Design And Research Of Low-complexity Transceivers And High-Dimensional Precoding Under MIMO System

In the Multi-input and multi-output(MIMO)system,a large number of antennas are installed at the base station,which greatly improves the spectrum efficiency and energy efficiency of the system.Moreover,since the channels between different users tend to be orthogonal,decoding by simple linear decoding methods,such as zero-forcing(ZF)decoding or minimum mean square error(MMSE)decoding,can achieve excellent performance.However,due to the installation of large-scale antennas,hardware implementation and software technical difficulties are brought about.First,a large number of antennas are deployed at the base station side of a massive MIMO system,and a large number of radio frequency(RF)chains are required to process signals.In the traditional full-digital transceiver architecture,each antenna has a RF chain connected to it to process the signal,which has superior performance.However,the RF chain will include devices with very high power consumption such as analog-to-digital converters /digital-to-analog converters(ADCs/DACs).Recently,some hybrid architectures have been proposed in literatures,which keep the number of RF chains acceptable,but they are all based on the assumption that the base station side or the user side has complete channel state information.To obtain perfect CSI is very difficult under the hardware architectures of the hybrid systems,so extra hardware systems or complex signal processing might be needed,which also increase the cost.Second,in the Frequency Division Duplex(FDD),the uplink and downlink signals are transmitted in different frequency bands,the downlink CSI cannot be obtained according to the channel reciprocity like time division duplex(TDD).FDD systems need to use limited feedback technology to obtain downlink CSI.But the number of antennas at the base station side is large,which will also lead to a sharply increase in feedback overhead.In response to the above problems,this paper proposes two corresponding solutions.First,this paper proposes a scheme of a transmitter and receiver system using a small number of RF chains,and a signal processing method based on this system,including a channel estimation method and a corresponding data decoding method.And simulation results show that the proposed scheme is close to the full-digital system in performance,but the number of RF chains has dropped a lot.Second,this paper proposes a high-dimensional segment precoding method that can effectively reduce the feedback overhead,and a quick way to search for codewords.This segment precoding method can achieve a high channel utilization,and using the quick way can quickly find a codeword matched with the channel in a large size codebook,which reduces the complexity of the search.