Joint user association and resource allocation in small cell networks with backhaul constraints

Heterogeneous networks potentially provide significant capacity gains by overlaying the traditional cellular network with a layer of small cells (SCs) served by access points (APs). However, the limited backhaul capacity of the SC APs, combined with increased interference from neighboring cells, necessitates careful resource allocation to realize the gains. In our work, we consider maximization of the weighted sum rate in small cell networks with carrier aggregation while enforcing a backhaul constraint on each SC AP. We propose an efficient, waterfilling-like, algorithm which converges to a locally optimal solution of the non-convex optimization problem. This algorithm differs from existing works by using a computationally efficient bisection-like search that ensures the sum-power and sum-rate constraints at each AP are satisfied via an one-dimensional search. An added advantage is that this approach also allows for a decentralized implementation.