Fast time series adaptive filtering algorithms based on the QRD inverse-updates method

We derive, from first principles, a new adaptive filtering algorithm for time-series data based on the QRD inverse-updates method of Pan and Plemmons. In common with other fast algorithms for time-series adaptive filtering, this algorithm only requires O(p) operations for the solution of pth order problem. Unlike previous fast algorithm based on the QRD technique, the algorithm presented here (in both square-root and square-root-free forms) explicitly produces the transversal filter weights. The results of some preliminary computer simulations of the algorithm, using finite-precision floating-point arithmetic, are presented.