Research On Low-bit Fast Precoding Method In Massive MIMO Systems

The use of massive multiple input multiple output(MIMO)communication systems at the base station(BS)has the potential to achieve high spectral efficiency and high reliability and massive connectivity,however massive MIMO systems will increase hardware cost and RF power consumption of the base station.Nowadays,the use of low-resolution analog-to-digital/digital-to-analog converter(ADC/DAC)has been considered an important way to reduce the hardware cost and power consumption of massive MIMO systems.However,low resolution DACs,especially one-bit DACs,will bring in serious quantization noise,which leads to an increase in the received symbol error probability(SEP)of the system.To alleviate the influence of this drawback,this thesis introduces a ΣΔ modulation framework for the downlink of massive MIMO systems,and designs a one-bit low-power Zero Forcing(ZF)precoding method based and a symbol-level precoding(SLP)method under the framework of ΣΔ modulation.In the research of one-bit low-power ZF precoding method under the framework of ΣΔmodulation,this thesis constructs a one-bit and SEP constrained transmission power minimization problem.To solve this problem,we first convert the one-bit constraints to box constraints under the framework of ΣΔ modulation,which is difficult to solve.We thus resort to ZF precoding to transform it into a transmission power control problem,which has a closed form solution.Finally,the simulation results are used to verify the effectiveness of the proposed methods.In order to further reduce the transmission power of the base station and improve the performance of the system,this thesis considers the use of SLP technology.Unlike ZF precoding,SLP not only uses the channel state information(CSI),but also combines the constellation structure of the original information symbols to design the transceivers.In the research of one-bit low-power SLP precoding method under the framework of ΣΔ modulation,this thesis constructs a one-bit and SEP constrained transmission power minimization problem.To solve this problem,we first convert the one-bit constraints to box constraints under the framework of ΣΔ modulation,which is difficult to solve.Next,we convert the symbol error rate constraint in the optimization problem into an executable form through the standardized symbol error rate analysis method,and then convert the non-convex problem into a convex optimization problem.After obtaining the final optimization problem,this thesis designs an algorithm based on Accelerated proximal gradient(APG)to solve the optimization problem.Finally,the simulation results are used to show that the method of using SLP technology can further reduce the transmission power consumption of the base station while ensuring the SEP performance of the system.