Adaptive beamforming system using blind deconvolution with sharp mainlobe

There are many researches to develop signal enhancement systems which can work under a common workspace environment. This paper proposes a method of signal enhancement which has an iterative process based on the DSA (delay and sum array) method for estimation of DOA (direction of arrival) and AMUSE (algorithm for multiple unknown signals extraction) for blind deconvolution using a new cost-function. The blind deconvolution extracts the information of the target signal and estimates a coherent signal by a deconvolution of the impulse responses defined as delay factors between sensors. The proposed method can obtain the sharp directivity pattern that the half-bandwidth is narrower than 2/spl deg/ for a frequency range from 100 Hz to 3 kHz when the SNR is varied from -10 dB to +10 dB.