On The Research Of MIMO Transmission Optimization With Channel State Information

It has been proved theoretically that wireless communication system can get additionalspatial freedom degree by set multiple antennas at the transmitter and the receiver undercertain channel fading condition. Then, the MIMO system can get additional spatial free-dom degree gain. By using the additional freedom degree, the system can realize multiplexaccessing, and then achieve capacity improvement. Therefore, MIMO has become a verypopular research topic recently. This dissertation focus on the research of transmit strategyoptimization by exploiting different degree of channel state information in wireless commu-nication system. The optimal techniques will decrease BER, and improve system capacity.The main work of this dissertation is as follows.Firstly, we study the transmission strategy optimization for single user MIMO systemwith limited feedback. The key work of limited feedback based transmit strategy optimiza-tion is codebook construction. Therefore, we propose three criteria for codebook designing,which are MMSE based criterion, BER based criterion and capacity based criterion. Incodebook construction, we propose novel optimal transmitter structure, and design new dis-tortion function to measure codebook performance, and we also modify the general Lloydalgorithm. Based on the modified Lloyd algorithm, we get new style codebook. Comparedwith the existing codebook, the proposed codebook construction algorithm can get betterperformance. At the same time, we extend the research to MIMO-OFDM system. Based onthree criteria, we propose new power allocation strategy with limited feedback of channelstate information for MIMO-OFDM system.Then, we study the transmit strategy for multi-user MIMO system. By analyzing thepower optimization in MIMO-OFDM system, we propose a new power optimization algo-rithm, which can minimize the transmit power while satisfy each user's rate requirement.Thus, the system is not only with good fairness, but also achieve wireless resource optimalallocation. We also deduce the power duality relationship between multiple access channel (MAC) and broadcast channel (BC). Based on the power duality relationship, the power op-timization scheme of uplink and downlink can transform from on to the other. In multi-userMIMO system scheduling, we propose a codebook based low complexity user schedulingalgorithm. Compared with greedy user selection algorithm, the proposed scheduling algo-rithm is with very low complexity. The codebook in user scheduling is designed by usingFFT matrix. We employ the orthogonal relationship among codewords in codebook to groupusers, and then we select optimal user set from the user groups. In practical system, thetransmitter can not get perfect channel state information. Motivated by this, we study thetransmit strategy for multi-user MIMO system with partial channel state information. In thisdissertation, we propose a novel low complexity algorithm with limited feedback. In the al-gorithm, we feedback channel direction information (CDI) and channel quality information(CQI) for transmitter scheduling, which only has very low complexity in user scheduling.As the last part of our research, we study the transmit strategy optimization by crosslayer design, and propose transmission strategy optimization algorithm by using channelstate information and queue state information. Here, we propose the conception of stabilityand queue control function, and analysis the characteristic of queue control function, that is"to get system stability, the second order differential coefficient should be zero". Based onthe conception of queue control function, we derived the upper bound of the weighted sumrate by queue control, and we also propose a cross-layer design based multiuser schedulingalgorithm.