基于局部投影算法的舰船信号降噪处理研究

Noise Reduction of Ship Signals Based on the Local Projective Algorithm

利用基于非线性动力学系统的局部投影降噪算法研究了舰船辐射噪声的非线性降噪问题。分析了局部投影降噪算法邻域半径参数与降噪效果的关系,研究了利用噪声强度估计邻域半径参数的方法。通过对含噪Lorenz模型和实际舰船信号进行降噪处理,得到了较为满意的结果。

Nonlinear noise contained in underwater acoustic signals greatly affects the correct extraction of characteristic parameters, such as Lyapunove exponent and fractal dimension. In this paper, we focus on the application of well-known local projective noise reduction method to underwater acoustic signals. We discuss in much detail what parameters to select and how they are selected in using the local projective noise reduction method, especially for the parameter of local neighborhood size ε, which affects greatly the result of noise reduction. In order to improve both the efficiency and the convergence of the local projective noise reduction method, we propose a new method to estimate local neighborhood size ε. The new method is based on the estimation of the noise level of underwater acoustic signal which is easier to calculate than other algorithms. As an example, we use Logistic map and three types of real ship radiated noise as samples to evaluate the efficiency of the algorithms. The results show that the new algorithm is superior to the old one. Besides, the new algorithm is relatively simple and converges quickly.