Research On Sound Source Localization Based On Distributed Microphone Array

Sound source localization has been widely used in video conference,intelligent transportation,security surveillance,human-computer interaction,and other field.In recent years,with the development of network communication and mobile computing,distributed microphone array was gradually developed.Distributed microphone array is the combination of distributed system and microphone array.Its system features are similar with wireless sensor network and have the ability of speech processing.Distributed microphone array has a number of advantages over traditional microphone array,such as greater flexibility,and good fault tolerance.At present,sound source localization based on distributed microphone array is a hot topic.This thesis studies on sound source localization based on distributed microphone array,and the major works are listed as follows:(1)In order to simplify the system complexity and reduce the computational load,in this thesis,optimal arrays is selected to participate in orientation.Firstly,estimate the direct to reverberation energy ratio(DRR)of each array,then according to the DRR,choose the microphone array.The larger the DRR,the smaller the sound that signal received by microphone is affected by reverberation.(2)An improved time difference of arrival(TDOA)estimation method is presented by the generalized cross-correlation.In order to enhance the robustness of the algorithm to noise and reverberation,two parameters related to the signal-to-noise ratio(SNR)and DRR are introduced in the weighting function.The experimental results show that the proposed algorithm can achieve higher accuracy when noise and reverberation are serious.(3)A novel hybrid localization algorithm based on SRP-??-v is presented.SRP-??-v is the improved steered response power(SRP)method which takes into account the effects of noise and reverberation.The geometric algorithm is first used to estimate the approximate range of the sound source;then,SRP-??-v is exploited to determine the precise location of sound source.The experimental results verify the validation of the proposed algorithm...