Optimizing Multi-Objective Computation Offloading In Heterogenous Environments Using Adaptive Offloading Cat Hunt Optimization Algorithm

Abstract

The advent of smart mobile devices has ushered in a new era of computing, but their performance is inherently constrained by factors like processing power and battery capacity. To optimize task execution, it is critical to strike a balance between tasks executed on the devices and those offloaded for remote processing, as proper offloading greatly enhances the quality of service. Prevailing techniques often emphasize on single objectives and make computationally complex, lacking a universal approach that balances targets and complexity effectively. To tackle these issues, this research proposes an Adaptive Offloading Cat Hunt Optimization (AOCHO) algorithm, which is designed to optimize computation offloading in mobile edge computing, with a primary focus on minimizing time, energy consumption and resource utilization. Primarily, this research starts by formulating the problem using Directed Acyclic Graphs (DAGs) in heterogeneous environments, aiming to reduce energy consumption for mobile users. Subsequently, the AOCHO-based offloading algorithm tackles multi-objective problems. The experiments conducted in the MATLAB environment, yield superior results. The simulation demonstrates a substantial reduction in delay by 0.0172 sec, a decrease in energy consumption by 0.251 (10^-3 J), and a cost reduction of 0.387. These results clearly reveal that the proposed algorithm surpasses other benchmark algorithms in various situations. This underscores the algorithm's effectiveness in enhancing offloading efficiency for mobile devices in the realm of mobile edge computing.