Comparison of STFT based direction of arrival estimation techniques for speech signal
Anukul Anand, Manoj Kumar Mukul · 2016
Direction of Arrival (DOA) estimation of signal deals in estimating the direction from where the signal is originating by joint processing the multi-channel signals captured by the array of sensors. This paper presents the comparison between various Direction of Arrival Estimation techniques for the speech signal. Short-Time Fourier Transform (STFT) is used as a tool to transform the time domain signal into frequency domain. The performance analysis of these methods such as Classical, Capon's Beamforming, Multiple Signal Classification (MUSIC) and Minimum Norm is analyzed in the frequency domain by selecting signal of individual frequency bin of interest. The experiment conducted uses a uniform linear array (ULA) of microphones to capture the spatial information of the speaker. MATLAB is used for evaluating the performance of these methods.