Video Transmission Control Method Based on Deep Q-Network

With the wider application of the Internet, video transmission has gradually become an important way to obtain information. However, traditional video transmission control methods still have problems such as large transmission delays, poor bandwidth utilization, and so on. This paper aims to optimize video transmission control using Artificial Intelligence (AI), to better improve video quality and enhance the stability of video transmission. A video transmission control model was constructed using deep Q-networks, and buffer overflow and buffer underflow were dynamically analyzed; furthermore, video transmission control was achieved through user experience QoE (Quality of Experience) and adaptive bit rate algorithms; finally, the effectiveness of AI in video transmission control was verified. Through empirical research, it was found that the average transmission delay of short video transmission under the AI method in this paper was only 14.194 milliseconds, and the average bandwidth utilization rate reached 80.37%; the average transmission delay under traditional methods was 50.999 milliseconds, and the average bandwidth utilization was 49.73%. The method proposed in this paper can significantly reduce transmission delay and improve bandwidth utilization when applied to video transmission control. This study demonstrates the effectiveness of artificial intelligence in video transmission control, providing more methods for improving video transmission quality.