Signal-level Simulation Of Passive Sonar Direction Finding

With the perimeter that deployment around our country is increasing, the underwater security situation has become increasingly serious. To promote the performance of detecting underwater targets, it studied signal-level simulation of passive sonar direction finding to deepen the research of the process of passive sonar direction finding based on the concealment of passive sonar direction finding.It simulated the source of signal of passive sonar. The marine environment noise was obtained by Gauss white noise, and the radiated noise that can be trusted was simulated by modulation spectrum noise which contained continuous spectrum noise and line spectrum noise. To deal with the signal, it established the model of beamforming based on the directional model of array. To obtain the direction information, it simulated uniform weighted, Chebyshev weighted and adaptive beamforming respectively, the simulation results show that uniform weighted beamforming has a higher precision, and the adaptive beamforming is good of anti-interference ability. What’s more, it introduced the particle swarm optimization(PSO) algorithm into passive sonar beamforming for the first time to promote the performance of direction finding. The simulation results of the standard PSO beam pattern optimization show that it has high accuracy and can suppress interference particular orientation. However, it also has defects of slow convergence and easy precocious. To overcome the defects, it put forward the T-PSO algorithm which combined the improved Taylor method and standard PSO algorithm innovatively.The result of signal-level simulation of passive sonar direction finding shows that T-PSO beam pattern optimization method can apply to both symmetric beam pattern optimization and asymmetric beam pattern optimization. Moreover, it has high precision measurement, low false alarm rate, and good anti-jamming capability. But, the convergence stability of T-PSO beam pattern optimization is to be improved.