MRPF: An Architectural Transformation for Synthesis of High-Performance and Low-Power Digital Filters

摘要: We present a graph theoretical methodology that reduces the implementation complexity of vector multiplied by scalar The proposed approach is called MRP (minimally redundant parallel) optimization and presented in FIR filtering framework to obtain low-complexity multiplierless implementation. key idea expand design space using shift inclusive differential coefficients together with computation reordering theoretic maximal sharing. transformed architecture filter obtained solving set cover problem graph. A simple algorithm based on greedy presented. merged common sub-expression elimination. simulation results show 70% 16% improvement terms computational over (transposed direct form) sub-expression, respectively, when carry, lookahead adder synthesized from Synopsys designware library 0.25 /spl mu/m technology.