A dynamic clustering algorithm in wireless sensor networks

Abstract

It is essential to prolong the lifetime of wireless sensor networks (WSN) via effective cooperation of its sensor nodes. Here, a dynamic clustering algorithm, named DCA, is presented to optimally and dynamically select the micro-sensor nodes to construct a dynamic sensor cluster at each time based on the integrated performance index including information acquirement and energy consumption. In distributed target tracking with WSN, the DCA can avoid the problem of too frequent cluster head (CH) switches, save more than 80% energy and remain almost same tracking accuracy, compared with the information-driven sensor querying (IDSQ).