Low Latency Live Streaming System with Congestion Control

In recent years, the traffic for live streaming on the web has been increasing. The current live streaming methods that use MPEG-DASH or HLS are simple and scale easily to many clients using HTTP. However, they do not take into account the communication between the distributor and the viewer. As a result, latency between the distributor and the viewer is relatively high. Therefore, in this paper, we propose a low latency live streaming system on the web using WebRTC. Since WebRTC uses UDP, it does not have a congestion control mechanism. Depending on the network congestion, it is possible to stream video with quality that exceeds the available bandwidth. Therefore, we propose a system to change the video quality based on the congestion status. The proposed system increases or decreases the video transfer rate by changing the quality of the streamed video depending on the network conditions. We have evaluated the proposed system in a real network environment. As a result, we showed that the delay of the proposed system is smaller than that of the MPEG-DASH system. We also showed that the proposed system can change the quality of the video and switch the transmission rate appropriately according to the network conditions.

Energy Efficient and High Dissemination Rate Method Considering Extended Transmission Distances on a Wireless Sensor Network

From the electric power consumed by sensor nodes comprising a wireless sensor network, a very large proportion is used to transmit and receive information wirelessly. Because most sensor nodes are battery-powered, long-term operation requires unnecessary wireless transmission to be minimized. There is currently insufficient research on cases involving increased wireless transmission distance and the accompanying increased power consumption. This paper proposes a new, effective method for information dissemination in situations with increased wireless transmission distances. The proposed method is able to minimize power consumption by not performing transmission if the message is considered already disseminated. Additionally, with increased wireless transmission distance, the number of instances of cancelled message transmission increases, further reducing the transmission and reception loads. Simulations demonstrated that the proposed method is effective at keeping power consumption increases to a minimum while greatly improving information dissemination rates, even in cases with increased transmission distances. Unlike the existing methods, the proposed method can improve dissemination rates without increasing power consumption in sensor networks, making it highly innovative and effective.