Game theoretical approach for utility-based distributed balanced data routing in wireless sensor networks

Abstract

In this thesis, two distributed algorithms for the construction of load balanced routing trees in wireless sensor networks are proposed. In such networks load balanced data routing and aggregation can considerably decrease uneven energy consumption among sensor nodes and prolong network lifetime. The proposed algorithms achieve load balancing by adjusting the number of children among parents as much as possible. The solution is based on game theoretical approach, where child adjustment is considered as a game between parents and child nodes, in which parents arc cooperative and children are selfish players. The gained utility by each node is determined through utility functions defined per role. Utility functions determine the behavior of nodes in each role. At the game termination, each individual node gains the maximum benefit based on its utility function, and the network reaches the global goal of forming the balanced tree. The proposed methods are called Utility Driven Balanced Communication (UDBC) algorithm which is designed for homogenous environment, where all nodes are assumed to produce equal amount of information, and Heterogenous Balanced Data Routing (HBDR) algorithm which is proposed for heterogenous environment, where different applications use different aggregation functions, and nodes can be vary in terms of the amount of produced information, energy levels, data transmission rate and available of the amount of produced information, energy levels, data transmission rate and available bandwidth for transmission. The advantage of this work over similar work in the literature is the construction of more balanced trees which results in prolonging network lifetime, with the capability of adaption according to specific application needs for sensitivity to delay and reliability of data delivery.