基于目标特征信息融合的潜艇目标识别系统

Submarine target recognition system based on characteristics information fusion

结合水下目标运动状态估计和目标声特征提取的关键技术,提出了基于目标运动状态特征和声特征信息融合的潜艇目标识别基本原理。首先,采用双通道Power-law水声瞬态信号检测方法和LMS自适应时延估计算法,建立了基于方位时延的目标运动状态跟踪的数学模型,并采用EKF算法实现了水下目标运动状态估计;其次,建立了提取潜艇典型不变声特征的原理模型;最后,提出基于多目标特征信息融合的结构模型,为潜艇目标识别系统的发展奠定了理论基础。

Combining with the key technologies of underwater target moving state estimation and target acoustic characteristics distillation, the elementary principle model of submarine target recognition system based on information fusion of acoustic characteristics and moving statement was put forward. Firstly, a method of detecting underwater acoustic transient signals based on power-law detection with dual-channel and the algorithm of LMS self-adaptive estimation was used. The mathematical model of target movement statement track based on orientation and time delay was built. And EKF algorithm was used to realize target movement statement track. Secondly, the principle model of extracting typical stable acoustic characteristics of submarine was also built. At last, the structure model of multiple information fusion was proposed, which is the theoretic basis of submarine target recognition system development.