Message-Passing Based User Association and Bandwidth Allocation in HetNets with Wireless Backhaul

This work presents a joint design of user association and resource allocation in a heterogeneous network, which is comprised of a single macro base station and a group of small base stations interconnected through wireless backhaul. In such a configuration, we optimize user association and resource allocation so that the total sum of generalized utilities is maximized. This problem is cast as a combinatorial formulation with a fractional objective. To handle this design challenge, we develop a novel message-passing framework to obtain an efficient joint autonomous solution for user association and resource allocation. The simulation results show that the proposed algorithm outperforms existing techniques with various network utility functions.