Efficient design of sparse FIR filters with optimized filter length

摘要: A large number of experiments have demonstrated that for an FIR filter the sparsity coefficients is highly related to its order. However, traditional sparse design methods focus on how increase zero-valued coefficients, but overlook impact orders performance. As attempt jointly optimize length and filter, a novel method proposed in this paper linear-phase filters. With peak error constraints, objective function problem formulated as combination measure effective Then, then recast weighted l 0 -norm optimization problem, which solved by efficient numerical based iterative-reweighted-least-squares (IRLS) algorithms. Experimental results illustrate can efficiently reduce order while enhancing filter.