基于遗传算法的IP核测试调度优化

摘要

测试调度能够很好的减少测试时间和降低测试成本.通过调度,SOC中尽可能多的IP核可以进行并行测试,然而过度的并行测试会引起功耗过高,对SOC产生不利影响.为了改善这个问题,考虑峰值功耗因素的限制,提出一种基于遗传算法的IP核测试调度优化方案,寻求最短测试时间.通过对ISCAS标准电路组成的SOC进行仿真实验,验证了该方案良好的优化效果,实现了IP核测试调度的可靠性和经济性.%Test scheduling will reduce testing time and decrease testing cost. As many IP cores as possible can be tested in parallel by scheduling. But too many parallel tests will lead to over power and da harm to the SOC. To solve the problem, an IP cores test scheduling optimization method based on genetic algorithm is proposed. Searching for the shortest test time under considering power constrain. The experiment results of an ISCAS circuit based SOC verify the good optimization effect of the method, achieves reliability and economic for IP cores test scheduling.