Of The Sound Source Localization Based On Subspace

The sound source localization is an important issue in the field of speech signalprocessing. It has widely used in multimedia systems, intelligent robot systems, videoconference systems and the pervasive computing. The sound source localizationtechnique receives sound signals through a microphone and analysis signals data inorder to got sound signal’s position. There are imprecise and computational problemsin the existing positioning technology, and how to solve these problems is a challenge.This paper studies sound source localization based subspace and the existingtechnology has been improved in order to expect to achieve better results.In thispaper, some basic knowledge of the array signal processing and speech signalprocessing are introduced first. The knowledge is the basis for further theoreticalanalysis and experimental. Then the article describes the positioning of the multiplesignal classification algorithm. in the process of sound signal, Microphone array willhave a greater error in the magnitude of the speech signal. the positioning accuracy rateof distance will decline. This paper proposes an improved algorithm to improve theaccuracy of distance positioning. The algorithm use phase information to locate thespeech signal position. In this paper, we further introduce how to solve the broadbandspeech signal. This paper explained the basic principles and the advantages anddisadvantages of the ISM algorithm. According to the characteristics of the speechsignal presented an improved algorithm. The experiments show that the improvedalgorithm can locate the sound signal in the case of less number of sub-band.