基于PTIDR编码的测试数据压缩算法  

A Test Set Compression Algorithm Based on PTIDR Code

作　　者：李国亮[1] 冯建华[1] 崔小乐[2] 

机构地区：[1]北京大学信息科学技术学院,北京100871 [2]北京大学深圳研究生院信息工程学院,深圳518055

出　　处：《计算机辅助设计与图形学学报》2008年第2期161-166,共6页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(90207018,60576030)

摘　　要：为减少测试数据存储量,提出一种有效的新型测试数据压缩编码--PTIDR编码,并构建了基于该编码的压缩/解压缩方案.PTIDR编码能够取得比FDR,EFDR,Alternating FDR等编码更高的压缩率,其解码器也较简单、易实现,且能有效地降低硬件开销.与Selective Huff man,CDCR编码相比,PTIDR编码能够得到较高的压缩率面积开销比.特别地,在差分测试集中0的概率满足p≥0.7610时,PTIDR编码能取得比FDR编码更高的压缩率,从而降低芯片测试成本.We present a new test data compression and decompression architecture based on a novel and efficient code, named PTIDR code. The proposed approach can acquire better compression efficiency than that of FDR （frequency-directed run-length）, EFDR （extended FDR）, alternating FDR etc. The decoder of PTIDR is also simpler, easier to realize and needs less hardware consumption. Compared with algorithms such as selective Huffman and CDCR （combining dictionary coding and LFSR reseeding）, PTIDR can acquire higher CR/AR （ratio of compression ratio and area ratio）. Especially, when the probability of 0s in the difference test set is greater than or equal to 0.7610, it can acquire better compression efficiency than FDR code, and thus, reducing the test cost of the chip.

关 键 词：测试数据压缩 哈夫曼编码 FDR编码 PTIDR编码 

分 类 号：TN47[电子电信—微电子学与固体电子学]