Multi-Metric Clusterhead Selection Using Classification In Wireless Sensor Networks

Multi-metric clusterhead selection is a multidimensional problem in wireless networks whose optimum solution cannot be found in real time. In this paper, we design an approximation algorithm called MMCSC for this problem using SOM classification techniques. SOM (Self Organizing Map) converts the multidimensional problem into a one-dimensional problem, thus makes it fast to solve. MMCSC considers multiple metrics in clusterhead selection including remaining energy, number of neighbors, and distance to sink. Our evaluations show that MMCSC surpasses the existing algorithms in terms of shorter execution duration, higher remaining energy of clusterheads, and achieving unequal clustering.