目标噪声特征提取和目标分类器设计是被动声纳目标识别系统的关键技术。本文针对被动声纳目标识别,提出了一种新的连续谱特征提取方法。此外,为了训练神经网络目标分类器,将遗传算法和BP算法相结合,提出了一种新的自适应遗传BP算法。最后,对海上实录的三类目标噪声进行了分类识别。实验结果表明,设计的被动声纳目标识别系统具有很好的分类效果。 Target noise feature extraction and target classifier design are the key technologies of passive sonar target recognition system. Aiming at the passive sonar target recognition, this paper proposes a new method to extract the continuous spectrum feature. In addition, in order to train the neural network target classifier, a new adaptive genetic BP algorithm is proposed by combining genetic algorithm and BP algorithm. Finally, three types of target noise at sea were classified and identified. The experimental results show that the designed passive sonar target recognition system has a good classification effect.