Spiking neural network for sound localization using microphone array

Sound source localization is very useful in various fields of engineering applications. Due to remarkable ability of humans for sound source localization, this paper describes a simple biological inspired model based on spiking neural network for localizing sound source. In this paper, a simple method using analog and digital combinational circuits is proposed for generating spikes. Because of simplicity of the proposed generating spikes method, in this paper, microphone array is utilized instead of using two microphones in order to increase accuracy. Then, a neural structure based on spiking neural network is proposed which works by means of microphone´s signals. This structure is designed in way that can be implemented on Field Programming Gate Array (FPGA) properly. Simulation results show that implementing of this model for different types of microphone array is not only very simple, but also shows high accuracy of localizing sound source.