Low power filtering using approximate processing for DSP applications

Abstract
The growing demand for portable multimedia devices has placed increased importance on low power solutions for DSP tasks such as filtering and source coding. An approach to power reduction in digital CMOS filter design using approximate processing is presented. This involves adaptively reducing the number of operations switched per sample based on signal statistics. Speech filtering examples demonstrate that power consumption can be reduced over conventional solutions by an order of magnitude for wireless applications.

This publication has 6 references indexed in Scilit: