Optimal Frequency Band Allocation for Multi-Cell IoT Networks Over Unlicensed Spectrum

Optimal frequency band allocation for multi-cell Internet of Things (IoT) networks that operate on an unlicensed spectrum and adopt grant-free ALOHA random access is crucial for effectively supporting massive IoT devices but has yet to be closely studied. This paper systematically investigates optimal frequency band allocation under frequency reuse for a slotted multi-cell IoT network that operates on an unlicensed wide band with multiple frequency bands and adopts the slotted multi-band ALOHA scheme. Specifically, we consider three cases of frequency band allocation, i.e., adjustment cost-free case, adjustment cost-constrained case, and adjustment cost-penalized case, and respectively study the average rate maximization, the average rate maximization under a frequency band adjustment cost constraint, and the maximization of a weighted sum of the average rate and frequency band adjustment cost, which are challenging large-scale integer programming (IP) problems. Then, in each case, we propose an efficient algorithm with parallel updates per iteration and a low-complexity greedy algorithm with finite iterations to obtain optimal and approximate optimal points, respectively. We also analyze the convergence results of the proposed optimization algorithms and the differences and connections of the three problems. We compare the proposed methods with existing frequency band allocation methods. Analytical and numerical results offer design insight and validate the values of the proposed formulations and algorithms.