Research On Key Technologies Of Channel Quality Estimation In Multiple-Input Multiple-Output Systems With Distributed Antennas

The channel quality in multiple input multiple output (MIMO) with distributedantennas, such as the extent of channel fading, the received signal to interference plusnoise ratio (SINR) and the inter-carrier interference (ICI), etc., are the essentialparameters of space-time decoding, error correction code decoding and adaptive codedmodulation. The accuracy of the channel quality estimation have a direct impact on thethroughput of the system link, thus the channel quality are the key reference of thewireless resource allocation and management.There are a lot of room for improvement on channel quality estimation andoptimization in the next generation mobile communication systems such as LTE-A. Themainly aspects are summed up. Firstly, the larger the number of transmit antennas, themore the spectrum resources occupied by pilot symbols. Secondly, a high precision andlow complexity method for the SINR estimation is needed in the actual project. Thirdly,in the LTE-A scenarios with distribted antennas, the existence of heterogeneous Dopplerbecomes possible, therefore the calculation of ICI need to be reconsidered.In order to further improve the spectrum utilization, around the shortcomingsdescribed above, this thesis focuses on the study of estimation and optimizationmethods for channel quality in distributed MIMO, including: location design for pilotsymbols, antenna number selection, SINR estimation, and ICI analysis.Firstly, pilot design for achieving full frequency reuse is proposed in downlinkdistributed MIMO systems. Using the imbalance of the large-scale fading of distributedchannel, the pilot symbols of adjacent antennas are orthogonal in time and frequencydomains, and the pilot symbols of distant antennas are multiplexed. Based onminimizing the mean squared error of channel estimation, the pilot density is optimized.Under the condition of orthogonal space-time block coding with three transmit and tworeceive antennas, multipath Rayleigh fading channels, and antenna spacing of500meters, at the area averaged mean square error of102, the proposed methodoutperforms the centralized MIMO based method by1.2dB in signal to noise ratio.Secondly, a method for antenna number selection is presented in the line-shaped cell downlink distributed antenna systems. In the different position, the performance iscompared between single transmit antenna with full power and the Alamouti scheme.The antenna number is selected based on minimizing the bit error probability. Under thecondition of distance between transmit antennas500meters and standard deviation ofshadowing8dB, when the distance between mobile station and transmit antenna is lessthan80meters, at the area averaged bit error probability of102, the transmit power canbe reduced by1.5dB after optimization for antenna number selection.Thirdly, a SINR estimation method is proposed for LTE-A over frequency-selectiveRayleigh fading channels. A reduced-rank matrix is constructed utilizing the correlationbetween subcarriers, and the smallest eigenvalue of the covariance matrix is extracted toestimate SINR. The simulation results show that, when SINR is less than30dB, theproposed method requires only continuous four pilot subcarriers for realizing estimationerror is less than0.2dB, and that the proposed method is insensitive to the Doppler shift.Finally, the different locations of receiving antennas may lead to different types ofDoppler power spectrum in uplink distributed MIMO. For this scene, the efforts ofDoppler spreading on ICI are analyzed. The instantaneous SINR is presented usingzero-forcing detector at first, and then the carrier-to-interference ratio (CIR) expressionis derived. The numerical results show that, under the conditions of two transmit andtwo received antennas,256subcarriers and subcarrier spacing of7.81kHz, the CIRperformance of the flat spectrum has4.8dB gains relative two-path spectrum.This thesis studies the location design for pilot symbols, antenna number selection,SINR estimation, and ICI analysis for the channel quality estimation problem indistributed channel envirenment. The research results can be used in the broadbandaccess based on a distributed multi-antenna, the mobile communications and thewireless communication systems.