摘要
研究如何在复杂噪声和严重混响的室内环境下,提高声源定位系统的性能.首先分析室内麦克风阵列的信号模型,确立一种合适的阵列拓扑结构;其次在复倒谱域对滤波器的各项参数进行仿真比较,得出滤波器参数的最佳值;然后提出一种广义互功率谱相位法,改进了峰值不明显等问题.在Matlab中对多种加权函数进行了仿真比较,进一步改善声源定位结果的准确性.最后对提出的算法进行验证,结果表明,当改进后算法的角度误差和距离误差分别为±10°和±15 cm时,定位成功率能达到80%左右.
The paper mainly researches how to improve sound source localization effects in indoor environments with complex noise and serious reverberation. Firstly,the signal model of the indoor microphone array is analyzed and a suitable array topology is established. Secondly,a lot of parameters of the filter in the cepstral domain are determined optimally by simulation. Then,the generalized cross-power spectrum phase is proposed to make the peak value more obvious. In addition,many weight functions are simulated and compared in Matlab to further improve the accuracy of sound source localization. Finally,the proposed algorithm is verified. The result shows that the localization success rate is up to 80% when the angle error and distance error of the improved algorithm are ± 10°and ± 15 cm,respectively.
作者
张传义
卢晓
ZHANG Chuan-yi;LU Xiao(School of Information Science ＆ Engineering,Northeastern University,Shenyang 110819,China.)
出处
《东北大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2018年第8期1075-1079,共5页
Journal of Northeastern University(Natural Science)
基金
国家自然科学基金资助项目(61473075)
关键词
声源定位
麦克风阵列
倒谱法
广义互功率谱相位法
sound source localization
microphone array
cepstrum
generalized cross-powerspectrum phase
