POLARIZED-X ALGORITHM FOR USE IN FIXED-POINT PROCESSING

Using the operation by fixed-point processing, cheap and fast processo rs can be realized, which helps to reduce the production cost. This pa per uses a newly defined cost function. It discusses also the converge nce of the ''polarized-x'' method. This method updates the coefficient of the nonrecursive filter based on the product with the reference si gnal polarity, which is an operation suited to fixed-point processing. A normalized recursive adaptation algorithm is derived from the new c ost function, which is useful in the application to the system with th e reference signal with a rapidly varying power. It is shown that the stability condition is given as 0 < K < 4/pi, and the estimation error increases only by approximately 2 dB compared to the conventional nor malized least mean square (NLMS) algorithm. The obtained structure is modified to construct an adaptive algorithm so that the normalization is applied individually to the coefficients and the convergence is det eriorated little for the operation by the fixed-point processing.