Towards ad-hoc cloud based approach for mobile intrusion detection

As the usage of smart devices is increasing, malware affecting such devices is rapidly evolving as well. Security risks affecting the confidentiality, integrity, and privacy of smart devices are rapidly emerging. Mobile security suites exist to defend the device against malware and other intrusions. However, they require extensive resources which is a constraint of the device itself. In this paper, we address the problem of intrusion detection for smart devices taking into account the devices' limited resources such as energy, CPU usage, and internet connectivity. We provide an ad-hoc mobile cloud based intrusion detection framework that takes advantage of Wi-Fi Direct as means of achieving connectivity and sharing resources and services for providing security. The proposed framework allows exchange of data with or without the availability of internet connection. The paper also provides experiments, carried out using real devices, showing various improvements of using our approach. Up to 61% and 40% enhancement in energy consumption and response time respectively is reached compared to local execution.