An efficient computational offloading framework using HAA optimization-based deep reinforcement learning in edge-based cloud computing architecture

Mobile Cloud Computing (MCC) has emerged as a popular model for bringing the benefits of cloud computing to the proximity of mobile devices. MCC's preliminary goal is to improve service availability as well as performance and mobility features. Edge computing, a novel paradigm, provides rich computing capabilities at the edge of pervasive radio access networks close to users. Designing an efficient offloading technique for edge computing is a major research challenge given the constrained resources. Offloading speeds up processing, which has an impact on service quality in heterogeneous devices. Due to the difficulties of the network states' distribution environment, allocating computing resources is a difficult process. In this study, inputs from various sorts of devices such as cars, mobile phones, and heterogeneous building sources are considered. When an accurate energy estimation model is established to compute the energy consumption of the tasks during offloading, an effective task offloading technique can be derived. The model should select whether or not to conduct offloading based on the computed energy cost. This study proposed a Hybrid Arithmetic Archimedes Optimization algorithm-based Deep Reinforcement Learning model for computation offloading in heterogeneous devices. The effectiveness of the proposed method is evaluated using different state-of-art methods such as First Upload Round and Second Upload Round (FUR-SUR), Efficient Dynamic-Decision Based Task Scheduler, Context-aware computation offloading and Price-based distributed offloading. The proposed method offers superior results to other existing methods. The user's average utility of the proposed method increases by 410% contrasted to FUR-SUR.