摘要
针对基于XNOR/OR的固定极性Reed-Muller电路(FPRM)逻辑电路面积优化方法搜索最优解速度较慢,易陷入局部最优等问题,提出一种新的FPRM逻辑电路面积优化方法,利用基于分群游走机制的灰狼优化算法(GDGWO)搜索电路面积最小的FPRM电路.GDGWO在初始化种群后,采取“轮盘赌”选择算法选出合适的新群体头狼,以提高种群多样性;执行种群分裂机制,防止因原始种群陷入局部最优而降低算法的鲁棒性;在分群搜索开发过程中引入改进后的随机游走策略,使灰狼种群能够更快地包围猎物,提高算法的收敛速度.基于北卡罗来纳微电子中心Benchmark测试电路的实验结果表明,GDGWO与粒子群算法相比,电路面积优化率提升57.42%;与黑猩猩算法相比,提升41.94%;与原始灰狼优化算法相比,提升43.68%.
Area optimization for fixed polarity Reed-Muller(FPRM)circuit is searching in the polarity optimization space for a polarity corresponding to the circuit with the smallest total number of XNOR terms and OR terms,and is a dual-valued combinatorial optimization problem.For the existing XNOR/OR-based area optimization for FPRM circuit problems such as slow search for optimal solutions and easiness to fall into local optimum,an area optimization for FPRM circuit approach is proposed,which used the groups in a division of grey wolf optimizer(GDGWO)based on the subgroup wandering mechanism to search for the circuit area with the smallest FPRM circuit.The algorithm applied a roulette selection algorithm to select a suitable new group of alpha wolves after initializing the population,aiming at improving the population diversity.It incorporated a population splitting mechanism to prevent the robustness of the algorithm from being reduced by the original population falling into local optimum;during the development of the split group search,an improved random wandering strategy was introduced,and the gray wolf population could encircle the prey faster to improve the convergence speed of the algorithm.The experimental results based on the microelectronics center of North Carolina Benchmark test circuit showed that the highest circuit area optimization rate of the GDGWO,compared with the particle swarm algorithm,was 57.42%;compared with the chimp optimization algorithm,the highest rate was 41.94%;and compared with the original gray wolf optimization algorithm,the highest rate was 43.68%.
作者
曹新龙
何振学
王伊瑾
赵晓君
张艳
肖利民
王翔
CAO Xin-long;HE Zhen-xue;WANG Yi-jin;ZHAO Xiao-jun;ZHANG Yan;XIAO Li-min;WANG Xiang(Key Laboratory of Agricultural Big Data of Hebei Province,Hebei Agricultural University,Baoding 071001,Hebei,China;School of Computer Science and Engineering,Beihang University,Beijing 100191,China;School of Electronic Information Engineering,Beihang University,Beijing 100191,China)
出处
《兰州大学学报(自然科学版)》
CAS
CSCD
北大核心
2024年第4期521-527,共7页
Journal of Lanzhou University(Natural Sciences)
基金
国家自然科学基金项目(62102130)
中央引导地方科技发展资金项目(226Z0201G)
河北省自然科学基金项目(F2024204001,F2020204003)
河北省青年拔尖人才计划项目(BJ2019008)
河北省高等学校科学技术研究项目(QN2024138)
。