基于模拟退火算法的稀疏系数抽取滤波器设计  

Design of sparse coefficient decimation filter using simulated annealing algorithm

作　　者：吴尘[1] 徐新洲[1] 王浩[2] 赵力[1] 

机构地区：[1]东南大学水声信号处理教育部重点实验室,南京210096 [2]杭州电子科技大学通信工程学院,杭州310018

出　　处：《东南大学学报（自然科学版）》2015年第4期631-634,共4页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金资助项目(61231002;61273266;61375028);教育部博士点专项基金资助项目(20110092130004)

摘　　要：为了减少系数抽取滤波器的乘法器个数,提出了一种基于模拟退火算法的稀疏系数抽取滤波器设计方法.该方法将系数抽取滤波器的非凸稀疏设计转化为一个寻找最稀疏的系数抽取滤波器的零系数位置集合的组合优化问题,然后利用模拟退火算法来求解该问题.该方法结合贪婪思想逐步地增加系数抽取滤波器的稀疏度,直到没有更加稀疏的设计结果存在.在每一步中将系数抽取滤波器的稀疏度固定,利用模拟退火算法来寻找满足给定设计标准的系数抽取滤波器的零系数位置集合.实验结果表明,该方法可以有效地减少系数抽取滤波器所需的乘法器个数.In order to reduce the number of multipliers of the coefficient decimation filter, a design method based on the simulated annealing （SA） algorithm is proposed. The proposed method trans- forms the non-convex sparse design of coefficient decimation filter into a combinatorial optimization problem which finds the sparsest set of the positions of the zero coefficients, and then uses the simu- lated annealing algorithm to solve it. Combining the greedy theory, the method successively increa- ses the sparsity of the coefficient decimation filter until no sparser design result exists. At each step of the method, the sparsity of the coefficient decimation filter is fixed, and SA is used for finding the set of the positions of the zero coefficients that satisfies the design specifications. Simulation results demonstrate that the proposed method can effectively reduce the number of multipliers of the coeffi- cient decimation filter.

关 键 词：系数抽取滤波器 稀疏 模拟退火 线性优化 

分 类 号：TN911.72[电子电信—通信与信息系统]