Wang Xinfeng
ABSTRACT
Internet Digital Cannon is a visualized name of ZMW attack, in which a BGP connection is periodically reset. When a number of BGP connections are attacked, a great lot of BGP notification error messages would surge and Internet performance would be depressed badly. As for ZMW or CXPST (which is proposed after ZMW), importance of vital links is twofold: Vital links can give a range of important defense links and instruct network defense, and on the other hand vital links can also enhance attacking effect observably by partitioning a network into several subnets. Based on graph theory, this paper proposes a cost function and applies spectral clustering algorithms which are mainly used in machine learning into vital links search of network topology partitioning. Partition effect of single and multi-eigenvector spectral clustering algorithms are simulated and result show: Multi-eigenvector spectral clustering algorithms have a lower cost and can partition a network topology into any number of subnets, and single eigenvector spectral clustering algorithms can only partition a network topology into 2n subnets in the nth iteration. Results of this paper can be used in theory research and engineering project of network confronting.
Recommendations
Study on multi-center fuzzy C-means algorithm based on transitive closure and spectral clustering
Fuzzy C-means (FCM) clustering has been widely used successfully in many real-world applications. However, the FCM algorithm is sensitive to the initial prototypes, and it cannot handle non-traditional curved clusters. In this paper, a multi-center ...
Self-tuning clustering for high-dimensional data
Spectral clustering is an important component of clustering method, via tightly relying on the affinity matrix. However, conventional spectral clustering methods 1). equally treat each data point, so that easily affected by the outliers; 2). are ...
Ensemble learning using three-way density-sensitive spectral clustering
AbstractAs one popular clustering algorithm in the last few years, spectral clustering is advantageous over most existing clustering algorithms. Although spectral clustering can perform well in many instances, the algorithm still has some ...
Comments