Hun-Je Yeon

Dynamic media buffer control scheme for seamless streaming in wireless local area networks

Multimedia streaming between smart devices has come into general use in Wi-Fi networks, for example, streaming a video content in mobile phone to smart TV. However, drastic and frequent changes of link quality caused by wireless channel characteristics such as multipath fading and Doppler effect leads the network throughput unstable. In order to support seamless play even in unstable wireless link, the buffer control for media playback should be smarter and more robust against radical network variations. To this end, we propose a novel dynamic buffering mechanism for HTTP streaming. This is evaluated by experiments: streaming a multimedia content from a smartphone to a display device when they are connected to the same Wi-Fi Access Point. We show that the proposed mechanism outperforms an existing solution in terms of the number of buffering and the total buffering time. We also verify how fast the proposed method adapts to sudden changes of wireless link condition.

Realtime throughput optimization protocol in IEEE802.11 networks

To achieve efficient adaptive Transmission Opportunity (TXOP) and real-time scheduling for high system saturation throughput, we suggest a novel Real-time Throughput Optimization Protocol (RTOP) based on Dynamic Aggregation and Fragmentation (DAF) scheme. DAF scheme aims mainly at best TXOP usage and minimizing retransmissions by adjusting the frame and fragment size dynamically. RTOP is designed in accordance with the channel and traffic conditions, and complies with error adaptation mechanism. The run-time error adaptation algorithm has been applied regarding the various channel conditions and diverse traffic categories for mobile hand held devices. Optimum frame and fragment sizes are calculated using this model under different Bit Error Rate (BER) conditions. Analytical model for the suggested RTOP is developed to evaluate the throughput, delay performance, error adaptation and stability with respect to different traffic classes and scenarios, and proved with NS-3 simulator. The simulation result shows that the RTOP achieves improved throughput while ensuring long term fairness and performance stability to channel conditions and underlying physical rates.