摘要
基于2DMesh结构的片上网络在设计之初就要考虑模块映射问题,以满足通信功耗的约束。提出一种基于遗传蚁群映射算法的方法解决片上网络设计中通信功耗最小化问题。该算法针对标准蚁群算法易于出现早熟停滞等缺陷,引入轮盘赌选择机制及染色体杂交等手段,使映射功耗函数快速收敛,达到良好的全局寻优效果。
At the beginning of Network on Chip(NoC) design, cores of mapping are considered based on 2D Mesh topology such that the communication energy constraint is satisfied. A Genetic And Ant Colony(GAAC) mapping algorithm is presented to solve NoC design problem with an objective of minimizing the communication energy. To solve the problem that the ant colony algorithm often gets stock into premature stagnation in the iteration process, roulette wheel selection and crossover of chromosome are introduced to improve the result of energy-aware mapping function fast converged and optimized solution.
出处
《计算机工程》
CAS
CSCD
北大核心
2010年第22期262-264,共3页
Computer Engineering
基金
江苏省专用集成电路设计重点实验室开放课题基金资助项目(JSICK0801)
南通大学自然科学基金资助项目(07Z122)
关键词
片上网络
遗传蚁群映射算法
轮盘赌选择
染色体杂交
Network on Chip(NoC)
Genetic And Ant Colony(GAAC) mapping algorithm
roulette wheel selection
crossover of chromosome