The Research And Implementation Of Channel Estimation And Efficient Precoding Technique In MIMO System

Wireless communication technology is an important driving force for the development of today's society.In particular,the introduction of MIMO technique has greatly improved the quality of wireless communication and increased the channel capacity of the communication system.On the other hand,the variability of the channel environment still brings great challenges to the quality of the received signal.In order to solve the error caused by the transmission channel and improve the signal-to-noise ratio of the communication link,a symbol-level precoding(SLP)technique is proposed.However,with the application of millimeter wave technology in MIMO systems,the number of antennas and users has surged.The use of traditional digital baseband precoding technique will continue to increase the hardware complexity and RF link power consumption of the entire communication system,and reduce energy efficiency.Therefore,how to improve the energy efficiency of the precoding system and reduce the power consumption is the key to the current research on high efficiency precoding technique.Based on the above research background,this thesis studies and implements a symbol-level digital precoder that an approximate closed form algorithm,and applies it in the precoding system,aiming to achieve high energy efficiency,low power consumption,high throughput and low complexity.This thesis mainly carried out and completed the following aspects of work.First,this thesis introduces an overview of channel estimation and symbol-level precoding techniques in precoding systems.For channel estimation algorithms,the commonly used algorithms are analyzed in combination with actual applications.In addition,three different precoding methods were introduced,and the advantages and disadvantages of their structures were analyzed separately,which laid a theoretical foundation for the subsequent design.Secondly,the algorithm principle of the SLP technique,which is an approximate closed-form algorithm,is introduced.This algorithm reduces the average power of precoded symbols while improving the energy efficiency of the system.In order to verify and compare the performance of this algorithm,it is compared with a variety of algorithms based on the Zero Force(ZF)precoding criterion.After being verified by simulation in MATLAB,the algorithm can achieve excellent performance under certain channel conditions.Thirdly,in order to build the hardware test platform of the precoder,the key technique of the SLP system was implemented,which mainly includes a channel estimation module,a precoding matrix generation module,and a symbol-level precoder module.The general framework and process of hardware implementation are introduced respectively,and the functional test and interface introduction of the designed IP are carried out.Finally,based on the optimization methodology of HLS,the performance of the designed precoder IP is optimized,which reduces the delay and improves the throughput of the design.By building a verification test platform,the optimized IP is tested for energy efficiency on the FPGA.The test results show that the energy efficiency of IP has reached the expected goal.In summary,this thesis has obtained a symbol-level digital precoder that can be used in specific projects.It has high throughput performance in symbol processing,while reducing the complexity,the average power consumption,and improving system energy efficiency.