Mobile Cloud Computing is an emerging paradigm that concatenates Cloud Computing and Mobile Computing domains where repository and concoct of data occur externally and results are sent back to mobile devices. In earlier research, offloading frameworks requires special compilation, modification in applications or binary executable to be modified during offloading process. These frameworks mainly focused on how to do offloading worthwhile and not concentrated on how Virtual Machines are allocated to execute offloaded task in cloud environment which is a major countenance for authentic-time applications. In this study, a Decentralized Framework for Offloading Mobile Applications to Clouds (DFOMAC) is developed with the utilization of Adaptive Cuckoo Search Optimization Algorithm (ACSOA) and Neuro-Fuzzy Controller (NFC) to minimize energy and time consumption of mobile devices. An offloading decision model is developed using Neuro Fuzzy for predicting the execution environment of a task either at mobile or at the cloud. The decentralized resource allocation of Virtual Machines (VMs) in the cloud is carried out with the utilization of Adaptive Cuckoo Search Optimization Algorithm (ACSOA). The proposed novel framework is implemented in real-time environment using Android platform and battery performance of mobile devices are evaluated using Accubattery application. After, the planned process is compared with the existing method like Support Vector Machine (SVM) and Artificial Neural Network (ANN) methods and shows better energy efficiency as compared to existing methods.
Volume 12 | 07-Special Issue
Pages: 1834-1846
DOI: 10.5373/JARDCS/V12SP7/20202296