Intervention in Power Control Games With Selfish Users

Abstract

We study the power control problem in wireless ad hoc networks with selfish users. Without incentive schemes, selfish users tend to transmit at their maximum power levels, causing significant interference to each other. In this paper, we study a class of incentive schemes based on intervention to induce selfish users to transmit at desired power levels. An intervention scheme can be implemented by introducing an intervention device that can monitor the power levels of users and then transmit power to cause interference to users. We mainly consider first-order intervention rules based on individual transmit powers. We derive conditions on design parameters and the intervention capability to achieve a desired outcome as a (unique) Nash equilibrium and propose a dynamic adjustment process that the designer can use to guide users and the intervention device to the desired outcome. The effect of using intervention rules based on aggregate receive power is also analyzed. Our results show that with perfect monitoring intervention schemes can be designed to achieve any positive power profile while using interference from the intervention device only as a threat. We also analyze the case of imperfect monitoring and show that a performance loss can occur. Lastly, simulation results are presented to illustrate the performance improvement from using intervention rules and compare the performances of different intervention rules.Comment: 33 pages, 6 figure

Repeated resource sharing among selfish players with imperfect binary feedback

“…[3]). Recently, a new class of resource sharing policies based on "intervention" [12] were proposed in 1 MAC games [14] and power control games [15].…”

“…However, the above resource sharing policies [2]- [15], designed under one-shot game models, neglect an important feature of many resource sharing systems: the repeated interaction among the players. Specifically, the (resource sharing) policies in [2]- [15] assume that the players interact only once, and require the players to consume resources at constant levels over the time horizon in which they interact 2 .…”

“…Specifically, the (resource sharing) policies in [2]- [15] assume that the players interact only once, and require the players to consume resources at constant levels over the time horizon in which they interact 2 . We call such policies in [2]- [15] static policies.…”

“…Specifically, the (resource sharing) policies in [2]- [15] assume that the players interact only once, and require the players to consume resources at constant levels over the time horizon in which they interact 2 . We call such policies in [2]- [15] static policies. When the negative externality among the players is strong, we can significantly improve static policies by using dynamic policies, which allow players to choose time-varying resource usage levels.…”

“…In this paper, we design dynamic resource sharing policies to achieve Pareto optimal equilibrium outcomes that are not achievable by existing static policies [2]- [15]. We provide a systematic design approach, which first characterizes the set of Pareto optimal equilibrium outcomes, and then achieves any outcome in this set by a deviation-proof policy, which is a policy followed by the players in their self-interests.…”