Energy-Efficient HTTP Adaptive Streaming with Hybrid TCP/UDP Over Heterogeneous Wireless Networks

In this work, we propose an energy-efficient HTTP adaptive streaming with hybrid TCP/UDP over heterogeneous wireless networks. Unlike the existing HTTP adaptive streaming system, the proposed system can utilize both TCP and UDP to overcome the performance degradation problem of TCP and improve the energy efficiency. The proposed system handles the types of transport protocols, requested video quality, and amount of data received through each wireless network as control variables based on network conditions, buffer occupancy, and energy condition of mobile terminal. When UDP is selected, the proposed system employs Raptor code as a forward error correction scheme to support a reliable data delivery, and adaptively chooses the proper Raptor code parameters such as symbol size and amount of redundant data, and requests redundant data from the server to reduce the overhead. The proposed system is fully implemented in an Android-based mobile terminal and tested in real WiFi and LTE networks. The experiment results show that the proposed system can achieve good energy efficiency while providing a seamless high-quality HTTP adaptive streaming service.