Archives

Detection of High Accuracy Classifier and Most Influencing Offloading Parameters for Mobile Cloud Applications


V. Suganya and Dr.M. Kannan
Abstract

Due to significant development in the mobile software and hardware technology and high speed internet, the usage of mobile phone increased day by with desktop like user experience. Also, the applications targeted for these devices, are developed in parallel with complex and high power demanding resources. But mobile phones are limited in processing these resource hungry applications in terms of energy and processing. One solution to such a problem is to utilize the interdisciplinary paradigm called mobile cloud computing which is a combination of mobile devices and cloud computing. The computation offloading in mobile cloud computing provides a solution to enhance the battery power and processing capacity of the mobile phone by outsourcing the compute intensive part of the mobile application to highly resource rich cloud servers for execution and send the computed results back to the mobile phone. This offloading decision is taken by the decision engine in the mobile device. And these offloading decisions are highly based on profiling the environment, device and application parameter on the mobile device. This paper investigates and determines the optimal evaluation parameter with enhancing precautious solution and uses stacking ensemble method for improved accuracy rate in the offloading process. The result shows that the ensemble stacking method outperforms its existing base classifiers inaccuracy rate by 89.02 percent.963-

Volume 11 | 04-Special Issue

Pages: 963-972