基于云进化算法的NoC测试规划

Test Scheduling of NoC Based on Cloud Evolution Algorithm

针对NoC测试时,如何在功耗限制下利用有限的片上资源最大化并行测试,以优化NoC测试时间的问题,文中提出一种利用云进化算法进行测试规划的方法,可以有效提高测试效率。该方法复用NoC的片上资源作为TAM,采用非抢占式测试和XY路由方式,通过云进化算法优化待测IP核在各条TAM上的分配方式寻找最佳方案。在ITC'02标准电路上的实验结果表明,该方法有效降低了测试时间,提高了测试效率。

In order to reduce the test time of NoC, the problem of how to use the limited on-chip resources to maximize parallel test in power constraints is discussed. This paper presents a method for using the cloud evolution algorithm to schedule the IP cores, which effectively improves the test efficiency. The method reuses the on-chip network as TAM （test access mechanism）, using non-preemptive test and XY routing mode and optimizing the allo-cation of IP cores in each TAM by the cloud evolution algorithm. Experimental results from the ITC＇O2 system-on-chip benchmarks show that this method effectively reduces the test time and improves the test efficiency.