A GENERALIZED LEAST-SQUARES FAST TRANSVERSAL FILTER ALGORITHM FOR THE DECISION FEEDBACK EQUALIZATION OF DISPERSIVE CHANNELS

Recently computationally efficient recursive least squares adaptive filtering algorithms have been introduced in transversal filter form as well as in lattice form. Fast transversal filter implementation of the generalized recursive least squares adaptive algorithm, which is appropriate for multichannel adaptive filtering and estimation, is presented in this paper. The above algorithm is derived in particular for the decision feedback equalization of dispersive channels, when the number of stages in the forward and backward filters are different. The algorithm considered in this paper is advantageous over the generalized least squares lattice algorithm considered by Ling and Proakis, because of the less computational effort required. The convergence rate of the mean square error for the decision feedback equalization of QPSK data transmitted through the dispersive channel using the generalized fast transversal filter algorithm is obtained using computer simulations. © 1990.