提出一种基于部分测试向量奇偶位切分的LFSR重新播种测试方法.针对确定测试集中各个测试向量包含确定位的位数有较大差异以及测试向量所含的确定位大多连续成块的特点,通过奇偶切分部分确定位较多的向量,使得编码压缩的LFSR度数得到有效降低,从而提高了测试数据压缩率.其解压缩电路仍然采用单个LFSR进行解码与切分向量的合并.与目前国际同类编码压缩方法相比,具有测试数据压缩率高、解压硬件开销低、测试数据传输协议简单等特点.
A novel scheme of LFSR reseeding based on dividing some test cubes with many specified bits is presented in this paper. As the numbers of specified bits of test cubes vary widely and the specified bits always appear consecutively in a test cube, the proposed scheme divides the test cube with many specified bits by separating odd bits from even bits. The cube is divided into two new cubes each of which has almost half of specified bits of the original cube. It can reduce the largest number of specified bits in any test cube in the test set. Thus it can reduce the degree of LFSR encoding and increase the encoding efficiency. Only a single LFSR is used for decompressing the LFSR seeds and combining the divided cubes. Its control circuit is simple. Thus the hardware overhead is less than LFSR reseeding alone. Compared with the previous schemes, the proposed scheme can increase the encoding efficiency with less hardware overhead. And the LFSR seeds are same length so the community protocol for transmitting test data with the ATE is simple.