PERFORMANCE COMPARISON OF FCM AND K-MEAN CLUSTERING TECHNIQUE FOR WIRELESS SENSOR NETWORK IN TERMS OF COMMUNICATION OVERHEAD

Abstract

Wireless Sensor Networks are consists of several sensors which are connected through a gateway using wireless communication. These smart sensors are directly interacting with environmental condition such as temperature, pressure, humidity, vibration etc. and communicate these over the wireless network. Communication overhead is increase when any sensor node is fail. The protocol will follow multiple path to destination node. So energy consumption is more. Multiple paths are increased when sink node is mobile. In this paper we introduce cluster technique for making cluster in network to reduce communication overhead due to less connection with sink and make the network hierarchical. We use two technique k-mean and fuzzy c-mean algorithm for making cluster .k-mean is useful for stationary sink node and fuzzy c-mean is useful for mobile sink node due to fuzziness of nodes in cluster. In this paper we compare both algorithms.