一种用于测试数据压缩的自适应EFDR编码方法

摘要

该文提出一种用于测试数据压缩的自适应EFDR(Extended Frequency-Directed Run-length)编码方法。该方法以EFDR编码为基础，增加了一个用于表示后缀与前缀编码长度差值的参数N，对测试集中的每个测试向量，根据其游程分布情况，选择最合适的 N 值进行编码，提高了编码效率。在解码方面，编码后的码字经过简单的数学运算即可恢复得到原测试数据的游程长度，且不同 N 值下的编码码字均可使用相同的解码电路来解码，因此解码电路具有较小的硬件开销。对ISCAS-89部分标准电路的实验结果表明，该方法的平均压缩率达到69.87%，较原EFDR编码方法提高了4.07%。%An adaptive Extended Frequency-Directed Run-length (EFDR) code method for test data compression is presented in this paper. The method is based on EFDR code, and adds an additional parameterN, which is used to represent the code length difference between tail and prefix. According to the distribution of the runs in each test vector of the test set, the method selects the most suitableN values to code, and it can improve the compression ratio. For the decompression, according to the size of the codeword, the run length of the original test data can be obtained with a simple mathematical operation. Meanwhile, those codeword under different parameter values can be decoded by the same decompression circuit. Thus, the decompression circuit can keep in a low hardware cost level. The experimental result shows that the average compression rate of the proposed method can achieve to 69.87%, over 4.07% than original EFDR code method.