Trust-based resource allocation and task splitting in ultra-dense mobile edge computing network

The number of mobile users are rapidly growing as 5G technology advances. Although technologies like Device-to-Device (D2D) communication related mobile edge computing (MEC) is helping to share the burden of edge server but privacy of data is a great perturb. This paper presents resource optimization by considering trust factor to meet the latency and energy usage requirements. The proposed solution is divided into two sub-parts. In first part resource allocation done based on the trust factor. In the second part, task splitting among resources has been optimized. The Stanford Large Network Dataset Collection (SLNDC) is used to analyze the system's performance for dynamic networks. The computational findings show that minimizing edge server (ES) usage, the suggested approach performs better. It's clear from the simulation results that the proposed solution Trusted Resource Allocation and Task Splitting Optimization (TRATSO) without ES, gives minimum energy consumption. The energy usage is reduced by 85.99%, 93.89%, 83.06%, 37.80%, 47.04%, 83.83%, 28.41% and latency is reduced by 70.09%, 64.70%, 61.82%, 88.92%, 89.22%, 73.58%, 21.58% as compared to Random Task Splitting (RTS) without ES, RTS with ES, Pattern Search Task Splitting (PSTS) without ES, PSTS with ES, GA Multi-objective (GAMO) without ES, GAMO with ES and TRATSO with ES respectively. CPU utilizations and throughput are also calculated for the same.