Matching-Based Resource Allocation for Critical MTC in Massive MIMO LTE Networks

Supporting critical Machine-Type Communications (MTC) in addition to Human-Type Communications (HTC) is a major target for LTE networks to fulfill the 5G requirements. However, guaranteeing a stringent Quality-of-Service (QoS) for MTC, in terms of latency and reliability, while not sacrificing that of HTC is a challenging task from the radio resource management perspective. In this paper, we optimize the resource allocation process through exploiting the additional degrees of freedom introduced by massive Multiple-Input Multiple-Output (MIMO) techniques. We utilize the effective bandwidth and effective capacity concepts to provide statistical guarantees for the QoS, in terms of probability of delay-bound violation, of critical MTC in a cross-layer design manner. In addition, we employ the matching theory to solve the formulated combinatorial problem with much lower computational complexity compared to that of the global optimal solution so that the proposed scheme can be used in practice. In this regard, we analyze the computational complexity of the proposed algorithms and prove their convergence, stability and optimality. The results of extensive simulations that we performed show the ability of the proposed matching-based scheme to satisfy the strict QoS requirements of critical MTC with no impact on those of HTC. In addition, the results show a close-to-global optimal performance while outperforming other algorithms that belong to different scheduling strategies in terms of the adopted performance indicators.