Energy efficient secured K means based unequal fuzzy clustering algorithm for efficient reprogramming in wireless sensor networks

摘要

Wireless Sensor Networks (WSNs) is a collection of tiny distributed sensor nodes that have been used to sense the physical parameters of the environment where it has been deployed. Data dissemination is an important activity performed in WSNs in order to administer and manage them. Gossiping makes the network to transmit the same data item multiple times by multiple sensor nodes to their neighbors until they reach the required nodes which are in need of them. These multiple transmissions result in a problem called a Redundant Broadcast Storm Problem (RBSP). Moreover, the RBSP results in too many senders' problem and also leads to the consumption of more energy in the network. In data dissemination, providing energy efficiency and security are the two major challenging issues. In such a scenario, the attackers may make use of the weakness in security provisions available in the network and they can perform unauthorized activities to disrupt the process of data dissemination. Hence, it is necessary to address the issues of RBSP, energy consumption, security and too many senders problem in order to enhance the reliability and security of communication in WSN for data dissemination. In this paper, a novel protocol named Cluster based Secured Data dissemination Protocol (CSDP) has been proposed for providing energy efficient and secured dissemination of data. The proposed protocol is a distributed protocol which considers the route discovery process, cluster formation, cluster head selection, cluster based routing and security through the design of a new digital signature based authentication algorithm, trust based security enhancement and encryption techniques for effective key management. The major contributions of the proposed work include the proposal of cryptography based public key and private key generation algorithms, techniques for trust score computation and malicious node identification and finally the effective prevention of malicious activities for enhancing the security of the network. Moreover, this work considers node identification techniques for effective clustering of nodes and performs optimal route discovery and secured transmission of packets. This work is novel with respect to multicast based data dissemination protocol, proposal of combined signature generation and verification schemes, encryption based key management and distributed data collection and communication techniques. In addition, an Intelligent Fuzzy based Unequal Clustering algorithm is used to perform effective clustering process and the traffic analyzer to identify the intruders by monitoring the node's behaviors and their trust values. The proposed protocol has been extensively tested with realistic simulation parameters using NS2 simulator. The simulation results obtained from this work have proved that the proposed protocol improves the level of security through the proposal of a time efficient encryption and decryption algorithm with increase in packet delivery ratio and network throughput and at the same time it reduces the energy consumption as well as delay in data dissemination.