Profit Optimization for Mobile Edge Computing using Genetic Algorithm

The mobile edge computing has been widely recognized as a key enabler for new latency-sensitive applications and services on resource starved mobile terminals. The idea to offload a computationally intensive task to cloud has been extensively researched since the last decade. These are generally aimed at optimizing system energy consumption or latency reduction. In this paper we attempt to examine the profitability of computation offloading from the perspective of a network operator. The offloading decisions and joint optimization of radio and computational resources result in a mixed integer nonlinear optimization problem which is NP hard. To tackle this challenge, we decouple the offloading decisions from the radio and computational resource allocation. Firstly, the offloading decision is arrived at using a heuristic based genetic algorithm. It then goes as input to resource allocation optimization problem. The proposed genetic algorithm outperforms spectrum efficiency based offloading algorithm as per the simulations performed.

Automated summary

Mobile edge computing is crucial for new latency-sensitive applications and services, focusing on computation offloading and system optimization. Research shows that using a heuristic genetic algorithm can effectively improve offloading performance to solve NP-hard problems.