Combining MUSIC Algorithm and Adaptive Beamforming to Improve Online Call Quality

In this paper, proposing a uniform circular microphone array (UCA) model that combines the multiple signal classification algorithm (MUSIC) and the adaptive beamforming technique to improve the quality of online calls. MUSIC algorithm is used to accurately detect the direction of arrival (DOA) of signal sources, while adaptive beamforming using the least mean square (LMS) algorithm can eliminate unacceptable sources and noise. As a result, the UCA system can actively select the desired signal source. Based on simulation results for three narrowband sinusoidal signal sources, the proposed system shows that it meets the requirements of direction detection, changing appropriate adaptive weight values, and limiting the influence of unwanted sources and noise. From there, the desired signal is accurately filtered with consistent filtered power.