基于TDOA算法的四麦克风阵列三维声源定位研究

发布时间：2018-04-27 16:18

本文选题：声源定位 + 麦克风阵列　；参考：《兰州理工大学》2014年硕士论文

【摘要】：智能服务机器人是在非结构环境下为人类提供必要服务的多种高技术集成的智能化装备,具有多种功能,其中听觉就是重要标志之一,是机器与人以及环境之间重要的交互手段。在现实环境中,听觉可以实现360°的“听觉场景”,它可以搜索到不在可视范围内的声源目标。也就是说,听觉能够搜索定位出由于物体遮挡而产生模糊的目标,或者位于转角处的声源目标。虽然,相较于视觉,对它的研究依然处在前期阶段。但是,近些年来对听觉的研究已经成为一个重要课题,基于听觉的声源定位更是一个新的研究热点。基于听觉的声源定位,即由麦克风阵列拾取语音信号,继而通过对信号的处理,进一步确定声源的位置,这在语音信号处理领域已是一个研究热点,具有重要应用前景和实际意义。本论文学习研究基于麦克风阵列的声源定位技术的原理及特点,主要进行了以几方面的下工作：首先,对采用麦克风阵列进行声源定位的研究现状以及定位原理进行了总体概述,论述了基于麦克风阵列声源定位的三大技术,并总结了在定位过程中所遇到的难点,主要介绍研究了基于到达时间延时的定位技术。其次,基于到达时间延时的定位算法很多,本文对其中的广义互相关算法进行了研究,针对定位过程中存在的噪声影响,提出了一种改进的互功率谱相位加权算法。该算法运算量少,经实验证明,在强噪声环境下该算法具有较好的定位性能。最后,本论文以Lab VIEW软件为开发平台驱动北京双诺公司的MP420采集卡,设计采集程序实现实际环境中语音信号实时采集和存储。利用Matlab软件对采集到的数据结合本论文提出的改进算法进行仿真验证。实验证明,该算法能更精确的实现声源的定位。
[Abstract]:Intelligent service robot is a kind of high-tech integrated intelligent equipment which can provide essential services to human beings in non-structural environment. It has many functions, among which hearing is one of the important symbols. It is an important means of interaction between machines and people as well as environment. In real environment, hearing can achieve 360 掳"auditory scene", which can search the target of sound source which is not in visual range. That is to say, the hearing can search for the target which is blurred by the object occlusion, or the sound source target located at the corner. Although, compared to vision, the study of it is still in the early stages. However, in recent years, auditory research has become an important issue, auditory sound source location is a new research hotspot. Acoustic source location based on hearing, that is, pick up speech signal by microphone array, and then determine the position of sound source through signal processing, which has become a research hotspot in the field of speech signal processing. It has important application prospect and practical significance. This paper studies the principle and characteristics of sound source location technology based on microphone array. First of all, the research status and principle of sound source location using microphone array are summarized, three technologies based on microphone array sound source location are discussed, and the difficulties encountered in the positioning process are summarized. This paper mainly introduces the localization technology based on time delay of arrival. Secondly, there are many localization algorithms based on time-of-arrival delay. In this paper, the generalized cross-correlation algorithm is studied, and an improved cross-power spectrum phase weighting algorithm is proposed to deal with the influence of noise in the localization process. Experimental results show that the algorithm has good localization performance in strong noise environment. Finally, this paper uses Lab VIEW software as the development platform to drive the MP420 acquisition card of Beijing Shuangnuo Company, and designs a collection program to realize the real-time acquisition and storage of voice signals in the actual environment. The Matlab software is used to simulate the collected data and the improved algorithm proposed in this paper. Experimental results show that the algorithm can more accurately achieve sound source location.
【学位授予单位】：兰州理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN912.3

【参考文献】