When detecting an underwater target by using acoustic wake, the useless backscattering noise affects the useful target reflected signal and makes signal detection and processing very difficult. In terms of this problem, we propose a method in which a multiwavelet transformation algorithm is found to eliminate the useless backscattering noise. In this paper, according to the different characteristics of signal and white noise, a denoising algorithm based on the correlation of multiwavelet coefficients is presented for better denoising performance. First, the correlative coefficient of different blocks is calculated, and then it is compared to the normalized correlative coefficient and selected according to the rule. The reserved correlative coefficient is then processed with the neighboring coefficient method. Using the algorithm, the correlation of neighboring coefficients and different blocks is simultaneously taken into account. The denoising performance is better than the conventional algorithm, which only uses neighboring coefficients. The experimental results show that the interference of noise is reduced and the break point is more visible, which is of great practical significance to underwater target detection.