Resource Reservation in Backhaul and Radio Access Network With Uncertain User Demands

Resource reservation is an essential step to enable wireless data networks support a wide range of user demands. In this paper, we consider the problem joint resource in backhaul and Radio Access Network (RAN) based on statistics demands channel states, also network availability. The goal maximize sum expected traffic flow rates, subject link access point budget constraints, while minimizing outage downlinks. formulated turns out be non-convex difficult solve global optimality. We propose efficient Block Coordinate Descent (BCD) algorithm approximately problem. proposed BCD optimizes capacity using novel multi-path routing that decomposes down link-level parallelizes computation across links, transmission resources RAN carried via scalable distributed Successive Upper-bound Minimization (BSUM). prove converges Karush-Kuhn-Tucker (KKT) solution. Simulation results verify efficiency efficacy our approach against two heuristic algorithms.