Jamming games in wireless networks with incomplete information

Due to their broadcast nature, wireless networks are highly susceptible to jamming attacks resulting in denial of service. Game theory provides powerful tools to model and analyze such attacks. This article discusses a class of such jamming games played at the MAC layer among a set of transmitters and jammers. The equilibrium strategies resulting from these jamming games characterize the expected performance under DoS attacks and motivate robust network protocol design for secure wireless communications. A key characteristic of distributed wireless access networks is that users do not have complete information regarding the other user's identities, the traffic dynamics, the channel characteristics, or the costs and rewards of other users. Bayesian games are shown to be useful for modeling such uncertainties under different models for the MAC layer. Different wireless network scenarios with single-stage and repeated game formulations are described to illustrate the performance loss of jamming attacks with increasing network uncertainty.