A Learning Automata based clustering Approach for underwater Sensor Networks to Reduce Energy Consumption

Wireless sensor networks that are used to monitor a special environment, are formed from a large number of sensor nodes. The role of these sensors is to sense special parameters from ambient and to make connection. In these networks, the most important challenge is the management of energy usage. Clustering is one of the methods that are broadly used to face this challenge. In this paper a new distributed clustering protocol based on learning automata is proposed for underwater wireless sensor networks. The proposed algorithm that is called LA-Clustering forms clusters in the same energy level, based on the energy level of nodes and the connection radius. regardless of size and the structure of sensor network. The proposed algorithm is compared with Leach, Leachswdn, Leachdcsh, and Heed algorithms. In comparison with these protocols, the superiority of this algorithm in terms of prolonging the lifetime of the network and keeping more number of alive nodes and also increasing the number of transmitted data is proved.

Keywords:

Underwater Sensor Networks , Clustering , Learning automata , Energy Consumption