A GRASP based algorithm for efficient cluster formation in wireless sensor networks

In wireless sensors networks (WSNs) the efficient use of the sensors´ energy is a key point to extend the network lifetime and has been the center of attention by many researchers. There are several different techniques to reduce the sensor´s energy consumption. Sensor node clustering is one of these techniques. However, finding an optimal clustering in WSNs is a NP-Hard problem, thus efficient heuristics are needed to find good clustering in reasonable time. In this work we propose and analyze a Greedy Randomized Adaptative Search Procedure (GRASP) coupled with Path Relinking intensification to obtain near-optimal clustering. We develop a clustering protocol to simulate the clusters formation and data transmission. The good performance of our protocol is compared with the well known and commonly used clustering protocol for WSNs, LEACH and LEACH-C. The obtained results reveals that the GRASP algorithm determines better clusters extending the network lifetime.