Online Measurement-Based Adaptive Scalable Video Transmission in Energy Harvesting Aided Wireless Systems

We consider the point-to-point adaptive transmission of scalable video coding streams over a fading channel in energy harvesting (EH) aided communication systems, where the transmitter is equipped with an EH device and a rechargeable battery. The aim of this paper is to enable the transmitter having the ability of adaptively adjusting the number of video layers to be transmitted according to the available energy. Since the depletion of the energy in the battery acts as a trigger for the interruption of the video transmission, we define the energy starvation probability for equivalently characterizing a quality-of-experience (QoE) metric. We formulate the problem of EH-aided video transmission as a constrained optimization problem, maximizing the number of video layers for transmission while keeping the energy starvation probability below a threshold. Classic large deviation theory is invoked for estimating the energy starvation probability from online measurements. Consequently, our dynamic layer switch algorithm operates with no prior statistical knowledge of both the EH process and of the channel quality. Furthermore, the technique of perturbation analysis that constructs a perturbed sample path from an original sample path is invoked for improving the performance of the proposed method. Our experimental results verify that the algorithms proposed have the adaptation capability to accommodate both the energy-dynamics and the channel-dynamics for improving the video quality subject to an unobjectionable level of transmission interruption rate.