基于粒子群神经网络的声目标识别系统设计研究

Design Research of an Acoustic Target Recognition System Based on the Particle Swarm Neural Network

为了提高声目标的自动识别率,开展了基于粒子群神经网络算法的声目标识别系统设计研究。首先给出了试验中粒子群优化算法的初始参数及算法流程;其次,分别设计了包含电源硬件、信号调理、铁电存储器、触发以及通信接口等电路的声目标识别硬件电路;最后,通过声信号采集试验及系统样机的制定,开展了目标识别试验研究,试验结果验证了声目标识别系统的有效性及稳定性。

In order to raise automatic recognition rate for acoustic targets,design research is carried out on the acoustic target recognition systems based on the particle swarm neural network algorithm. First,this paper gives initial parameters and algorithm procedure for the experiment on the particle swarm optimization algorithm. Second,it designs hardware circuits for acoustic target recognition,including circuits of power hardware,signal conditioning,FRAM,trigger and communication interface. Finally,through the experiment on acoustic target acquisition and making of the system prototype,we have made experimental research on target recognition. The experimental results validate the effectiveness and stability of the acoustic target recognition system.