Broadband Signal Enhancement Method Using Adaptive Beamforming

The final goal of communication is to transmit information without error from a transmitter to a receiver.However,a speech data transferred in the form of a signal is always interrupted/disturbed by interference or noise signals which cause the information to not be received clearly by the receiver.The study of signal processing has for a long time focused on this issue,in order to combat and eliminate the interference and noise signal perturbing the desired signal.An important technique called beamforming to eliminate the unwanted signals in the direction of the desired signal was proposed in the early twentieth century.The technique has an objective to produce beams that can receive a signal from a specific direction while blocking the signals from the other directions.Adaptive beamforming is part of the beamforming technique for the purpose of enhancing the desired signal and suppress the interference effectively.For a while,the technique is known as a significant technology of the adaptive array antenna(or smart antenna).The smart antenna use the technique to obtain the optimum weight by using certain adaptive beamforming algorithms to further set the main lobe of the beamformer to focus on the direction of arrival of the desired signal and suppress at the same time the disturbing or the interfering signals.The technique have been widely used in many application areas such as wireless communication,radar,sonar,medical imaging and so on as an effective solution to signal detection and estimation in harsh environment.In this thesis,we focus on the broadband signal enhancement method using adaptive beamforming.For that reason,broadband signals have been implemented to a newly proposed adaptive beamforming method called robust adaptive beamformers based on worstcase optimization and constraints on magnitude response(RAB-CMR-WC)in order to extract a desired speech signal and reject the unwanted signals.Although the proposed method is a narrowband adaptive beamformer,we cannot directly apply the wideband speech signals to the proposed method.Therefore,a technique called fast Fourier Transform(FFT)have been used to narrow the wideband speech signals at the output of each sensor array element.Moreover,the frequency bins at the output of the FFTs are sampled into different array snapshot vectors and then the narrowband adaptive beamformer algorithm of the new proposed method is used.Finally,the inverse of Fast Fourier Transform(IFFT)is used to generate the desired speech signal.