Research On Transmission Technologies For Massive MIMO Using Finite-resolution Converters

In order to meet the ever increasing demand of multimedia broadband services in the future mobile communication network,wireless communication technologies along with its network architecture require new breakthroughs urgently.As a candidate technique for the next generation wireless system,massive multipleinput and multiple-output(MIMO)can realize high-speed wireless communication by exploiting large-scale antenna arrays and spatial division multiplexing.However,as the number of antennas increases,the system hardware cost and power consumption increase dramatically.To address this issue,a potentially effective solution is to utilize finite-resolution digital-to-analog converters(DACs)and analog-to-digital converters(ADCs).This dissertation investigates the massive MIMO technology exploiting finite-resolution DACs/ADCs from the aspects of performance analysis and transceiver design.Firstly,we study the achievable rate of a massive MIMO system equipped with both finite-resolution DACs at BS and finite-resolution ADCs at user side.Since the quantization of both finite-resolution DACs and ADCs are nonlinear,it is in general difficult to conduct exact characterizations on the quantization operations.Fortunately,a linear representation has been widely adopted by using the Bussgang theorem,which decomposes the quantized signal into two uncorrelated parts.By applying this quantization model and exploiting a zero-forcing(ZF)precoder,we derive an asymptotic expression for the downlink achievable rate.The rate loss due to transceiver quantization is accordingly characterized.To maintain the desired achievable rate at high signal-to-noise ratio(SNR),it is discovered that the number of BS antennas should increase by four times when the resolution of DACs decreases by one bit.While at low SNR,the quantization distortion is ignorable because thermal noise dominates in this condition.Numerical simulations verify the proposed linear quantization model and the derived asymptotic achievable rate.Secondly,we study the user loading scheme of a downlink multiuser massive MIMO network with finite-resolution DACs equipped at BS and finite-resolution ADCs equipped at user side.RZF precoding is exploited and transmit-side spatial correlation is characterized by Kronecker Model.By applying the asymptotic random matrix theory,we optimize the regularization parameter to maximize the asymptotic signal-to-interference-quantization-and-noise ratio(SIQNR).It is revealed that the obtained optimal regularization parameter increases linearly with respect to the user loading ratio.On the other hand,it is independent of the ADC resolution and the channel correlation.Numerical simulations verify that our proposed optimal RZF precoding outperforms conventional RZF,ZF,and MRC precodings in the aspects of achievable rate.Moreover,by maximizing the sum rate per antenna,the user loading ratio is optimized and a closed-form solution is obtained for uncorrelated channels at low SNR.Although the obtained closed-form expression is derived for a special case without correlation,it is verified also valid for slightly correlated channels by simulation results.Thirdly,we investigate a multi-cell millimeter wave communication system with a large-scale antenna array,where low-resolution ADCs are used at the BS.Massive MIMO matches well with millimeter wave communication since that the decrease in wavelength enables to pack a large number of antenna elements into small form factors.We assume that each cell serves multiple users and each user is equipped with multiple antennas but driven by a single RF chain.Analog beamforming is therefore conducted at user side.We introduce a two-step channel estimation method for the multi-cell hybrid system,which first estimates the angle of incidence at each user and then estimates the equivalent MIMO channel.Different from conventional wireless channels in cellular networks,spatial sparsity emerges as a dominant nature in millimeter wave propagations.Hence,the popular tools,such as the law of large numbers and the central limit theorem,do not apply here and the achievable rate does not converge with a large number of antennas.Fortunately,we innovatively derive a tight lower bound for the user ergodic rate with the help of stochastic calculations and Jensen's inequality,considering both channel estimate error and ADC quantization error.Based on the derived lower bound,the impacts of various system parameters,including the ADC precision,signal and pilot SNRs,and the numbers of users and antennas,on the system performance are characterized.Specifically for a typical scenario of a single-cell network,we find that the received SIQNR can be expressed as a scaling value of the original low SNR and the scaling factor relies on the abovementioned parameters.Simulation results verify the tightness of the derived lower rate bound and implies that the desired user rate can be obtained by adjusting these system parameters.Finally,we investigate secure transmission in a multiuser massive MIMO downlink network equipped with low-resolution DACs at the BS.We assume that there exists a multi-antenna eavesdropper that intends to eavesdrop the information transmitted from the BS to multiple legitimate users.The eavesdropper is passive in order to conceal its presence.We consider two popular artificial noise methods for injecting artificial noise at the BS in order to prevent the unintended receiver from eavesdropping.One method is based on artificial noise which lies in the null-space spanned by the channels of all the desired users,while the other assumes random artificial noise.It is noteworthy that the DAC quantization noise can be regarded,in some sense,as a special type of artificial noise since both are transmitted along with the information-carrying signals and produce interference at the eavesdropper.Using the DAC quantization model,a tight lower bound for the achievable secrecy rate of each user is derived.Interestingly,for a fixed power allocation factor,low-resolution DACs typically result in a secrecy rate loss,but in certain cases they provide superior performance,e.g.,at low SNR.Specifically,we derive a closed-form SNR threshold which determines whether low-resolution or high-resolution DACs are preferable for improving the secrecy rate.Furthermore,power allocation is optimized and a closed-form expression for an approximately optimal factor is derived.Then,we verify the tightness of the derived bound and the obtained insights via numerical simulation.