Improving Delivery Rate for Real-Time Mobile Streaming Video

摘要

Due to the limited computing resources of cell phones and the relative lack of bandwidth of cell phone networks, broadcasting live streaming video from cell phone camera is challenging. Prior work has shown that data compression using wavelet and sub-band aligned integer run-length encoding can produce real-time streaming video. However, such transmissions can result in numerous dropped video frames. The work presented here, a real-time decision making mobile streaming video algorithm, builds upon this earlier work by adapting the number of video frames transmitted per second based on the current load of the cell phone network and the data size of the video frames. Using artificial intelligence, the algorithm makes real-time decisions to improve the video delivery rate by an average of 32%.