J4 ›› 2015, Vol. 42 ›› Issue (3): 90-96.doi: 10.3969/j.issn.1001-2400.2015.03.016

• Original Articles • Previous Articles     Next Articles

Novel sensor node importance evaluation method based on the agglomeration contraction principle

LIU Kui1,2;LIU Sanyang2   

  1. (1. College of Mathematics and Information Science, Henan Normal Univ., Xinxiang  453002, China;
    2. School of Mathematics and Statistics, Xidian Univ., Xi'an  710071, China)
  • Received:2014-01-02 Online:2015-06-20 Published:2015-07-27
  • Contact: LIU Kui E-mail:liukui_1980@163.com

Abstract:

Node importance evaluation based on removal of nodes and their incident links cannot accurately reflect the importance of the nodes. Because this way may change the topology of wireless sensor networks, and even splits the network into several disconnected parts. To solve this question, we propose a novel node importance evaluation method based on the agglomeration contraction principle. This method does not need to remove the assessment node from the network, and thus this method may not cause the network to be fractured. In this method, with the help of a spectral method, the vital gateway nodes can be obtained by using the nontrivial eigenvectors of the Laplace matrix of the network diagram firstly. After that, by using some super-energy nodes to provide targeted protection for vital gateway nodes in a wireless sensor network, this way can prolong the life of the network and improve the robustness of the system effectively. Above all, this method can reduce the number of evaluation nodes in a wireless sensor network, and decrease the workload of calculation accordingly. Final examples verify its efficiency and the result is consistent with our intuitive judgments.

Key words: spectral method, clustering contraction principle, gateway node, wireless sensor networks

CLC Number: 

  • TN393