基于恒定束宽波形保真及干扰抑制的水下目标识别方法
Underwater Target Recognition Based on Constant-Beamwidth Waveform Fidelity and Interference-Suppression

水下目标辐射噪声通过阵列波束形成后易造成波形失真,且受干扰的影响,严重降低了目标的识别效果。为此提出了恒定束宽波形保真及干扰抑制的水下目标识别方法。该方法通过二阶锥规划方法设计恒定束宽波束形成器,用以实现波形保真并滤除干扰源的影响,结合人耳听觉感知机理和经典的DEMON分析,提取出有效反映目标类别属性的联合特征矢量。针对直线阵,利用单水听器获得的真实目标辐射噪声,仿真了干扰条件下阵列输出信号并开展识别实验,结果表明所提出的水下目标识别方法可以有效实现目标的准确分类,且对目标形态有着良好的宽容性。
In the presence of interference, the recognition performance of underwater target radiated noise severely degrades due to waveform distortion after array beanforming. Based on constant-beamwidth and interference-suppression, a new method of underwater target recognition algorithm is proposed to improve the fidelity of waveform in this paper. In this method, the second-order cone programming (SOCP) is exploited to deal with signal distortion of broadband constant-beamwidth beamforming. Combining SOCP with auditory filtering techniques, we extract intrinsic features of the targets to effectively distinguish them from each other. Based on array signal simulated from the ship-radiated noise collected by a hydrophone, the results and their analysis demonstrate preliminarily the effectiveness of the proposed algorithm in target classification