摘要
主要采用TDOA算法,重点研究了基于麦克风阵列的声源定位技术。在对广义互相关时延估计算法分析的基础上,通过引入基于C_0复杂度的端点检测技术、线性预测和IIR平滑滤波,设计了基于线性预测残差的广义互相关算法。同时给出了平面四元十字麦克风阵列的设计及定位原理,完成了基于DM368的嵌入式语音定位系统设计。测试结果表明,所设计的系统有效改进了时延估计的准确性,提升了噪声环境下的定位性能。
This work focuses the application of TDOA algorithm on microphone array based sound source localization.After the analysis of generalized cross-correlation phase transform(GCC-PHAT)algorithm,the GCC-PHAT algorithm based on LP residue is proposed,which adopts C0 complexity based endpoint detection technique,linear prediction(LP)and IIR smoothing filtering.A cross-array,with four microphones symmetrically located in the same plane is designed,also the source-locating equations are deduced using the estimated delay information.An embedded sound source localization system based on DM368 is designed.The test results show that the proposed system can improve the accuracy of delay estimation and enhance the performance of localization under noisy environments.
作者
李瑶天
汤剑
张霞
刘丽娜
胡剑凌
LI Yaotian;TANG Jian;ZHANG Xia;LIU Lina;HU Jianling(School of Electronic and Information Engineering,Soochow University,Suzhou Jiangsu 215006,China)
出处
《电子器件》
CAS
北大核心
2018年第6期1612-1617,共6页
Chinese Journal of Electron Devices
基金
国家自然科学基金项目(61271359)
关键词
麦克风阵列
声源定位
TDOA算法
广义互相关算法
线性预测
microphone array
sound source localization
TDOA algorithm
generalized cross-correlation algorithm
linear prediction
