Detection And Quantization In Non-Orthogonal Multiple Access System

From the basic low-rate speech service in the second generation of mobile communication system(2G)to the high-rate multimedia data services in the fourth generation of mobile communication system(4G),wireless communication techniques have a significant development in the past few decades.However,with the huge growth of the number of users,and the different types and requirements of the multimedia services,the transmission rate of the present wireless communication system is far from meeting the requirement of our highly developing society.Therefore,how to achieve a higher transmission rate in the future wireless communication systems is the main direction of the next generation of mobile communication system.Firstly,Non-Orthogonal Multiple Access(NOMA)has attracted a lot of attention and been recognized as one of the key radio access technologies for the fifth generation(5G)mobile networks.The key concept behind NOMA is that all the users are allowed to be superimposed at the receiver in the same time/code/frequency domain to significantly increase the spectral efficiency and reduce latency in the 5G communication systems.Secondly,the Multiuser Multiple-Input and Multiple-Output(MU-MIMO)is a key technology for wireless communication standards like IEEE 802.11(Wi-Fi),Wi MAX and Long Term Evolution(4G).In addition,the massive MU-MIMO is proposed to further enhance the system throughput and energy efficiency.Finally,the multiple access relay channel(MARC)is a fundamental unit of the communication network.Therefore,a high-efficiency communication strategy for the MARC is an intriguing problem.In this paper,we design,analysis and optimize the detection and quantization of the MIMO-NOMA system to meet the high quality and high rate requirements in the next 5G mobile communication system.Firstly,this paper considers a low-complexity Gaussian Message Passing Iterative Detection(GMPID)algorithm for massive MIMO-NOMA systems,in which a base station with Nrantennas serves NuGaussian sources simultaneously.Both Nuand Nrare very large numbers.Firstly,we prove that the variances of the GMPID definitely converge to the mean square error of Linear Minimum Mean Square Error(LMMSE)detection.Secondly,two sufficient conditions that the means of the GMPID converge to that of the LMMSE detection are proposed.However,the means of GMPID may still not converge when(?2-1)2?Nu/Nr?(?2-1)-2.Therefore,a new convergent GMPID with equal complexity,called SA-GMPID is proposed,which always converges to the LMMSE detection with a faster convergence speed.Finally,numerical results are provided to verify the validity of the proposed theoretical results.Secondly,we consider an iterative LMMSE detection for the uplink MIMO-NOMA systems,in which all the users interfere with each other both in the time domain and frequency domain.It is well known that the iterative LMMSE detection greatly reduces the system computational complexity by departing the overall processing into many low-complexity distributed calculations.However,it is generally considered to be suboptimal and achieves relatively poor performance.In this paper,we firstly present the matching conditions and area theorems for the iterative detection of the MIMO-NOMA systems.Based on the matching conditions and area theorems,the achievable rate region of the iterative LMMSE detection is analysed.Interestingly,we prove that by properly designing the iterative LMMSE detection,it can achieve(i)the optimal capacity of symmetric MIMO-NOMA system,(ii)the optimal sum capacity of asymmetric MIMO-NOMA system,(iii)all the maximal extreme points in the capacity region of asymmetric MIMO-NOMA system,(iv)the whole capacity region of two-user and three-user asymmetric MIMO-NOMA systems.Finally,a practical iterative LMMSE detection design is also proposed for the general asymmetric MIMO-NOMA systems.Finally,the quantize-and-forward strategy for the half-duplex Non-Orthogonal multipleaccess relay channel(NOMARC)is investigated in this paper.Different from the orthogonal MARC,the multiple users in such an NOMARC interfere with each other both in the time domain and frequency domain.For the case that the relay node cannot recover the initial bits correctly,a de-noising mutual information preserving quantizer(DMIPQ)is proposed,which can remove the noise from the received signal while preserving the useful information.A suboptimal searching algorithm is designed to reduce the complexity of DMIPQ.An approximation of DMIPQ is further discussed to get an analytical solution.It is shown that the approximate DMIPQ is equivalent to MMSE Lloyd-max quantizer.Numerical results illustrate that the proposed DMIPQ outperforms the MMSE Lloyd-max quantizer.It can also be shown that in the NOMARC,the bit error rate of DMIPQ is close to its compressand-forward bounds.