Bayesian Coalition Game-based optimized clustering in Wireless Sensor Networks

Wireless Sensor Networks (WSNs) have gained a lot of popularity in recent years because these are being used in wide range of applications. A collection of randomly/planned deployed tiny Sensor Nodes (SNs) can perform the task according to the need of a specific application. Utilization of energy of SNs is one of the key issues in these networks as nodes are battery operated and recharge or replacement of battery is a difficult task to be achieved. To address this issue, we propose a Bayesian Coalition Game-based optimized clustering in WSNs. To formulate the game, we propose a new Hybrid Homogeneous LEACH (HHO-LEACH) protocol for SNs in WSNs. We have used the concepts of Learning Automata (LA), and Bayesian Coalition Game (BCG) in which SNs are assumed as the players in the game with dynamic thresholds-based coalition formation among themselves, i.e., coalition among the nodes are formed using distance-based thresholds which makes a partition of the network field. SNs near to base station use direct communication for data transfer with or without single hop to the Base Station(BS) after interacting with the environment. During this process, each player may get a reward, or a penalty with respect to the finite number of actions performed. Performance of the proposed protocol is evaluated using extensive simulations by selecting various evaluation metrics. The results obtained show that proposed coalition game achieved better stability, and network lifetime in comparison to other existing protocols such as LEACH, and DD.