AN EFFICIENT RECURSIVE TOTAL LEAST-SQUARES ALGORITHM FOR FIR ADAPTIVEFILTERING

An algorithm for recursively computing the total least squares (TLS) s olution to the adaptive filtering problem is described. This algorithm requires O(N) multiplications per iteration to effectively track the N-dimensional eigenvector associated with the minimum eigenvalue of an augmented sample covariance matrix. It is shown that the recursive le ast squares (RLS) algorithm generates biased adaptive filter coefficie nts when the filter input vector contains additive noise. The TLS solu tion on the other hand, is seen to produce unbiased solutions. Example s of standard adaptive filtering applications that result in noise bei ng added to the adaptive filter input vector are cited. Computer simul ations comparing the relative performance of RLS and recursive TLS are described.