Dual-microphone sound source localization in enclosed space based on time reversal and genetic algorithm

For sound source localization in enclosed space, the position and array shape of microphones have significant impact on localization performance. There has been a lot of researches on array shape of the microphones, while the study on the position of microphones is very few. In this paper the problem is studied using a dual-microphone localization method based on time reversal and genetic algorithm. The method is based on the adaptive focusing, strong anti-reverberation capability and the ability to overcome multipath effects of the time reversal algorithm. The optimal position of dual-microphone is found by genetic algorithm firstly, then the localization performance under different bandwidth and SNR in enclosed space is studied. The results show that the method has strong robustness, and it can achieve good localization effect under low SNR and low frequency.