Design And Implementation Of Audio And Video Synchronization On Real-Time Streaming Media System

In the 21st century, with the rapid development of economy, unrest of the whole society accumulates gradually, thus people’s safety awareness is gradually strengthening. And computer technology, data compression technology with high compression ratio and the development of broadband communication technology have brought new growth point to the development of real-time streaming system, whose typical application is security. With the diversity of application, single video monitoring system can’t satisfy the demand of customers any more, and high performance monitoring system with both audio and video is widely needed by many areas, involving banks, railways, passenger stations, to provide a comprehensive security service. Due to network transmission unreliability, data, which should keep synchronous relationship, has lost their inherent time relationship arriving at sink end. So the performance of audio and video synchronization in real-time streaming media system has become a research hotspot.Therefore, this paper has done some research on the issue of audio and video synchronization in real-time streaming media system. Meanwhile, it designs and implements a simple real-time streaming media system, which is composed of source end, streaming media server and sink end. Besides, it has proposed a real-time streaming media synchronization scheme with network adaptive control, which is based on RTP/RTCP protocol. At the source end, it collects, encodes, packets and sends audio and video data synchronously. Each packet will be appended a timestamp before sending, which maps different media into the same timeline. At the sink end, the first step is to preprocess the received audio and video, which assures the packets are in orders. Then extracts the audio data and makes video packets into frame because video frame is too large and is split into several pieces at the source end. At last, synchronization control between audio and video proceeds according to the timestamp. In order to realize the adaptability to the network, the sink end counts network delay jitter and packet loss. It not only adjusts the buffer size and playback rate of audio and video, but also transmits the statistics to the source end. Thus the source end can adjust the video compression ratio on the basis of the received information. Experimental results show that this real-time streaming media system has excellent synchronization performance under normal network circumstance.