一种交替游程编码的SOC测试数据压缩方法

Method of compression of SOC test data based on alternative run-length encoding

以减少系统芯片SOC测试时间和测试数据量为目标,引入量子进化算法完成层次型SOC在功耗约束条件下的建模和算法设计并得到相应的测试集,通过共享广播技术整合多个芯核的测试集,采用交替游程编码的方法压缩测试集,该方法同时考虑测试数据中的"0"和"1"游程,可以大大减少长度较短的游程数量,针对国际标准片上系统芯片验证表明,与其他算法相比,量子进化算法有效满足了功耗要求同时获得了较短的测试时间,与其他压缩编码方法相比,提出的方法获得了更有效的压缩效果。

Aiming at the reduction of SOC test time and test data volume, this paper introduces quantum evolutionary algorithm to complete establishing the module and designing the algorithm for hierarchical SOC under the condition of the power constraint and get the corresponding test set.In succession, the test set of multiple cores is integrated through share broadcast technology, and then the method of altemative run-length encoding is combined to compress the test set.This method takes into account both the ＂0＂ and ＂1＂ run length,so it can greatly reduce the number of the shorter run-length.The experimental results for SOC benchmark show that： Compared with other algorithm, quantum algorithm can efficiently meet the demand of test power while it acquires shorter test time.Compared with other compression encoding method, the method of this paper acquires a more effective compression.