The thesis presents a new adaptive notch filter (ANF) algorithm that is more accurate and efficient and has a faster convergent rate than previous ANF algorithms. In 1985, Nehorai designed an infinite impulse response (UR) ANF algorithm that has many advantages over previous ANF algorithms. It requires a minimal number of parameters with constrained poles and zeros. It has higher stability and sharper notches than any ANF algorithm until now. Because of the special filter structure and the recursive prediction error (RPE) method, however, the algorithm is very sensitive to the initial estimate of the filter coefficient and its covariance. Furthermore, convergence to the true filter coefficient is not guaranteed since the error-performance surface of the filter has its global minimum lying on a fairly flat region. We propose a new ANF algorithm that overcomes the convergence problem. By choosing a smaller notch bandwidth control parameter that makes the error-performance surface less flat, we can more easily detect a global minimum. We also propose a new convergence criterion to be used with the algorithm and a self-adjustment feature to reset the initial estimate of the filter coefficient and its covariance. This results in guaranteed convergence with more accurate results and more efficient computations than previous ANF algorithms.

Adaptive filtering may be applied in areas where an optimal filtering algorithm may not be known a-priori and where the filtering operation may be non-stationary. This field, or more generally, the field of adaptive systems, is one which may be regarded as mature, having been the subject of considerable research effort in the areas of control and signal processing for almost four decades. DFE (decision feedback equalization) in various forms has been proposed for detection on magnetic recording channel. An allpass filter is an alternative to the FIR (finite impulse response) forward equalizer which is normally implemented with DFE. This is because the allpass filter is a lower power and complexity alternative, though its behavior and performance are not very well understood yet. Here, an allpass structure implemented as first and second order IIR (infinite impulse response) filters is examined. Convergence for the LMS (least mean square) adaptation algorithm is studied and, moreover, some convergence conditions and bounds are developed, similarly to the well known FIR case. This thesis provides an useful analytical study of convergence of IIR adaptive filtering. This is accomplished by a systematic approximation of the covariance terms of the adaptive coefficients. The range of the step-size parameter of the LMS algorithm is developed under some simplifying assumptions. All the results obtained are verified by simulation (Matlab and C routines are used). / Graduation date: 1998

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

Thesis (Ph. D.)--University of Hong Kong, 2001. / Includes bibliographical references.

No description available.

No description available.

Thesis (M. App. Sc.)--Carleton University, 2004. / Includes bibliographical references (p. 78-80). Also available in electronic format on the Internet.

Thesis (M.App.Sc.) - Carleton University, 2005. / Includes bibliographical references (p. 87-88). Also available in electronic format on the Internet.

Thesis (M.App.Sc.) - Carleton University, 2006. / Includes bibliographical references (p. 108-112). Also available in electronic format on the Internet.

Thesis (M.S.)--Ohio University, August, 1994. / Title from PDF t.p.