Spectrum-aware mobile computing using cognitive networks

Abstract

With the advent of mobile cloud computing, the expectation of the mobile users for anywhere, anytime, content-rich experience will see a significant increase. The users' expectation on quality of experience for content-rich applications can only be met through offloading computationally intensive application tasks to a remote cloud since mobile devices are still constrained by their battery power. This, however, leads to an increase in mobile web traffic. The success of computation offloading techniques, therefore, depends on being able to effectively trade-off resource usage at the mobile device against efficiently managing the spectrum for mobile computing. Hence it is essential for cloud offloading techniques to take advantage of recent advances in cognitive network- ing and spectrum-aware scheduling of application components. The convergence of cognitive networking and spectrum-aware mobile computing is propelling research in this area. The current state-of-the-art includes techniques that offload application data using all viablemultiple radio interfaces (e.g.,WiFi, LTE, etc.) in multi-RAT-enabled devices, while being adaptive to the conditions of the mobile network. This chapter presents a survey of the existing spectrum-aware mobile computing techniques and proposes a vision for the future for a 5G-enabled, cognitive mobile computing platform. Implementation setups using real data measurements from an HTC phone running multicomponent applications and using different cloud servers such as Amazon EC2 and NSFCloud over LTE and WiFi are also discussed.