An Global Uneven Clustering Protocol Based on Collision Decreasing in Environment Integrated Surveillance

In the application of environment integrated surveillance, sensors are different in many aspects, such as category, initial energy and energy consume rate, etc. A lot of clustering protocols are proposed to distribute the energy consumption among nodes in each cluster and extend the network lifetime. However, most of them are not completely fit for environment integrated surveillance sensor network. To solve the problem, we proposed a distributed clustering protocol GUCA-CD which extends two classical clustering protocols LEACH and EEUC. Each node has a probability of becoming a tentative cluster head (TCH) to compete for final cluster heads (FCH) in each round. The probability depends on the number of consecutive rounds in which a node has not been FCH, node´s residual energy and distance to the Sink. In order to decrease the collision of TCHs, our protocol produces clusters from far to near with global uneven radius. Simulations show that our protocol can balance the energy consumption well among all sensor nodes and achieve an obvious improvement on the network lifetime.