Secure Data Aggregation with Mean Field Extensive Game Theoretic Framework

Abstract

In this paper, game theoretic framework is used as a mathematical tool to address security problems in mobile wireless sensor network (MWSN). In most of the existing works, only attacker and a defender are considered as the two players of the game. With the use of recent advancements made in the mean field game theory, we have proposed a novel extensive game with multiple players for MWSN security, such that the aggregator node can securely compute aggregation data even in the presence of an attack. The proposed scheme enables the aggregator node and an individual node in MWSN to strategically make defense decisions and hence, the confidential data could be faithfully transmitted to base station (BS) for security. Also, each node in this dynamic distributed network, knows the information about its own state and the information about the other nodes’ aggregate effect in the MWSN. Optimal strategic equilibrium solution to our proposed mean field extensive game is given, such that the utility of each player is maximized in the game. Extensive game analysis and study shows that our proposed dynamic algorithm strategically outperforms existing static approach.