一种峰值功耗约束下的SoC蚁群测试调度算法  

Effective SoC Test Scheduling under Peak Power Constraints Based on the Ant Colony Algorithm

作　　者：崔小乐[1] 程伟[1] 李崇仁[1] 

机构地区：[1]北京大学深圳研究生院集成微系统科学工程与应用重点实验室,广东深圳518055

出　　处：《微电子学与计算机》2011年第7期5-8,共4页Microelectronics & Computer

基　　金：国家自然科学基金项目(61006032);广东省科技计划项目(2009A011604001;2009B090600084);深圳市科技计划项目(JC200903160361A;08CXY-02)

摘　　要：基于扫描链技术的SoC芯片测试可产生比正常使用模式下更大的功耗,这将会对器件可靠性产生不利影响,故在测试时需要将芯片测试功耗控制在允许峰值功耗之下.文中采用蚁群优化思路设计SoC测试调度算法,用于在峰值功耗和TAM总线最大宽度约束下降低SoC测试时间.实验结果表明,本方法优于先前已发表的相关方法.The SoC IC generates much more power dissipation during testing which may cause reliability problem to the device.Hence,it is desirable to schedule the tests under the allowable peak power constraint.This work presents,based on the Ant Colony Optimization（ACO） algorithm,a test scheduling to reduce the test application time under the peak power constraint as well as the constraint of maximum TAM bus width.Experimental results show that the algorithm is more effective in reducing test time than the earlier published methods.

关 键 词：测试调度 测试功耗 蚁群优化 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]