CAJ | 학술논문

针对在 KAD 网络中核心节点的识别问题,提出了一种基于 BP 模型对节点重要程度进行实时判定的方法.结合KAD 网络测量的结果,对网络中核心节点的属性特征进行提取和归一化处理,获得了一组可分离度较高特征集合.采用MatLab 设计相应的学习算法对 BP 网络进行训练,使结果收敛于预定误差区间.将完成训练的 BP 网络模型应用于对测试节点的判定,实验结果表明该方法可以实时地完成核心节点的判定,并且识别准确率可达到约70%.
침대재 KAD 망락중핵심절점적식별문제,제출료일충기우 BP 모형대절점중요정도진행실시판정적방법.결합KAD 망락측량적결과,대망락중핵심절점적속성특정진행제취화귀일화처리,획득료일조가분리도교고특정집합.채용MatLab 설계상응적학습산법대 BP 망락진행훈련,사결과수렴우예정오차구간.장완성훈련적 BP 망락모형응용우대측시절점적판정,실험결과표명해방법가이실시지완성핵심절점적판정,병차식별준학솔가체도약70%.
@@@@In view of the core node recognition in the KAD(Kademlia), a model based on BP is presented to determine whether a node is core node in real time. According to the result of the measurement for KAD, some attribute characteristics extraction and normalization processing is implemented to obtain an attribute set with higher separable degree. An algorithm in MatLab is designed to train the BP network until the results limit in a predetermined error range. In addition, the trained BP model is adapt to test prepared data, the results of the experiment show that the method can judge a node degrees of importance, and the recog-nition accuracy rate is up to about 70%.