中文摘要：

提出了一种基于迭代列消元法的线性分组码参数全盲识别算法。该方法首先对截获二进制码流构造截获矩阵，然后对截获矩阵进行迭代列消元法，利用相关列的归一化数目最大值来识别码字长度和同步时刻。同时，对截获矩阵进行迭代列消元法后的矩阵，选取其中一个小矩阵窗内相关列都是全零列，将其对应的转移矩阵中的列向量横向放入校验矩阵，完成校验矩阵的识别。此外，根据相关列和独立列中的码元0的比例减去1的比例的统计特性差异，提出了判别相关列和独立列的门限。仿真结果证明，在误码率为0．01时，该文算法仍能取得很好的效果。

英文摘要：

In order to solve the blind identification problem of linear block code, an iterative column elimination algorithm is proposed in this paper. First of all, a matrix is filled with intercepted bit stream received from binary symmetric channel, and then an iterative column elimination algorithm is introduced which attempts to eliminate parity bits in codewords of noisy data. Second, the length and synchronization of codewords are estimated when the normalized number of dependent column reaches the maximum. Third, After an iterative column elimination algorithm is applied in the intercepted matrix made of the right length and synchronization, we choose a window whose dependent columns is all zeros column, so the corresponding column in the transition matrix is placed in parity-check matrix. What is more, the threshold is introduced, according to the statistical characteristics differences between dependent column and independent column theoretically. Finally, the simulation results show that the probability of correct recognition in the proposed algorithm is good when the bit error rate is one percent.