Massive Access for Machine-Type Communications in Backhaul-Constrained Heterogeneous Networks

Massive number of machine-type communications (MTC) devices may lead to access overloading. In heterogeneous networks, the load balancing between macrocells and small cells and the backhaul capacity limitation bring new technical challenges due to their mutual coupling. In this paper, we adopt access class barring (ACB) for access control and jointly optimize access control, load balancing and preamble allocation in backhaul-constrained heterogeneous networks. Since this joint optimization problem is intractable due to its combinatorial nature, we decompose the problem and solve it in two steps. First, by introducing an auxiliary variable and exploiting the monotonicity property of objective function, we derive a closedform solution for the optimal ACB parameter with the given cell load and number of preambles. Second, we propose an iterative algorithm to further optimize load balancing and preamble allocation with optimal ACB. Furthermore, we prove its convergence property by monotone convergence theorem. Simulation results show that the proposed massive access algorithm achieves significant performance gain compared to existing algorithms.