Abstract:

The real-time video services in wireless networks are fragile to the dynamic network environment.Therefore, proper buffering management techniques are adopted to mitigate the impact of network fluctuation on user experiences. An adaptive buffer management algorithm based on queue prediction was proposed for wireless video streaming applications. By judging the current state of the buffer, the scheme adjusted the video playout rate and buffer size according to the real-time frame arrival rate and end-to-end delay. The experiments were conducted to validate the proposed algorithm.The experimental results show that this method is able to adaptively adjust buffer size and playout rate according to network fluctuations,which reduces the interrupt frequency and the number of lost frame effectively.

Key words: adaptive buffer management, queue prediction,, buffer status, delay contro