Distributed Resource Allocation for D2D Communications Underlaying Cellular Networks in Time-Varying Environment

In this letter, we address joint channel and power allocation in a device-to-device (D2D) network underlaying a cellular network in a time-varying environment. A fully distributed solution, which does not require information exchange, is proposed to allocate channel and power levels to D2D pairs while ensuring the quality of service (QoS) of the cellular user equipments (CUEs). The problem is modeled as a Stackelberg game with pricing. At the leader level, base station sets prices for the channels to ensure the QoS of the CUEs. At the follower level, D2D pairs use an uncoupled stochastic learning algorithm to learn the channel indices and power levels while minimizing the weighted aggregate interference and the price paid. The follower game is shown to be an ordinal potential game. We perform simulations to study the convergence of the algorithm.