Power minimization in DSP application specific systems using algorithm selection

Abstract

We introduce the algorithm selection problem for power minimization. After demonstrating the high impact of this synthesis task on the power consumption of the final implementation using a case study, we studied its computational complexity. We present an efficient optimization intensive algorithm for power minimization using algorithm selection. We applied the marginal utility-based algorithm for algorithm selection on three DSP examples. A table illustrates the effectiveness of the power optimization using algorithm selection on one audio (LMS DCT transform domain filter) and two video (NTSC formatter and DPCM coder) applications. On several DSP examples more than an order of magnitude reduction in power is demonstrated.