集群网络可组合信息流多阶段优化检测仿真

Cluster Network Combinable Information Flow Multi-Stage Optimization Detection Simulation

为了提高整个集群网络可组合信息应用系统的可靠性,需要对信息流进行多阶段优化检测,提出基于K-means聚类的集群网络可组合信息流多阶段优化检测方法。结合K-means算法的相关思想以及集群网络可组合信息流的相关特点,将欧式距离作为指标,比较不同数据流的相似度并划分聚类。在上述基础上,算法引用经过改进后的萤火虫算法(BGSO)对集成模型进行优化,获取最优子集。并使检测模型随着信息流的变化进行自适应更新,提高整体的检测准确性节省检测时间。实验结果验证了所提方法相比传统方法在各个方面都有了一定的改进,也充分证明了所提方法的有效性。

In order to improve the reliability of the combinable information application system,it is necessary to optimize and detect information flow from multiple stages.Therefore,a method of multi-stage optimization detection for combinable information flow method in cluster network based on K-means clustering was proposed.Combining the related ideas of K-means algorithm with the related characteristics of combinable information flow,the Euclidean distance was used as the index to compare the similarity of different data streams and partition clusters.On this basis,the binary glowworm swarm optimization(BGSO)was used to optimize the integrated model and get the best subset.Finally,the detection model was adaptively updated with the change of information flow,so that the overall detection accuracy was improved to save detection time.Simulation results show that the proposed method has some improvements in various aspects.The effectiveness of the proposed method can be fully proved.