基于线谱特征的水下目标识别仿真实验研究

水下目标识别技术是世界各国十分重视的研究课题之一，在国民经济尤其在军事领域中，具有重要的应用价值。船舶辐射噪声中有着很强的线谱成份，这些成份分类或识别往往是非常有价值的。因此，利用线谱特征对舰船目标进行识别是水声领域研究的重要内容。本文在研究舰船辐射噪声线谱提取逄法的基础上，对多种型号和多种工况的大量噪声样本，进行了线谱提取。针对舰船辐射噪声线谱分布的特殊性，提出了筛选法和模糊推理分类法。利用这些

A Simulation Experiment of Underwater Targets Identification Based on Linespectrum Features

Underwater targets identification, which possesses a crucial application value to state economy and defense, is a very important topic in advanced country. Ship radiated noises include a lot of linespectrums, which are valuable for targets identification, therefore it becomes an important approach in the studies of underwater acoustics. Based on the algorithm of extracting ship radiated-noise's linespectrums, we extract the linespectrums of thousands of stylebooks in many operating conditions. In accordance with the particularity of the distributions of ship radiated-noise's linespectrums, two new methods, which are filtration and fuzzy reasoning classification, are presented respectively. Experiment results show that both methods are feasible and effective.