Decorrelating (DECOR) transformations for low-power adaptive filters

Abstract

Presented in this paper are decorrelating transformations referred to as DECOR transformations to reduce the power dissipation in adaptive lters. The coe cients generated by the weight update block i n an adaptive lter are passed through a decorrelating block such that fewer bits are required to represent the coe cients. Thus, the size of the arithmetic units in the lter F-block is reduced thereby reducing the power dissipation. The DECOR transform is well suited for narrow-band lters because there is signi cant correlation between adjacent coe cients. In addition, the e ectiveness of DECOR transforms increases with increase in the order of the lter and decrease in coe cient precision. Simulation results indicate reduction in power dissipation in the F-block ranging from 12 to 38 for lter bandwidths ranging from 0:15fs to 0:025fs where fs is the sample rate.