Energy and Delay Efficient Dynamic Cluster Formation Using Improved Ant Colony Optimization Algorithm in EAACK MANETs

摘要

A MANET is a wireless ad hoc network with neither a fixed infrastructure nor an administrator to monitor the network operations. Clustering the network helps manage the network better. MANETs are networks that are ad hoc in nature. This nature gives rise to issues like loss of packets, security etc. which needs to be addressed in order to make the network more efficient. One major problem in clustering is the delay involved in cluster formation and in the selection of a suitable cluster head. We have proposed an improved version of the ant colony algorithm that employs two strategies to deduce the movement of the nodes and use this information to reduce the overheads in communication. The first phase helps to dynamically determine the heuristic parameters of the network in order to select the appropriate nodes for the cluster. The second phase helps form clusters faster and selects the cluster head without any delay. A distinct dynamic broadcast algorithm is employed to transmit the node status throughout the network. This algorithm also helps the network to sustain the changes in the network with no hindrance to data transmission. This technique helps in reducing the communication overheads and also improves the efficiency of transmission in the network. The proposed algorithm exploits the features of the ant colony algorithm and improvises it to achieve maximum delivery of packets with minimal delay possible. The proposed algorithm also focuses on reducing the time delay usually associated with cluster formation and cluster head selection. The node information helps in reducing this delay and helps the network focus on data transmission. The advantage of this algorithm is that the cluster head is determined based on the node movement and is immediately selected when the cluster is formed.