Research On Precoding In MIMO System Under High-speed Mobility Scenarios

By the end of 2018,China's total high-speed railway mileage reached 29,000 kilometers,accounting for more than two-thirds of the world's high-speed railway operating mileage,ranking first in the world,which means that high-speed railway mobile communications face a severe test.In order to solve the problem of poor communication service quality and meet the demand of high-speed railway terminal users for mobile communication,the 3GPP organization proposed the Long Term Evolution Railway(LTE-R)technology.Multiple Input Multiple Output(MIMO)is one of the key technologies of LTE-R.This technology provides diversity and multiplexing gain in space by arranging multiple antennas at the transmitter and receiver,thereby realizing large-capacity transmission without increasing system bandwidth.MIMO uses multi-antenna technology,which causes interferences between each terminal user and each data stream.To effectively eliminate these interferences,it is required to use more complex detection methods at the receiver to recover the original data.However,in practical applications,receiver detection technology is limited by many factors,such as terminal size,high cost,and high complexity.So the 3GPP organization proposes that the base station adopts precoding technology,aiming at effectively eliminating mutual interference existing between users and reducing the complexity of processing the received signal at the receiver.This paper focuses on the MIMO system precoding technology in high-speed mobile environment.The main work is as follows:Firstly,the classical MIMO precoding algorithms and the proposed precoding algorithms in high-speed mobile environment are studied.The downlink channel model of multi-user MIMO system is introduced,and the channel capacity of multi-user MIMO system is derived.Then the characteristics of MIMO channel in high-speed mobile environment are analyzed,and based on this,the MIMO communication system model in high-speed mobile environment is established.Since the MIMO channel in high-speed mobile environment has fast time-varying characteristics,the channel state information(CSI)changes significantly during a symbol transmission period,so that the CSI obtained after the feedback delay is no longer accurately characterized.In view of this,an improved Tomlinson-Harashima precoding algorithm based on dynamic CSI model is proposed.This algorithm combines the channel statistics information with relevant characteristics to form a transmitter dynamic CSI model.The statistical characteristics of the channel are used to compensate the instantaneous CSI,which can reflect the current channel characteristics better.At the same time,the traditional Tomlinson-Harashima precoding algorithm is improved.The multi-user scheduling algorithm based on the maximizing signal-to-noise ratio criterion is combined with the Tomlinson-Harashima precoding to effectively eliminate multi-user interference,which can improve precoding precision as well as channel capacity.The simulation results show that the proposed algorithm has better error rate performance and channel capacity than traditional precoding algorithms.In actual high-speed railway mobile communication systems,due to the fast time-varying characteristics of the wireless channel,if the power distribution of the transmitter is not reasonable,the communication quality assurance cannot be obtained when the channel fades seriously,and the optimal channel capacity cannot be obtained at the same time.In order to further improve the channel capacity of the system,an improved Tomlinson-Harashima precoding algorithm based on the iterative waterfilling power allocation algorithm is proposed.When design the precoding under high speed mobile environment,the algorithm joint power allocation algorithm and user scheduling algorithm design.The iterative waterfilling power allocation algorithm and user scheduling algorithm based on greedy algorithm is combined with traditional Tomlinson-Harashima precoding algorithm,which not only reduces the bit error rate,but also further increases the channel capacity of the system.The simulation results show that the proposed algorithm has better channel capacity performance than traditional Tomlinson-Harashima precoding algorithm and linear precoding algorithm,and is more suitable for fast time-varying channel environment in high-speed mobile scenes.