Research Of Multimedia Transmission Performance Optimization In Wireless Network

With the rapid development of the Internet and wireless communication technologies,the traffic of video services has shown the explosive growth trend.Today's Internet service is based on the best effort delivery,it can not guarantee the minimum bandwidth required for multimedia services and other services such as service quality parameters such as latency,jitter and packet loss rate.Compared with the wired channel,wireless channel transmission has more challenges,such as channel decay,noise and interference and other factors,resulting in video transmission in the wireless channel has higher latency,lower bandwidth and higher error rate and so on.Multimedia applications are delay-sensitive,and faced with packet loss occurs on the radio channel,the traditional method is retransmission.Obviously,retransmission is no longer suitable for delay-sensitive multimedia applications.Therefore,how to compromise the transmission of reliable quality of service in the limited wireless bandwidth has become an essential point as well as a hot topic.It is also a very challenging task.In addition to the challenges posed by the network itself to transmission,the legacy problems of traditional video codecs also pose significant problems for storage and transport.The current coding method follows the Nyquist sampling theorem,sampling first,and then dropping most of the samples while compressing,which results in a great waste of resources.The predictive video encoding based on the encoded data with motion and texture information,the importance of these information is not the same,if lose the important information will have a greater impact on the video.We explore the traditional encoding method why is not suitable for wireless channel transmission.As a major breakthrough in the signal field in recent years,compressive sensing theory has been successfully applied in many aspects.Compressive sensing theory itself studies how to reduce the amount of data transmitted by minimizing the number of measurements by designing the measurement matrix and sparse basis.Taking into account the lossy nature of the network,its dynamics and the latency sensitivity of multimedia content,we innovatively reverse this technical approach by compensating for lossy packet loss in the network by increasing the number of measurements and,over a certain period of time,Not arriving lead to packet useless for decoding.In order to improve the robustness of wireless multimedia transmission.The reason why we can use it is that the compressed,quantized,uniform quantized data form unstructured data,and there is no correlation between these data.The quality of the received picture is only related to the amount of data received related.In this paper,the technique of compressive sensing is mainly used to improve the quality of wireless video transmission from two perspectives.On the one hand,it starts from the perspective of video codec and ensures the reliable transmission of video from the perspective of network transmission.The innovation of this method is embodied in:1.Video coding by compressive sensing can solve the problem of waste of measurement and can solve the inefficiency of current encoding is not suit for wireless transmission.Ultimately,achieve the ideal case that packet loss in channel is in direction proportion to the received video quality.2.Using the democracy feature of compressive sampling,we reverse the use of compressive sensing to compensate for lossy packet loss in wireless networks by increasing the number of measurements,in order to improve the robustness of the wireless multimedia transmission.3.Proposed rate-distortion model,deduced the delay probability of user queue delay based on queuing theory and calculated the average packet loss rate based on Gilbert packet loss model.Taking into account the dynamic of the network,we designed a dynamic adaptive sampling algorithm based on the rate-distortion model,the network status and the video quality received by users.After simulation,the algorithm shows superior performance in wireless environment.