Research On Performance Analysis And Optimization Algorithms For Cooperative NOMA Systems

Cooperative communication is an effective method to extend communication coverage,make up for pass-loss,and combat multipath fading.However,these advantages are achieved at the cost of consuming additional orthogonal resources.As a promising candidate,non-orthogonal multiple access(NOMA)introduces power-domain coding into multiple access technology,and can integrate with the traditional orthogonal resources.In this way,the restriction of the orthogonality between different user-resource-blocks can be broken,which can effectively improve the spectrum efficiency(SE)of the wireless communication in comparsion with the traditional orthogonal multiple access(OMA).Therefore,how to improve the cooperative communication system's SE by use NOMA has been attracted great attention.This dissertation focuses on theoretical performance analysis and optimization algorithm design for the cooperative NOMA systems.The cooperative NOMA system models are constructed according to the user cooperative transmission model and the dedicated relay cooperative transmission model.In order to investigate the cooperative NOMA,the performance of the cooperative NOMA systems is analyzed from the perspectives of error rate,outage probability,and data rate.Then,some optimization algorithms are proposed in terms of power allocation,beamforming,and users clustering to improve the spectral efficiency of the cooperative NOMA system.The innovations of this dissertation are summarized as follows:The user-error-rate performance of the NOMA system and the cooperative NOMA system under the Nakagami-m fading channel are studied.Firstly,a two-user NOMA system model is constructed,and the users' bit-error-rate(BER)expressions are derived by use of constellation space analysis method under the additive white Gaussian noise(AWGN)channel.Secondly,the users' BER expressions are derived by use of the moment generating function(MGF)under the Nakagami-m fading channel.The users' BER performance in the OFDM-NOMA system is also analyzed under the Nakagami-m fading channel.Finally,the two-user cooperative NOMA system model is constructed,the direct transmission signal and the cooperative signal are processed by use of maximum ratio combination method(MRC).In this way,users' BER expressions of the two-user cooperative NOMA system are derived under the Nakagami-m fading channel.Simulation results demonstrate that the OFDM-NOMA system and cooperative NOMA system can significantly improve the BER performance than the traditional NOMA system.The outage probability and the ergodic data rate performance of the NOMA with full-duplex(FD)/half-duplex(HD)relaying systems in the presence of two practical undesirable defects,namely channel estimation error(CEE)and in-phase/quadrature-phase imbalance(IQI)are investigated.Firstly,cooperative NOMA with FD/HD relaying system models under the joint influence of I/Q imbalance and CEE are constructed.Secondly,the expressions of the outage probability and the ergodic data rate are derived,which are critical for analyzing the impact of hardware damage on the performance of the cooperative NOMA system.Simulation results verify the analytical solution.In addition,the simulation results also show that FD cooperative NOMA is superior to FD cooperative OMA in terms of outage probability and ergodic data rate.And the performance of FD cooperative NOMA is also better than the performance of HD cooperative NOMA in the low signal-to-noise ratio(SNR)region.The outage probability of the cell-edge user is reduced.Firstly,a dynamic simultaneous wireless information and power transfer(DSWIPT)NOMA system model is proposed by making the time and power allocation coefficient in SWIPT flexible and variable.According to the DSWIPT protocol,the cell-center user can dynamically balance the energy distribution among signal reception,energy storage,and cooperative communication.Secondly,the outage probability and diversity gain expressions for the cell-center user and the cell-edge user in the DSWIPT NOMA system are derived,which are essential for analyzing the system performance.Finally,in order to obtain the optimal time and power coefficients,the total throughput optimization problem of the DSWIPT NOMA system is designed,and an optimization algorithm for the time and power coefficients is proposed based on the interior point method.Simulation results show that the proposed DSWIPT NOMA structure and its optimization algorithm can further improve the outage probability performance of the cell-edge user and the system throughput in comparision with the existing structure.In the beamforming-based multi-user relay cooperative NOMA system,the joint power-allocation and relay beamforming optimization NOMA(JPBO NOMA)algorithm is proposed.A downlink HD NOMA relay system and the system's weighted sum-rate(WSR)optimization problem is formulated.Since the optimization problem is nonconvex,the JPBO NOMA algorithm is proposed to maximize the NOMA AF relay system's WSR.In the proposed algorithm,the power allocation and the relay beamforming matrix are alternately optimized by use of successive convex approximation(SCA)method and Lagrange-multiplier(LM)method.Simulation results demonstrate that the proposed new algorithm can significantly improve the WSR with a faster convergence than the existing method.On the other hand,multiple users need to be clustered when the number of the users is large.Thus,a user clustering algorithm is proposed.Firstly,the beamspace multiple-input multiple-out(MIMO)NOMA system is considered in the millimeter-wave communication system.Secondly,a uniform beam selection algorithm is proposed according to the user classification and beam selection.Simulation results verify that not only the spectrum efficiency but also the computation efficiency of the users can be improved quite a lot with the proposed uniform beam selection algorithm.