基于模极大值小波域的去噪算法研究

根据信号与噪声在小波变换下的不同特性，提出了基于模极大值小波域的去噪算法。该算法先用Adhoc算法求出信号的模极大值，再根据模极大值小波域的定义求出信号的模极大值小波域，从而得到信号的小波系数，然后逆变换得到信号。实例分析表明：该算法能有效消除噪声，与交替投影模极大值算法相比，该算法在原理上更简单，程序实现更容易，去噪速度更快，能满足在线监测的要求。

Denoising Algorithm Based on Modulus Maximum Wavelet Field

Before the signal is detected, the noise included in signal should be wiped off. A novel denoising algorithm based on modulus maximum wavelet field is presented. The characters of noise and signal on wavelet transform are discussed. The processes of the algorithm are as follows: firstly the modulus maximum is obtained according to Adhoc algorithm; secondly the modulus maximum wavelet field is computed according to its definition to attain wavelet coefficient; thirdly the singularity is reconstructed through inverse wavelet transform. Compared with others, the algorithm has some advantages, such as constructing efficiently, wiping off the noise effectively, programming easily. Examples prove that the algorithm has better denosing performance, and can meet the demands of online detection. The singularity of signal is represented by Lipschitz index.