Research On Limited Feedback Technology In MIMO Communication Systems

Accompanied by the persistent evolution of the mobile communication system, the mobile media services will gradually become the main-stream services besides the traditional voice service. In order to support the promising services, even higher data rate is urgently required, therefore multiple-input multiple-output technology, aka MIMO, which can greatly improve the spectrum efficiency of the wireless communication systems will, without a doubt ,be the killer application in future mobile communication systems. Theoretically, considerable spatial-multiplexing gain, which will highly enhance the system performance, can be extracted by taking advantage of the extra spatial freedom which is offered by the application of MIMO technology, when perfect channel state information is available at the transmitter in a point-to-point wireless MIMO transmission model. In multiuser systems, because of the multiuser diversity, even higher data rate, compared to the point-to-point system, can be achieved by proper multiuser scheduling that extracts the multiuser diversity gain. However, all of the high rate transmissions mentioned above are totally based on perfect channel state information acquisition at the transmitter. For practical systems, it is unrealistic that the receiver feedbacks the full channel information, which will be regarded as a heavy burden to the system, through the feedback channel with finite data rate. That is why limited feedback based MIMO systems as well as the feedback scheme has become the key point of the research related.In this paper limited feedback scheme in MIMO systems is discussed. Firstly, we introduce the principle, trend, as well as the main topic of limited feedback technology. In terms of different limited feedback schemes used in different systems, an comparison and evaluation is also included in the introduction. Secondly, imperfect channel state information, including channel estimation error and channel delay, is considered. In terms of channel estimation error, in order to guarantee the multiplexing gain of the codebook based multiuser limited feedback systems, the basic requirement of channel estimation as well as the law of feedback bits is analyzed and derived. Taking single-user MIMO system as an example, we analyze the throughput loss caused by feedback delay. An upper bound of throughput loss is presented in high SNR scenario, so as the formulation of the throughput loss in low SNR scenario. Thirdly, we investigate the codebook design in LTE system, and propose an new method for codebook optimization, that will reduce the system performance penalty due to the feedback error. Finally, based on the result we have got, an perspective of limited feedback in future mobile communication is given.