语义相似度融合检错码跨层链路分流算法

Cross Layer Link Diffluence Algorithm Based on Error Detecting Codes Semantic Similarity Fusion

传统的检错码链路分流算法采用统计分析方法确定加密流量所属的具体应用协议,进行本层链路分流,当在链路层出现未加密数据时,性能不好。提出一种基于语义相似度融合检错码跨层链路分流算法。设计链路建立体系结构与检错码领域本体结构模型,为使分流后的链路特征向量的误差平方和最小,构建特征空间互信息区域语义相似度向量模型,计算语义相似度权值的微调参数,获得链路载波频率,采用检错码编码理论对链路载波频率进行冗余编码,提高语义相似度的融合性能。实现检错码跨层链路分流改进。仿真结果表明,采用该算法,链路建立过程的建立时间最短,有效避免了避免路由冲突,缩短了网络协议识别时间,有效提高检错码的抗干扰能力。

The traditional error detection code link distribution algorithm using the method of statistical analysis to determine the specific application protocol encryption flow belongs to the link layer, shunt, when unencrypted data in data link layer, the performance is not good. This paper proposes a semantic similarity fusion error detecting codes split algorithm based on cross layer link. Design of link establishment system structure and the error detection code domain ontology structure model, in order to make the link after the distribution to the feature vector of error square and minimum mutual information area, semantic similarity vector model construction feature space, to fine tune the parameters of semantic similarity calculation weights, obtain the link carrier frequency, using an error detection code theory of redundant code to link carrier frequency to improve the performance of fusion, semantic similarity. Implementation of cross layer link shunt improved error detection code. The simulation results show that, the link establishment process is established and the shortest time, it can effectively avoid routing conflict, shorten the network protocol identification time, effectively improve the error detecting code anti-interference ability.