Base Station Allocation for Users with Overlapping Coverage in Wirelessly Networked Disaster Areas
Yu Wang, Michael Conrad Meyer, Junbo Wang · 2019
After major disasters, temporary deployable cellular networks are often used to construct an emergency communication network. These networks do not have the same level of performance as a typical fog or cloud network, and in order to serve the users in a way that is sustainable, utilization of optimization algorithms must be considered. We proposed using a genetic algorithm (GA) to optimally allocate these users in the overlapping areas to a base station so that the system could provide an improved user experience. We tested our proposed algorithm against a greedy algorithm, a random algorithm, and allocating users to the closest MBS. The greedy algorithm outperformed the two other baseline algorithms, but the proposed algorithm was able to reduce the average and worst-case delay of the system by 80% compared to the greedy algorithm. The genetic algorithm had completely settled by 150 generations. This algorithm provided such an advantage that it warrants deeper study.