A Sonar Blind Beamforming Method Using a Neural Network

摘要：

In this paper a blind beamforming method based on Doppler information and a neural network is presented according to the characteristic of underwater acoustic environment and signal. Firstly, Doppler information of underwater acoustic signal is used to estimate weights of beamformer without array manifold. Then, we propose a blind beamtorming algorithm using a neural network with a high computational efficiency. The neural network with a parallel structure can process the data more rapidly. It can reach steady state within a circuit time constant at each snapshot, and network convergence time is unaffected by the number of sensors. Therefore, it is especially useful to arrays with many sensors.

作者: Li Hongsheng Zhao Junwei Wang Feng Guo Yecai Bian HongYu Sang EnFang Piao ShengChun

作者单位: Northwestern Polytechnical University, Xian, 710072, P. R C. Harbin Engineering University, HaErBin, 150001, P. R C.

会议类型: 国际会议

会议名称: 2007年国际声学工程与技术学术会议

会议地点: 北京

会议语种:英文

页码: 165-169

在线出版日期: 2007-10-01（万方平台首次上网日期，不代表论文的发表时间）