Multi-Objective Distributed On-Demand Small Cell Resource Allocation for eHealth

Small cell (SC) resource allocation for the next-generation cellular networks embraces ultra-low latency, energy efficiency, and reliable challenges. Conventional optimization algorithms may not be capable of supporting the abovementioned scenarios, with aggregated and centralized traffic burden causing excessive latency, especially for the conceived large-scale eHealth networks in Healthcare 4.0. In this paper, we propose a new Decentralized Integer-based Non-Dominated Sorting Genetic Algorithm (DI-NSGA), on top of the authors' previous work. Integer-based resource allocation process are formulated, and decentralized to mobile edge computing embedded SCs for releasing centralized traffic burden. Overall latency and achieved data rate are considered as the optimization objectives. Simulation analysis shows that the proposed DI-NSGA achieves low computation cost while maintaining high optimality by searching for the Pareto Front, compared with the selected benchmarks.