| In order to cope with the sustained and rapid growth of mobile communication services,5th generation mobile communication systems(5G)have put forward higher requirements on information transmission rate,delay and reliability,and the large-scale multiple-input-multiple-output(MIMO)has become the key technology.In addition,the MEMO precoding schemes with multi-antenna module at the terminal can further improve the performance of the entire large-scale MIMO system.At the same time,large-scale MIMO systems are facing new problems such as mass data generation,high hardware cost and high power consumption.Therefore,reducing the resolution of analog-to-digital converter(ADC)becomes an effective technology.In this dissertation,the key technology of 5G for multi-antenna terminal is researched,which includes the uplink rate and the user scheduling scheme considering the influence of ADC quantization and radio frequency(RF)chain,the design of downlink base station and user terminal baseband processing schemes,as well as the design and implementation of high-order MIMO prototype system.First,the ADC structure and additive quantization noise model,the RF mismatch channel model and the uplink transmission signal model are given.At the same time,based on the uplink transmission signal model,the uplink rate based on low-precision ADC and non-perfect RF is deduced under the assumption of channel state information known to the receiver.The simulation results under different power and quantization bits are given,as well as the tradeoffs with RF quality.In addition,the uplink user scheduling schemes based on low-precision ADC and non-perfect RF are studied,including the difference between single-user scheduling and multi-user scheduling.The simulation results show that the greedy user scheduling has higher uplink reachable rate,and the greedy user scheduling is more obvious than the random user scheduling as the number of service users decreases.Subsequently,the model of downlink large-scale MIMO system for terminal multi-antenna is given,including the block diagrams,the signal model and the physical resource distribution,among which the physical resource distribution refers to the long term evolution(LTE)standard.In addition,two block diagonalization precoding schemes are given,which are based on singular value decomposition(SVD)and QR decomposition,respectively.Subsequently,the synchronization scheme based on synchronization sequence,linear minimum mean square error(LMMSE)channel estimation scheme with low complexity,and the hardware-friendly scheme of LMMSE signal detection based on QR decomposition are introduced.The simulation results show that,with the increase of the complexity of the modulation and the increase of the number of user antennas,the performance of the bit error ratio(BER)curves get worse.In the time division duplex system,when the LMMSE channel estimation is used in the uplink,the BER curves perform better.Finally,a high-order MIMO platform with a specially designed 5G multi-antenna module is designed and implemented.To better explain the prototype system design,the basic parameters and hardware architecture are given.The basic indicators such as the number of orthogonal frequency division multiplexing subcarriers,as well as the basic techniques such as synchronization and demodulation refer to the LTE standard.In addition,the introduction of Lab VIEW programs is given,including the block diagram of the different modules and the corresponding description.Field trials are then conducted to test the performance of this high-order MIMO-OFDM system,including visual constellations,video transmission,BER,and signal noise ratio. |
