Research On Algorithm And Quantification For Precoding In MmWave Massive MIMO Systems

In recent years,the rapid development of mobile communication technology has brought great convenience to people's lives.At the same time,people have put forward higher requirements for future mobile communication services.However,wireless communication systems are also facing many severe challenges.The performance of the current systems are not far from the theoretical Shannon Limit.It also faces the shortage of spectrum resources in the microwave frequency band and pilot pollution.Considering that improving the efficiency of information transmission over a limited spectrum is not the preferred solution,the spectrum-rich millimeter-wave band is a new choice.Due to the harsh environment of millimeter wave channel and large-scale antenna array providing larger antenna gain and compensating the path loss of millimeter wave propagation,the integration of millimeter wave technology and massive MIMO technology has become a hot spot in 5G communication technology research.In order to suppress the effects of pilot pollution and interference,the application of precoding technology in MIMO systems play a key role.Traditional all-digital precoding techniques are not suitable for millimeter-wave massive MIMO systems due to power consumption and complexity issues.As a compromise,mixed precoding schemes with lower complexity and higher performance have attracted wide attention.This thesis studies the design of hybrid analog and digital precoder and analog combiner for multiuser millimeter wave MIMO systems.Considering the problem of signal interference between multiple users due to diffuse scattering of signal propagation,a robust hybrid precoding algorithm based on Successive Interference Cancellation(SIC)is proposed.By deducing the orthogonal decomposition formula of the channel matrix to eliminate the interference from the known users' signals,the multi-user links optimization problem with nonconvex constraints can be decompose into multiple single-user link optimization problems.The phase extraction algorithm is then used to search each user's optimal transmission link one by one,and the multi-user hybrid precoding matrix is obtained in combination with Minimum Mean Square Error(MMSE)criterion.Simulation results show that the proposed algorithm has significant performance advantages compared with the existing hybrid precoding algorithms under severe interference conditions.Then the quantization problem of the analog precoder is studied.The analog precoder is generally composed of phase shifters in the specific implementation due to power consumption and complexity.Studies have shown that if each coefficient of the analog precoder is implemented by using two phase shifters,the hybrid precoding can achieve the full multiplexing gain.In this thesis,we propose and consider two possible combinations of phase shifters and corresponding quantization schemes,each of which use two phase shifters for each coefficient.Without quantization,these two structures have the same performance.But considering quantization,they behave differently.The simulation results show that with 3-bit quantized analog precoder,the performance is close to that of the unconstrained system.In the case of the same quantization performance,these two structures require fewer RF chains and lower quantization resolution than the structure that each coefficient is implemented with a single phase shifter.