Research On Spatially Coupled And Power Domain Superpostion Multiple Access Key Technologies

As the mobile internet and internet of things development, and new services and applications continue to arise, this proposes the requirements of much higher throughput, lower latency and higher energy efficiency for future wireless communication network. To satisfy these requirements,the scholars at home and abroad propose novel technologies, such as massive multiple-input multiple-output, millimeter wave communication,ultra density network, non-orthogonal multiple access, and so on.According to code domain and power domain multiplexing, it can be classified into code domain and power domain non-orthogonal multiple access. Power domain non-orthogonal multiple access has been proposed to the Third Generation Partnership Project (3GPP) network standard, and it is named multiuser superposition transmission according to its technology characteristic. To make the name consist with 3GPP, this thesis names the code domain and power domain non-orthogonal multiple access as code domain and power domain superposition multiple access.Spatially coupled superposition multiple access (SCSMA) is one of the code domain superposition multiple access technology. So far, since the scholars at home and abroad use "Non-orthogonal multiple access(NOMA)" to denote power domain superposition multiple access.Without the special declaration, this thesis also adopts NOMA to represent power domain superposition multiple access.Superposition multiple access technology has many advantages,such as higher spectral efficiency, allowing lager scale users to access,better users' fairness, and so on. It can be applied into many scenarios.From the perspective of reducing detection complexity and energy efficiency improvement, this thesis investigates transceiver design of SCSMA, and proposes effective algorithm of resource allocation, such as power allocation, to reduce detection complexity and improve energy efficiency. The studies of the thesis can provide theory and algorithm support for applications of SCSMA and NOMA in the future wireless communication network. In all, the main works of the thesis can be summarized as follows:1. Different from the previous SCSMA systems transmitter design,this thesis introduces scramble to reduce the correlations of among data stream at transmitter. At receiver, joint demodulation and decoding are implemented by joint sparse graph to improve system performance and reduce detection complexity. The density evolution (DE) expression of the system including channel code is obtained by using Gaussian approximation belief propagation (BP) algorithm. The expression not only is used to calculate the decoding threshold and reflect the behavior of iteration decoding, but also guides the system design. In order to accelerate the iteration convergence rate, this thesis also proposes a serial schedule BP algorithm. If the system performance is same, extrinsic information transfer (EXIT) charts and simulation results demonstrate the proposed algorithm is able to reduce a half iterations.2. This thesis obtains the DE expression of SCSMA system which does not include channel coding by using GABP algorithm. Then, we utilize the DE expression and potential function to analyse, and prove that system performance of equal power transmission is optimal when system load is constant and the interference of among users is canceled completely. However, equal power allocation results that the convergence rate of iteration detection is slow. In order to accelerate the rate, this thesis proposes a power allocation algorithm so that unequal power transmission is introduced into the system. EXIT chart and simulation results verify the correctness of theory analyses and effectiveness of proposed algorithm.3. Energy efficiency has been regarded as an important performance metric in the future wireless communication network. This thesis investigates energy efficiency maximization problem through power allocation for multi-cell cooperation NOMA systems, under the constraints of all users' quality of service (QoS) requirements and total transmit power of base station. According to different multi-cell cooperation processing ways, this thesis studies on three multi-cell cooperation NOMA transmission schemes. Since the inter-user and inter-cell interference, the formulated energy efficiency problem is non-convex problem. This thesis utilizes the fractional programing and difference of convex programming to convert the original problem to a solvable optimization problem, and proposes an iteration power allocation algorithm to obtain all users' power allocation solution which approaches to global optimal one.4. This thesis investigates the total transmit power minimization problem through jointly optimizing user clustering, beamforming design and power allocation for multi-antennas NOMA systems, under the conditions of satisfying all users' QoS requirements. It is difficult to optimize the three key factors simultaneously due to the high computation complexity. According to the channel gain correlations and differences,this thesis first proposes a low complexity and suboptimum improved user clustering algorithm so that all users are divided to different clusters.Second, assume the user clustering is fixed. After transforming for the original formulated problem by some math operations, this thesis proposes a joint beamforming and power allocation algorithm. In order to reduce power consumption, this thesis also proposes an algorimic framework, in which user clustering algorithm and joint beamforming and power allocation algorithm are iteratively and jointly performed.Finally, the proposed user clustering algorithm, joint beamforming and power allocation algorithm and the algorimic framework are evaluated by numerical simulation.