一种新的四元阵列融合声源识别方法

针对声源识别中观测模型线性化误差,信号特征参数提取依赖于经验分析阈值判断而造成信息丢失的问题,本文从信号观测模型、预处理、特征提取与分类识别、半实物仿真试验等方面,提出了一种新的四元阵列融合声源识别方法。首先在系统坐标系下建立了四元阵列有色噪声环境下的观测模型；其次基于EMD理论,给出了四元阵列EMD融合算法,有效抑制了高频信号的干扰；再次基于MFCC-DTW方法,设计了阵列信号特征提取与分类识别算法；最后通过半实物仿真试验,并与相关研究基础对比,分别验证了本文提出的EMD融合算法及阵列信号特征提取与分类识别算法的有效性。

A Innovative Four-Sensor Acoustic Array fusion Identification Algorithm for Passive Acoustic Target

_{Aiming at the problem of linearization error of the observation model in sound source recognition, the extraction of signal characteristic parameters depends on the threshold value of empirical analysis to cause information loss, a innovative four-sensor acoustic array fusion identification algorithm is proposed from the aspects of signal observation model, preprocessing, feature extraction and classification recognition, and hardware-in-the-loop simulation experiments.Firstly, the observation model of the four-sensor acoustic array in the colored noise environment is established in the system coordinate system; secondly, based on the EMD theory, a EMD fusion algorithm is given to effectively suppress the interference of high frequency signals. And then the algorithm of array signal feature extraction and classification recognition is designed. Finally, the effectiveness of the proposed EMD fusion algorithm and array signal feature extraction and classification recognition algorithms are verified through semi-physical simulation experiments and comparison with related research foundations.}