Error-Exploiting Video Encoder to Extend Energy/QoS Tradeoffs for Mobile Embedded Systems

Abstract

Energy/QoS provisioning is a challenging task for video applications in power-constrained mobile embedded systems. Many error-resilient video encodings allow us to exploit errors and generate a range of acceptable tradeoff spaces by controlling the amount of errors in the system. This expanded tradeoff space allows system designers to comparatively evaluate different operating points with varying QoS and energy consumption by aggressively exploiting error-resilience attributes, and can potentially result in significant energy savings. Specifically, we propose an error-aware video encoding technique that intentionally injects errors (drops frames) while ensuring QoS in accordance with error-resilience. The novelty of our approach is in active exploitation of errorsto vary the operating conditions for further optimization of system aspects. Our experiments show that our error-exploiting video encoding can reduce the energy consumption for an encoding device by 37% in video conferencing over a wireless network, without video quality degradation, compared to a standard video encoding technique for a test video stream. Furthermore, we present the adaptivity of our approach by incorporating the feedback from the decoding side to achieve the QoS requirement under dynamic network status.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.