论文部分内容阅读
针对Ad Hoc网络的视频传输,提出了一种新的网络带宽预测及控制方案.该方案基于跨层、接收端反馈以及贝叶斯网络等技术.对视频传输过程进行公式化描述,从而推导出影响视频传输质量的几个主要因素,这些因素可以通过跨层机制或接收端反馈的方式获取.根据这些影响因素,确定变量集和贝叶斯网络拓扑结构,从而构建贝叶斯网络预测模型.预测结果作为Ad Hoc网络带宽,根据该带宽值来控制视频编码器,动态调节输出的实时视频流码率.为验证该方案,对整个视频通信系统进行了仿真.结果显示,跟传统的传输方案相比较,本方案能更好地利用网络带宽,减少了数据包的丢失,提高了实时视频质量.
Aiming at the video transmission in Ad Hoc networks, a new network bandwidth prediction and control scheme is proposed, which is based on the technologies of cross-layer, receiver feedback and Bayesian network, etc. The video transmission process is formulated and thus the impact is deduced Video transmission quality, these factors can be obtained by means of cross-layer mechanism or receiver feedback.According to these factors, the set of variables and the Bayesian network topology are constructed to construct the Bayesian network prediction model. The result is used as the Ad Hoc network bandwidth, the video encoder is controlled according to the bandwidth value, and the output real-time video stream rate is dynamically adjusted.In order to verify the scheme, the whole video communication system is simulated, and the result shows that with the traditional transmission scheme phase In comparison, the solution can make better use of network bandwidth, reduce the loss of data packets and improve the real-time video quality.