Research On Precoding For Intelligent Reflecting Surface-aided MIMO Systems With Low-resolution DAC

During the past decades,the development of Multiple-Input Multiple-Output(MIMO)has greatly improved the spectral efficiency of wireless communication systems,but the power consumption and hardware cost of the system are challenges for practical applications.The low-resolution Digital to Analog Converter(DAC)or Analog to Digital Converter(ADC)can effectively reduce the power consumption and hardware complexity of the system.In addition,thanks to the development of electronic devices,a new passive device Intelligent Reflecting Surface(IRS)can improve the system performance by changing the wireless propagation environment.Combining the IRS with the low-resolution DAC/ADC in the MIMO communication system could significantly reduce power consumption and improve system performance.However,it is necessary to take the influence of low-resolution quantization of DAC/ADC into account when designing IRS-related parameters.Based on the above background,this thesis focuses on the downlink of IRSaided MIMO system with low-resolution DAC and designs a precoding algorithm to meet the Quality of Service(Qo S).The measurement index of Qo S is the achievable rate.On the one hand,this thesis studies the precoding when the Base Station(BS)can obtain accurate Channel State Information(CSI).Based on the additive quantization noise model,the achievable rate of the multi-user system is derived.Under the constraint of users' achievable rate,a joint optimization model with the BS precoding matrix and IRS reflection matrix as optimization variables is established to minimize the transmission power of the BS.The objective function of the model is convex while the constraints are nonconvex.Based on the Semi-Definite Relaxation method and Alternating Optimization method,an algorithm is designed to solve this nonconvex optimization function.Simulation results show that the performance of the designed joint optimization algorithm significantly outperforms the algorithm based on maximum ratio transmission.On the other hand,because it is difficult for the BS to obtain accurate CSI in practical applications,this thesis further studies the robust precoding algorithm with the bounded CSI error.An optimization function is constructed to minimize the transmission power of the BS.The difficulty of solving the function is that the designed algorithm needs to meet the Qo S of users under any possible error.The S-lemma is used to convert this Qo S constraint into a semi-definite constraint,and then the model is simplified based on Semi-Definite Relaxation.The simplified model can be divided into two subproblems by Alternating Optimization method,which are used to solve the BS precoding matrix and IRS reflection matrix respectively.Both two subproblems are convex and can be solved by the convex optimization tool CVX.The performance of the designed robust precoding algorithm is demonstrated by simulation.