A Study on Mobile Cloud Anomaly Detection Model

Abstract

The mobile cloud service is inevitable alternative which provides mobile device users with high computing power and large storage. The cloud servers provide various services such as IaaS, PaaS and SaaS to mobile device users seamlessly and the situation means that the device's apps or mobile browser can interact with the VMs in clouds at anytime and anyplace. In this mobile cloud environment, our concern is the abnormal access of cloud resources. Since the users of cloud services pay based on the usage of the cloud resources, if there are a certain kind of malicious access to the cloud resources, it can cause financial damages to mobile cloud users. Especially, the mobile devices are always turned on, even if we are moving around the street. Thus the malicious behavior of mobile device can do more harm to users compared to PC-based cloud services. This research suggests an anomaly detection model in mobile cloud services. In our suggested model, the users' behavior and resource usages are considered and the model contains the usual behaviors and resource usages as the users' profile which is supposed to be compared with the current users' behavior for detecting the abnormalities. The research result gives insight to security managers about monitoring factors and recognizing the anomaly in the mobile cloud service.