Overview on HEVC Inter Frame Video Coding’s Impact on the Energy Consumption for Next Generation WVSNs

Abstract

With the advent of the High Efficiency Video Coding HEVC standard, wireless transmission of video data consumes more and more energy, a major concern in the field of Wireless Video Sensor Networks (WVSNs). The energy resources are limited, consisting only in the battery of the sensor nodes that determines their lifetime. In this paper, we propose an empirical parametric model to predict the energy consumption of an HEVC based video encoder in its inter prediction mode, used in the context of the next generation WVSNs. Such a model is of great interest to minimize the waste of energy of the encoding phase, while meeting the required video quality. The proposed model predicts energy consumption, considering the adopted Number of P frames (NP). A Raspberry Pi 2 card based video sensor node is used for modelling and validation, considering different configurations. The obtained results demonstrate that the proposed model describes well the occurred energy dissipation during the video encoding phase, with an average prediction error of 1.6%.