Exploring the use of heuristic-based algorithms for the ordering and partitioning of coefficients for power efficient fir filters realization

Abstract

This paper proposes the exploration of different heuristic-based algorithms for the best ordering and partitioning of coefficients in Finite Impulse Response (FIR) filters. Due to the characteristics of the FIR filter algorithms, which involve multiplications of input data with appropriate coefficients, the best ordering and partitioning of these operations can contribute for the reduction of the switching activity, what leads to the minimization of power consumption in the filters. Two heuristic-based algorithms named Nearest neighbor and Bellmore and Nemhauser are used in the tests. A new heuristic-based algorithm named Anedma is proposed to get as near as possible to the optimal solution for the ordering and partitioning of larger filter instances. Two set of coefficients were used for the tests, and the results are presented in terms of the Hamming distance between the consecutive coefficients. Sequential and semi-parallel FIR filter architectures were implemented with the coefficients ordered and partitioned. As will be shown, depending on the set of coefficients used, based on the guidance given by the heuristic algorithms, the power consumption of the FIR filter architectures can be reduced significantly.