User Association in Cloud RANs with Massive MIMO

This paper studies a resource allocation problem where set of users within specific region is served by cloud radio access network (C-RAN) structure consisting base-band units (BBUs) connected to remote heads (RRHs) equipped with large number antennas via limited capacity front-haul links. User association each RRH, BBU and link essential achieve high rates for cell-edge under limitations. We introduce two types optimization variables formulate this problem: (i) C-RAN user factor (UAF) including (ii) power vector. The formulated non-convex computational complexity. An efficient two-level iterative approach proposed. higher level consists steps where, in step, one these fixed derive the other. At lower level, applying different transformations convexification techniques, step broken down into sequence geometric programming (GP) problems be solved successive convex approximation (SCA). Simulation results reveal effectiveness proposed increase total throughput network, specifically users. It outperforms traditional approach, which, first assigned RRH largest average value signal strength, then, based on association, are optimized.