A NEW QUASI-NEWTON ADAPTIVE FILTERING ALGORITHM

A new algorithm for FIR adaptive filters based on the quasi-Newton cla ss of optimization algorithms is described, A series of theorems demon strating the stability of the algorithm, boundedness and positive defi niteness of the estimated autocorrelation matrix of the input signal a re provided, The internal variables of the algorithm and their effect are also investigated in order to provide a better insight of the algo rithm's behavior, Extensive simulation results are presented for fixed and floating-point implementation which show that the proposed algori thm has comparable convergence speed and superior robustness relative to other known Newton-type algorithms, Furthermore, robustness is guar anteed for highly correlated or even nonpersistently exciting input si gnals, which makes the proposed algorithm a powerful alternative to th e LMS and the RLS algorithms.