An Optimal Clustering Formation for Wireless Sensor Network Based on Compact Genetic Algorithm

Challenging task in many practical optimization applications arises from the restricted available hardware devices due to cost or space such as the sensor nodes in wireless sensor networks. This paper proposes a novel hierarchical clustering approach for wireless sensor networks to maintain energy depletion of the network in minimum using compact Genetic Algorithm. In this proposed method, new candidate solutions can be generated by learning explicit probabilistic models of promising solutions found so far and sampling the built models. The actual population is represented as a probability distribution over the set of solutions and is operationally equivalent to the order-one behavior of the GA. Simulation results compared with the original and the other methods in the literature as LEACH, LEACH-C, and HEED show that the proposed method provides the effective way of using a limited memory and the better performance in terms of the average residual energy, number of nodes alive, and number of received items to save the energy of nodes.