Pricing and Rate Optimization of Cloud Radio Access Network Using Robust Hierarchical Dynamic Game

We develop a robust one-leader multi-follower dynamic linear quadratic game for 5G wireless networks with a cloud radio access network setting. The network consists of a baseband unit (BBU) owner and multiple remote radio heads (RRHs). The users' traffic from the RRH is routed to BBUs based on their price and congestion which is modeled as a dynamic difference equation, considering the effect of decisions taken by the RRHs, price, and uncertainties. In the proposed dynamic game, the BBU owner is the leader and the RRHs are the followers. The uncertainties raised from compression noise and link imperfection, such as packet loss, is modeled as an exogenous disturbance to the system. The equilibrium point of the proposed robust hierarchical dynamic game is achieved, and the conditions for the existence and uniqueness of the equilibrium point are provided in two cases, including unconstrained problem and also by considering the fronthaul constraints. Finally, the simulation results show the advantages in terms of cumulative payoff compared with other existing baseline schemes.