwater targets are the key research topics in the field of underwater sound.The ship,as a kind of underwater moving targets,includes the radiation spectrum and the modulation spectrum,which involving a large amount of ship information,information full of high strength and good stability.Therefore,the spectral characteristics of the radiation noise in the ship can be extracted as the representation of the underwater target characteristics,and the detailed and accurate spectral analysis results can provide the basis for the target recognition.In complex,changing marine environment,due to the interference from channel and environmental,along with low signal noise,results of spectrum analysis are not ideal.In view of the above problems,this paper studies the problem of ship line spectrum enhancement to reduce the interference from complex environmental factors,laying the foundation for extracting the stable spectral characteristics of ship radiated noise.This paper combined the adaptive spectral algorithm and the high-order cumulants theory to realize the spectral component of the ship 's radiated noise and the enhancement of the helical modulation spectrum component,reaching the goal of improving the ratio of signal-to-noise.In this paper,we derive the analytical solution of the convergence equation and steady-state error of the minimum mean kurtosis algorithm with high-order theory under the ideal condition.In comparison of numerical simulation and conventional adaptive enhancement method(LMS algorithm and LMF algorithm),it is proved that this algorithm is more inclusive under complex conditions,and the adaptive line spectrum based on the proposed algorithm is used to achieve the goal of restraining abilities of noise.Aiming at the problem that the convergence rate and steady-state error of the minimum mean kurtosis algorithm are constrained by the step size,this paper proposes a new minimum step-size minimum peak kurtosis algorithm.Through the theoretical analysis and the numerical simulation,this paper verifies that the proposed algorithm is faster than the conventional minimum mean kurtosis algorithm in convergence rate,and higher in convergence precision,based on weighing the convergence rate and steady-state error.In order to obtain a stronger ability to suppress noise,this paper uses the improved algorithm and the conventional LMS algorithm to construct the second order cascade filter,making further improvement in enhancing the output ratio of signal-to-noise gain of adaptive line spectrum.Finally,the algorithm proposed in this paper is used to analyze the ship's line spectrum component and the propeller modulation spectrum component based on the experimental data.The experimental results show that the ratio of signal-to-noise is enhanced and the feasibility of the algorithm is verified,realizing to lay the foundation of the underwater target passive sonar detection and identification.