Research On Video Transmission Mechanism Based On Non-Orthogonal Multiple Access

With the rapid development of wireless communication technologies and the increasing popularity of smart mobile terminals,mobile network traffic is increasing rapidly,of which video traffic accounts for the vast majority.Especially the current network-video business has entered a period of vigorous development,video content has become the mainstream data of network services.However,the complex and changeable wireless environment and the massive demand for video services bring great challenges to the design of wireless video transmission.In order to adapt to the spatiotemporal changes of wireless channels and solve the problem of channel diversity between users caused by the wireless broadcast channels,it is necessary to study an effective video transmission mechanism that adapts to user channels to achieve robust scalable video transmission performance.At the same time,in order to achieve highefficiency video transmission,in the case of limited wireless resources,it is necessary to consider how to combine the emerging wireless access and cooperative transmission technologies to design a content-aware video transmission mechanism.In recent years,as one of the most potential key technologies for the next generation of wireless communication systems,non-orthogonal multiple access(NOMA)technology has attracted widespread attention and research.Its main advantage is to greatly improve spectral efficiency and system throughput by using superposition coding(SC)techniques in the power domain and successive interference cancellation(SIC)techniques at the receiver side.NOMA has its specific application potential and opportunities in scalable video transmission,which is mainly due to the ideal fit of the transmission gain brought by the SC of multiple video streams and the layered source video coding technology.However,the existing video transmission research based on NOMA is still in the exploratory stage.They basically follows the separation design of the source and channel,in which the characteristics of video content,the end-to-end transmission framework design and the data block scheduling and wireless resource optimization for improving the quality of video reception are not fully considered.In order to address the above challenges,this dissertation combines the different importance of video content and the different channel conditions of users to achieve highefficiency and robust scalable video transmission performance based on NOMA,and introduces the additional optimization degrees of freedom in the power domain.Specifically,this dissertation studies the video transmission mechanism based on NOMA,including multi-user unicast transmission scenarios,single cell multicast transmission scenarios and heterogeneous cellular network cooperative transmission scenarios.The work and major contributions of this dissertation are summarized below:1)For the multi-user unicast transmission scenarios,a multi-user video transmission scheme with the integration of spatial modulation and NOMA is proposed.The proposed scheme solves the problem of the joint source and channel cross-layer transmission design and optimization,including the content-aware cross-layer transmission mechanism and the power allocation algorithm driven by quality of experience(QoE).Through scalable video coding technology,the basic layer and the enhancement layers of video data are transmitted in the spatial domain and signal domain,respectively.In the spatial domain,layered spatial modulation is performed based on the base layer rate and the channel conditions for different users.In the signal domain,the enhanced layers are adaptively extracted based on the estimated users’ channel capacity to achieve dynamic scalable transmission performance.Then,different power levels are allocated to different users of the enhancement layer data,enabling the superimposed transmission in the signal domain.To optimize the users’ satisfaction when watching video,the QoE-driven power allocation algorithms are designed,including the optimal algorithm and the sub-optimal algorithm.Through the comparison with the reference schemes,the proposed multi-user video transmission scheme achieves better decoding performance and user experience.At the same time,the sub-optimal algorithm not only achieves near-optimal performance,but also greatly reduces the computational complexity.2)For the single cell multicast transmission scenarios,a NOMA-based video multicast scheme with statistical channels is proposed.To solve the problem of transformation from multi-user NOMA transmission mode to scalable multicast transmission mode of different video streams,the superimposed transmission scheme of video layers on the power domain is designed to achieve high-efficiency multicast while making the video quality decoded by users proportional to their channel conditions.In multicast scenarios,it is difficult for the base station to know the accurate channel information of all users.Therefore,in order to capture the realistic channel behaviors,a multicast model with statistical channels is designed,and the probability expression of the user successfully recovering the video layer is derived.At the same time,the power allocation optimization problem between video layers is modeled to maximize the average expected video quality of users on the basis of ensuring the basic service of users.Finally,based on the proposed transmission scheme,the power allocation algorithms oriented to the coverage of video layer services are designed to optimize the performance of the transmission system.Experimental results show that compared with the existing orthogonal and non-orthogonal multiple access schemes,the proposed power allocation algorithms have better transmission performance under different multicast scenarios.3)Considering that there are some users with enhanced video requirements in multicast scenarios,an inter-stream cooperative transmission scheme that supports enhanced services in NOMA-based video multicast is proposed.In order to solve the problems of higher service quality requirements of these enhanced users and efficient utilization of limited wireless resource,the design of inter-stream cooperative transmission is carried out between the enhanced service video streams and multicast service streams.The key lies in the selection of the cooperative transmission video layer,which can not only improve the quality of video reception,but also use the cooperative transmission signal to perform cooperative interference cancellation when decoding multicast signals.To improve the decoding performance of multicast signals,an optimal cooperative transmission strategy is derived to maximize the total decoding rate of SIC.In cooperative transmission phase,power pre-allocation is performed to improve the quality of service for enhanced users,and the closed-form solution to power pre-allocation is gave.In multicast stage,a power re-allocaiton algorithm oriented to the receiving video layers of users is designed to optimize the video reception quality of all users.Experimental results validate the superior performance of optimal cooperative transmission strategy in enhancing video quality and improving the total decoding rate.And through performance comparison,the proposed inter-stream cooperative transmission scheme shows an efficient capability to enhance quality of service.4)In order to take advantage of the cooperative transmission gain between the macro base station and the small base station,a cooperative video multicast scheme based on NOMA in heterogeneous cellular networks is proposed.Meanwhile,to solve the robustness problem of large-coverage video transmission,a linear transmission system is designed to achieve robust signal reception,that is to provide continuous quality change video reception performance by maintaining a linear transmission relationship between the received signals and the original video signals.Specifically,the video chunks are grouped at the small base station,and the superposition transmission of chunks is achieved within each group.The macro base station will select one of the chunks that are superimposed to provide cooperatively enhanced transmission services.At receivers,the user decodes the signals from both the small base station and the macro base station.In order to obtain the optimal cooperative transmission performance,a joint power allocation and chunk scheduling optimization algorithm is designed to minimize the total transmission distortion.Finally,through performance evaluation,it is confirmed that the designed scheme achieves remarkable performance gain using the cooperative transmission of macro base stations.At the same time,it shows the superior robust multicast performance of the linear transmission system and has a better visual effect in reconstructing video frames.