Multiscan-based Test Data Compression Using UBI Dictionary and Bitmask

In this paper, we propose a test data compression method, which combines the advantages of dictionary-based compression and bit mask-based compression. We present a kind of Unfixed-Based Index (UBI), which uses shorter indexes to represent the slices with higher occurrence frequency, while uses longer indexes to represent the slices with lower occurrence frequency. Then, we adopt efficient bit mask and Max-Degree based Clique Partition Algorithm (MD-CPA) to create as many compatible slices as possible. Moreover, we propose a new Variable Prefix Dual-Run-Length (VPDRL) code to compress the un-compatible slices. To demonstrate the usefulness of our approach, we have applied our scheme on ISCAS´89 benchmarks and compared our results with existing compression techniques. Our algorithm outperforms existing dictionary-based compression and other compression methods.