Ant colony optimization for design of FIR filters with signed-power-of-two coefficients

A heuristic approach based on ant colony optimization is proposed for the design of multiplierless finite impulse response (FIR) filters with signed-power-of-two (SPT) coefficients. The objective is to minimize non-zero SPT terms to reduce the number of adders that are needed for filtering with constant SPT coefficients, while meeting the required specifications of the filter. Each SPT coefficient is placed on several virtual paths whose length is determined by non-zero SPT terms. Each ant searches for the shortest path according to the residual pheromone that is dynamically calculated by measuring the distance between current path and previous optimal path. Simulation results show that suboptimal solutions can be obtained within reasonable computational time.