随着水下目标降噪技术的进展,利用回声信号来探测和识别水下目标的越来越重要,水下目标识别是水声领域中存在的难题之一.以大量的水下目标回声信号资料为依托,借助Walsh变换的分析法,研究了脉冲声的回声识别问题,给出了回声信号的时频特性及其相应的特征向量,提取了43维的Walsh谱特征.结合ART神经网络,对一定范围对真假两类目标进行分类,可取得80%以上的正确识别率.该方法具有较好的实用性,是一条解决实际回声信号识别的有效途径. With the progress of underwater target noise reduction technology, it is more and more important to use echo signals to detect and recognize underwater targets. Underwater target recognition is one of the difficult problems in underwater acoustic field. With a large number of underwater target echo signals Data, the Walsh transform analysis method is used to study the echo recognition of impulsive sound, the time-frequency characteristics of echoes and their corresponding eigenvectors are given, and the 43-dimensional Walsh spectral features are extracted. Combined with ART neural network, To classify the true and false targets in a certain range can get more than 80% correct recognition rate.The method has good practicability and is an effective way to solve the actual echo signal recognition.