摘要
本文阐述了一种非母板匹配的水中目标辐射噪声分类识别的思想、技术路线及算法的计算机实现关键技术。该算法基于提取水中目标辐射噪声信号功率谱中的最强谱带位置、线谱位置、线谱根数、线谱平均间隔等特征参量,通过特定处理后,直接给出识别结果,不需要附带庞大数据库。经试验验证,该技术突破了传统的母板匹配目标识别的算法思想,能够适用于复杂海况,做到宽容性匹配。识别效率高,识别率达到试验预期要求。
In this paper, a new idea about the identifying of the noise of underwater target is described. It＇s based on the pick-up of the characteristic parameters （such as the position of the frequency band. the line spectrum and others） of the noise signal radiated from an underwater target. The algorithm is non-motherboard-matching, which is completely different from the earlier ones in this area. It gives the result directly without the need of searching the motherboard database. As a result, this arithmetic matches most situations of the ocean. The efficiency of the identification is in a high level. And the identifying rate accords with our anticipation in an ocean.
出处
《应用声学》
CSCD
北大核心
2006年第5期265-269,共5页
Journal of Applied Acoustics
关键词
非母板匹配
特征参量
宽容性
识别效率
识别率
Non-motherboard-matching, Characteristic parameters, Allowance, Efficiency of ientification, Identifying rate
