系统工程与电子技术

• 通信与网络 • 上一篇    下一篇

基于关联规则的二进制线性分组码盲识别

张旻1,2,李歆昊1,2   

  1. 1. 中国人民解放军电子工程学院, 安徽 合肥 230037; 
    2. 安徽省电子制约技术重点实验室, 安徽 合肥 230037
  • 出版日期:2014-05-22 发布日期:2010-01-03

Blind identifying of binary linear block code based on association rules

ZHANG Min1,2, LI Xin-hao1,2   

  1. 1. Electronic Engineering Institute of the PLA, Hefei 230037,China; 
    2. Key Laboratory of Electronic Restricting Technique, Hefei 230037, China
  • Online:2014-05-22 Published:2010-01-03

摘要:

提出了一种基于关联规则的信道编码识别方法,解决了线性分组码的盲识别问题。在分析线性分组码中的信息位和监督位之间存在一一对应关系这一特性的基础上,利用关联规则寻找出隐藏在二进制序列中的对应关系,实现了分组码的盲识别。首先将按照一定规则排列的序列进行二分段划分,并依此构建相关事务项集,然后计算二分段项集中每项事务的支持度和置信度,在满足支持度条件下,通过遍历的方式寻找最大置信度和最少映射关系种类,达到正确识别编码的目的。由于采用了二分段划分的方法,因此极大地降低了关联规则搜索的复杂度,而且所提算法与现有识别方法相比对误码率要求不高。仿真实验验证了算法具有较高的鲁棒性和识别正确率,表明该算法具有一定的工程应用价值。

Abstract:

A novel approach of blind identifying of binary linear block code based on association rules is proposed. This method makes full use of the correspond relationship between each information bits to its parity check bits of the binary block code. If the inner correspond relationship of a binary sequence can be found by association rules, the linear block code of the sequence is correctly recognized. Firstly, the binary sequence is arranged to a matrix, then the matrix is divided into two parts. Therefore, each row of the matrix are becoming two items and two itemsets of the matrix are obtained. After that, the support and confidence of each itemset are calculated. According to the characteristics of the relationship of each information bits to its parity check bits of the binary block code, the linear block code is identified by traversing the way of finding the maximum confidence and the minimum mapping types. As only two parts of the matrix is concerned, the computing complexity of the association rule is degrade greatly. Compared with other mehods, the association method is suitable for low error rate of the sequence. Simulation results show that the proposed method has some strong points as high robustness,effectiveness and high accurate recognition, which indicate that the method has a certain value in future engineering application.