A Tree-Based Clustering Construction Algorithm (TBCCA) in Wireless Sensor Network

Aimed at the disadvantages on current clustered and topology control protocols in distributed wireless sensor network (WSN), such as high energy cost, non-balanced communication overhead, unsatisfactory computational complexity and extra consumption, etc., a novel tree based clustering construction algorithm (TBCCA) is presented. Firstly, it provides some premises and definitions about topologic model such as area radius, cluster radius and reachable adjacent domain, etc. Based on isosceles triangle clustered-tree structure, it proposes clustering strategy according to three types of threshold sets called near set, common set and medium set, and cluster radius controlled by RSSI (received signal strength indicator) value between neighboring nodes. Detailed procedures which contain cluster head selection and determination of candidate nodes are described in this paper, where some theorems about topology & coverage are in proof, and several optional strategies are provided in terms of different requirements on price of computation or communication. Performance analysis and simulation results illustrate that TBCCA has advantage over some existing algorithms in computational expense, for instance, TopDisc or DLMST; And compared with Leach protocol and HEED protocol, TBCCA is energy-efficient and energy-balanced, and expands the life-time of network while lower complexity, higher coverage and connectivity is guaranteed.