为了提高被动声目标识别率 ,该文研究了隐马尔可夫模型 (HMM)在被动声信号分类中的应用问题。然后 ,又提出了 2种混合分类器 :特征矢量混合的HMM分类器和 HMM/ ML PNN (多层感知机神经网络 )混合模型分类器。结果表明 ,这 2种混合分类器在性能上都优于单个特定的分类器 ,它们在被动声信号分类中具有良好的应用前景 In order to improve the recognition rate of passive acoustic target, this paper studies the application of Hidden Markov Model (HMM) in passive acoustic signal classification. Then, two hybrid classifiers are proposed: feature vector mixed HMM classifier and HMM / ML PNN (Multilayer Perceptron Neural Network) mixed model classifier. The results show that these two kinds of hybrid classifiers are superior to a single classifier in performance, which have good application prospects in the classification of passive acoustic signals