Game Theory-Based Wireless Resource Allocation for Full-Duplex Systems

Full-duplex systems enable simultaneous uplink and downlink transmission in wireless communications, significantly improving spectrum utilization efficiency. However, this concurrent transmission capability introduces challenges such as self-interference and inter-user interference, particularly the power allocation conflicts between uplink and downlink channels.

To optimize full-duplex system performance, the power allocation problem can be formulated as a joint uplink-downlink sum-rate maximization problem. Due to the non-convex nature of this optimization problem, traditional optimization methods struggle to find direct solutions. Therefore, game theory provides an effective analytical framework where the competitive relationship between uplink and downlink channels is modeled as a non-cooperative game. In implementation, this typically involves defining utility functions for each link that represent their transmission rates, with power levels as strategic variables.

In the non-cooperative game model, uplink and downlink channels act as independent players, each selecting optimal power allocation strategies to maximize their individual transmission rates. Through carefully designed iterative algorithms - such as best-response dynamics or water-filling based updates - the system gradually converges to an equilibrium state known as Nash Equilibrium. At this equilibrium point, neither link can unilaterally adjust its power strategy to improve its own utility, thereby achieving globally optimized resource allocation. The algorithm implementation typically involves iterative power updates where each link computes its optimal response given the current interference conditions from the other link.

This game theory-based power allocation approach not only effectively mitigates the impact of self-interference and inter-user interference but also achieves higher spectral efficiency and system throughput in full-duplex systems. The implementation can be enhanced with convergence acceleration techniques and practical constraints handling through barrier functions or projection methods in the optimization routine.