许川佩, 凌景. 基于ICEA的3D NoC测试规划研究[J]. 微电子学与计算机, 2016, 33(7): 74-78.
引用本文: 许川佩, 凌景. 基于ICEA的3D NoC测试规划研究[J]. 微电子学与计算机, 2016, 33(7): 74-78.
XU Chuan-pei, LING Jing. Test Scheduling for Three Dimensional Network-on-Chip Based on Improved Cloud Evolutionary Algorithm[J]. Microelectronics & Computer, 2016, 33(7): 74-78.
Citation: XU Chuan-pei, LING Jing. Test Scheduling for Three Dimensional Network-on-Chip Based on Improved Cloud Evolutionary Algorithm[J]. Microelectronics & Computer, 2016, 33(7): 74-78.

基于ICEA的3D NoC测试规划研究

Test Scheduling for Three Dimensional Network-on-Chip Based on Improved Cloud Evolutionary Algorithm

  • 摘要: 采用带分复用的测试方法, 结合自然进化策略的云进化算法(CEA)优化3D NoC测试规划, 依据带分复用的特点建立3D NoC测试时间优化模型, 在功耗和带宽双重约束下, 采用改进的云进化算法(ICEA)对该模型进行求解, 完成3D NoC测试.以ITC’02基准电路为实验对象, 实验结果表明该算法能迅速收敛到适应值最佳的测试规划, 实现3D NoC资源内核的最大化并行测试, 提高测试效率, 缩短测试时间.

     

    Abstract: This paper adopt the test method of bandwidth division multiplexing, based on cloud evolutionary algorithm that combined with the natural evolution strategies to optimize the test scheduling for 3D NoC. Establish test time model according to the features of bandwidth division multiplexing, under the dual constraints of power and bandwidth, the model is solved with improved cloud evolutionary algorithm to complete the test. With ITC'02 test circuit as the simulation object, experimental results show this algorithm can search best test scheduling quickly, maximize the parallel test on 3D NoC resources cores, improve test efficiency and reduce testing time.

     

/

返回文章
返回