Comparative Analysis of LMS and RLS Using Sub-band Adaptive Noise Cancellation Technique

Pages:104-105
Md. Fahim Ansari1, Anita2 (Department of EEE, BRCM College of Engineering and Technology Bahal, Bhiwani, Haryana1, Instrumentation and Control Engineering, Scholar2)

Noise cancellation technology is aimed at reducing unwanted ambient sound. Noise is the form of unwanted signal which attenuate the flow of information signal. So some techniques are used to avoid noise. The problem of the noise controlling in communication is tremendous amount of research. Adaptive filtering plays an important role in the field of signal processing and has various applications in fields of speech processing & communications. Adaptive filter is a filter that self-adjusts its transfer function according to an optimizing algorithm. Examples are speech enhancement, system identification, interference cancellation & speech coding. The applications where the required adaptive-filter order is high, as long impulse response in such applications. The adaptive-filtering algorithm is a large number of computations. In addition, the high order affects the convergence speed of the adaptive filter to overcome this problem sub band adaptive noise cancellation is used. In sub band adaptive filtering, both the input signal and the desired signal are split into frequency sub-bands via an analysis filter bank. Assuming that the signal decomposition in sub channels is effective, so do decimation of these sub-band signals and apply adaptive filtering to the resulting signals. Each sub-band adaptive filter usually has shorter impulse response than its full-band counterpart. The function of a filter is to remove unwanted part of signal. In sub band adaptive filtering different prototype filter are used in the analysis and synthesis filter banks. The analysis filter is modified the colored components and synthesis filter bank is optimized the input/output relationship to achieve minimum amplitude distortion. For doing this work we use the MATLAB simulink tool for the simulation. It provide an interactive graphical, developed algorithms, analyze and visualize, modeling environment, and define signal, parameter, and test data simulations. By simulation of sub band adaptive noise cancellation in matlab we have to conclude that LMS algorithm is the best as compare to RLS. As we increase the order of sub band then error signal will be low so signal output will be more correct. I.e. in 2order, 4order and 8 order and 16 order sub band noise cancellation 16order sub-band have high output and lowest error. Analyzing the best possible practices used 16 order sub-band for minimize the error. This thesis has been divided in seven sections.