Performance Analysis of Task Offloading in Mobile Edge Computing using Multi-Objective Optimization Algorithm

Mobile edge computing plays a vital role in emerging communication technologies; Edge computing is known for its low latency. Task are offloaded from the mobile users to the servers. For processing the data, edge server or cloud server can be utilized, but cloud server will control the entire operation in edge system. Edge servers are associated with micro base station whereas cloud servers are with macro base station. Task offloading is performed on either micro or macro-base station depending upon the traffic at the servers. Data processing at the servers can lead to computational complexity and also increase in overall cost of the system. By lowering computing complexity and time, the suggested technique would increase system welfare and lower the cost of edge systems. Particle swarm optimization (PSO) is used to maximize the system welfare and minimize the cost of the system. In this optimization tool, local best and global best particles are calculated for the objective function and optimized value is chosen from the global best particle. It is observed that PSO algorithm is computationally inexpensive and effective. The proposed approach is evaluated against a non-dominated sorting genetics algorithm (NSGA-II).