Coefficient Ratio Of Adaptive Filtering Algorithm And DSP Implementation

Adaptive filtering is an important branch of the digital signal processing field, adaptive filtering technology requires no prior knowledge, through self-regulating process to adapt to the external environment, or tracking unstable random signal and constantly changing, and ultimately achieve optimum filter performance. Currently, the technology has been widely used in radar, sonar, communications, earthquakes and biomedicine etc. With the development of digital signal processing theory and an increase in demand for a variety of applications, a variety of adaptive filtering algorithms emerge in endless, sparse adaptive filtering is one of them. The article based on the study of adaptive filtering technology and related algorithms,focused on applied to sparse systems of proportional coefficient adaptive filtering algorithms and build a adaptive noise cancellation hardware platform based on DSP.In this paper, the research methods involve theoretical analysis, software simulation and hardware design. Firstly, this paper analyzes the working principle of adaptive filter, structure and practical application. Discussed standard least mean square algorithm(LMS) and least square algorithm(RLS); In order to improve the performance of the algorithm and broaden the scope of application of the algorithm, this paper has made a further discussion about the normalized LMS algorithm(NLMS), simplified LMS algorithm, the improved variable step size LMS algorithm,to solve the contradiction between convergence speed and steady state error; On the basis of research PNLMS algorithm, we discussed the IPNLMS algorithm which has a better performance for different degrees of sparse on impulse response.Secondly, using MATLAB software, made simulation experiments on NLMS algorithm,two improved LMS algorithm, RLS algorithm, PNLMS algorithm for sparse system identification and noise cancellation. Through the regulation of key parameters-step, order,signal to noise ratio, analyze their impact on the convergence rate, steady-state error and filtering effect. Experimental results shows that compared to traditional adaptive algorithms,the proportion coefficient adaptive filtering algorithm shows better performance on sparse system.Finally, this paper takes high-speed digital signal processing chip TMS320F2812 as the core design of an adaptive noise cancellation based on PNLMS algorithm, through building peripheral circuit design adaptive noise cancellation system for sparse signal, the results verify the effectiveness of the algorithms and hardware systems.