Power Control Algorithm Research Based On Users' Decoding Order For The Downlink MIMO-NOMA Systems

Compared to OMA scheme,NOMA scheme can multiplex multiple users on the same time-frequency resource block,and thus achieve higher spectrum efficiency,which satisfies the requirement of mass access in the fifth generation mobile communication networks.At the same time,MIMO scheme,configuring multiple antennas at the transmitters and receivers to multiplex space domain resource to improve the system capacity,was proposed to meet the traffic explosive growth in the fourth generation mobile communication.In this thesis,we mainly research the combination of MIMO technology and NOMA technology,i.e.MIMO-NOMA.Three aspects,beamforming and SC and SIC are the key to the performance of the MIMO-NOMA systems.Jointly considering the above aspects,designing the low complexity power control algorithm is significant to the MIMO-NOMA systems.Based on the users' decoding order,power minimization and SINR region bound maximization for the downlink MIMO-NOMA systems are discussed.The main work contents are as follows:1.The joint optimal problem of the power allocation and users' decoding order and received detection vector in the downlink MIMO-NOMA systems is studied,which is aimed to minimize the total power consumption.Firstly,the optimality of the above three aspects are discussed respectively,and then,we formulate the joint iterative algorithm based on the optimality.Meanwhile,the convergence and optimality of the proposed algorithm are analyzed.In the simulation,we first compared the advantages and disadvantages of the MIMO-NOMA scheme and the MIMO-OMA scheme;secondly,we focused on the influence of whether to optimize the users' decoding order on the total transmission power.Simulation results demonstrate that the MIMO-NOMA scheme is superior to the MIMO-OMA scheme and optimizing the users' decoding order can obtain significant performance gain for the downlink MIMO-NOMA systems,that is,optimizing users' decoding order in MIMO-NOMA systems can decrease the total power consumption.2.The SINR region bound maximization for the downlink MIMO-NOMA systems is studied with the SINR threshold requirements and the total transmission power limited.Actually,the problem is aimed to maximize the average user rate and system capacity,that is,we can maximize the average user rate and system capacity as long as the region bound the SINR threshold vector can reach is maximized.In the process of solving the initial problem,we first decouple the initial problem because it is non-convex.In other words,we first jointly iterate the power allocation and the SINR multiplier and the received detection vectors,where the CVX tools are used to solve power allocation and the SINR multiplier.Then,jointly optimizing the users' decoding order,the joint bound finding algorithm based on CVX is formulated.Besides,a joint bound finding algorithm whose performance is equvilent to the above algorithm but without using CVX is also formulated.In the part of simulation,the convergence of the proposed algorithm is verified and the feasibility of the SINR threshold vector is analyzed according to the convergent results.Finally,compared to the MIMO-OMA scheme and the MIMO-NOMA scheme without considering the user's decoding order respectively,we verify the performance gain of the joint bound finding algorithm considering the users' decoding order in average user rate.