Key Technology Research On Cross Layer Optimization For Video Transmission Over Wireless Network

In wireless networks, the wireless access rate may fluctuate seriously be-cause of the user’s mobility, the migration of wireless access point and the signal interference. All the factors will influence the end-to-end transmission performance. In the traditional layered network structure, each layer has it-s own specific goal. The system transfers the data using the corresponding interfaces between the adjacent layers to complete the end-to-end communica-tion process. The hierarchical structure is easily extensible and standardized. However, with the media traffics (including video, audio, game) rapidly devel-opment, the characteristic between the media traffics is significantly different. The traditional layered network structure limits the sharing of the information about the characteristic of different traffic. It is hard to adapt the network vari-ations and get the optimal resource utilization to meet the quality of service (QoS) requirements. The researchers start to consider the cross-layer method to break through the limits of the traditional hierarchical design, and improve the transmission performance of wireless network.At present, lots of academic research about cross layer design focuses on the multimedia transmission. The aim is to enhance the video transmis-sion performance in wireless network using cross layer strategy, by consider-ing the video encoding/deconding and the wireless transmission key mecha-nism. Some emerging technologies such as dynamic adaptive streaming over HTTP(DASH), free viewpoint video (FVV) improve the user experience and are attractive. A large number of researchers begin to study how to enhance the transmission effect for these videos.The wireless access rate varies frequently, and the data in the wireless n-ode’s buffer may increase dramatically to lead to the buffer overflow. If the key information of video is lost, the video quality will decrease seriously. If the active queue management mechanism in the buffer could consider the wire-less rate variation and the characteristics of video data, the video quality will be improved effectively, virtual view synthesis technology is a key method to enhance the 3D users’ experience. Synthesising the virtual view at the client side will increase the demand of wireless access rate, however, synthesising the virtual view at the server will consume lots of cloud resource to achieve the real-time video encoding/transcoding. Thus it need to find a optimal synthesis policy to make a tradeoff and improve the system performance. Adjusting the video’s bitrate according to network condition could improve the transmission performance. However, in the multi-user scenarios, the unreasonable band-width allocation may influence the video transmission effect and fairness. A wireless resource allocation strategy by considering the video’s characteristic is important. Based on the above mentioned problem, this paper study the cross layer design for video transmission over wireless network, and the results are shown as following:1. In this part, we consider DASH-based transmission of scalable videos in wireless broadband access networks (e.g., LTE and WiMAX), and pro-pose three methods to enhance the quality of experience (QoE) of wire-less DASH users. First, we design an improved mapping scheme from SVC layers to DASH layers that can provide the desired bitrates, enhance the video end-to-end throughput, and reduce the HTTP communication overhead. Second, we develop a DASH friendly scheduling and resource allocation (DFSRA) algorithm by integrating the DASH-based media de-livery and the radio-level adaptation via a cross-layer approach. It utilizes the characteristics of video content and scalable video coding, and greatly reduces the possibility of video playback interruption by considering the client buffer status. The optimization problem is formulated as a Mixed Binary Integer Programming (MBIP) problem, and is solved by a subgra-dient method. Finally, a DASH proxy-based bitrate stabilization (DPBS) algorithm is proposed to improve the video playback smoothness, which can achieve the desired tradeoff between playback quality and stability.2. Wireless environment has frequently varying access rate and high error rate, it is hard to ensure the video transmission performance. In order to reduce the influence of bandwidth variations, this paper proposes a new active queue management mechanism (context-aware active queue man-agement, CAAQM). It could discard the non-important video data active-1y and ensure the important video data to be transmitted successfully, if the congestion happens. The proposed method is based on the cross layer principle, moreover it considers the average queue length, wireless rate variations and videos’characteristic, the response is quite sensitive.3. Interactive multiview video streaming (IMVS) could provide the stereo experience. it allows viewers to periodically switch viewpoint and lead to a fully immersed experience. Moreover, the user experience can be further enhanced by creating virtual views from neighboring coded views using view synthesis techniques. Dynamic adaptive streaming over HTTP (DASH) is a new standard that can adjust the quality of video streaming according to the network condition. In this work, we propose an improved DASH-based IMVS scheme over wireless networks. In order to reduce the bandwidth requirement, we also propose a adaptive virtual view syn-thesis strategy, which allows to synthesis the virtual view either at the cloud-based server or the client based on the network condition and the cloud computing cost.4. In wireless environment, in order to provide high-quality servers for the multiple DASH-based IMVS users, we proposes a new cross-layer scheduling policy and wireless resource allocation algorithm, which com-bines the content of the video features, virtual view synthesis cost and wireless channel quality. It could get the optimal bitrate level selection at the application layer and the wireless resource allocation at the MAC layer. The cross layer design could improve the videos’transmission performance and reduce the server cost for virtual view synthesis. The scheduling problem is summarized as a multiple choice knapsack prob-lem. In order to reduce the computing complexity, a greedy algorithm is proposed to solve the problem.