The Research Of The Transmission Solution To Improve QoS In Video Conference

Compared with traditional on-site meetings,online video conferences are more convenient and can further facilitate communication.At the same time,the current network is developing in the direction of heterogeneous networks,such as wireless networks and mobile cellular networks,which has caused new problems in the use of network video conferencing in heterogeneous networks.The new problem is that the transmission system has not considered the diversity of packet loss and bandwidth fluctuations in heterogeneous networks,so it will degrade system performance.Therefore,increasing the applicability of the transmission system to the heterogeneous network will be of great significance to improve the system performance of the network video conference.The thesis studies the congestion and playback control in video transmission schemes.First of all,through researching the video transmission scheme,two key control modules in the transmission scheme,video transmission congestion control and video playback control are pointed out,and the research progresses related to these two modules are introduced in detail,and their advantages and disadvantages are analyzed analyzing and summarizd.Secondly,the thesis proposes a Kalman filter-based packet loss differentiation algorithm to solve the problem of lack of effective packet loss differentiation algorithm for congestion control in the transmission scheme,in order to improve the accuracy of packet loss differentiation of congestion control algorithm in the heterogeneous network.The algorithm performs Kalman filter to calculate the packet arrival time interval for improving the accuracy of time delay estimation.Through comparing the packet arrival interval in the packet loss period with the packet arrival interval in the stable period,the algorithm can determine the packet loss type and use the judgment value for information feedback.In this thesis,the proposed packet loss differentiation algorithm is applied to the TFRC congestion control algorithm,and a video transmission system is constructed in NS2 to verify the effectiveness of the algorithm.The simulation results show that the improved TFRC algorithm has lower misjudgment rate of packet loss and higher throughput than other congestion control algorithm.Finally,this thesis proposes a video playback algorithm based on state transition probability,aiming at solving the problem that the traditional video playback control algorithm adjusts the playback rate so fast that the degree of user experience has been reduced.The algorithm calculates the state transition probability of the buffer firstly.And then the algorithm combines the transition probability and the current buffer state to estimates the next state.Finally,the algorithm adjusts the playback rate according to the estimated state.The state transition probability is calculated through the sliding window of the remaining frames of buffer.This thesis sets up video transmission system on NS2,and carries on the simulation experiment of the video playback algorithm on it.The simulation results show that compared with the traditional AMP-Mean and the playback algorithm based on probability estimation,this algorithm has a more stable playback rate adjustment and playback delay,which can better improve the system QoS.