Reserch On Limited Feedback Technique For Multicell Cooperative MIMO Systems

MIMO (Multiple-Input Multiple-Output, MIMO) technology has been a researchfocus in the field of wireless communications in recent years, since it could improvethe spectrum efficiency effectively and increase the system diversity gain and thechannel capacity.However, it is a grand challenge that how to inhibit or reduce co-channelinterference between the adjacent cells when it comes to the actual deployment ofMIMO technology for multi-cell system. Multi-cell cooperative technology collectsand shares channel state information via adjacent base stations, which results in theimplementation of collaborative scheduling or the joint design of transmission signalsso as to avoid co-channel interference.Therefore, it is an important condition for achieving collaborative MIMO systemadvantage that whether the base station is enable to effectively obtain accuratechannel state information which need to be got by the uplink feedback with limitedbandwidth. Whereas, under the MIMO architecture, feedback information wouldcause a great deal of overhead and even offset the performance improvement resultedfrom multi-cell cooperation. For this reason, the establishment of efficient andpractical limited feedback mechanisms needs to be solved urgently for the sake of theeffective realization of multi-cell MIMO system.This paper focuses on the limited feedback technique for multi-cell cooperativeMIMO system. We make a deep analysis on the design concept and constructionmethod of four existing codebook in the MIMO system with limited feedback basedon codebook and indicate the comparison on the performance of these codebooks,which provides strong theoretical support and reference value for the selection of theappropriate codebook in a a practical application.In addition, in the light of the characteristics and demand of the limited feedbackfor multi-cell cooperative MIMO system, three kinds of feedback mode and the corresponding optimal feedback bit allocation schemes are designed in this paper.Each mode chooses different quantization accuracy of channel state information, andthe optimal feedback bit allocation scheme of each mode is optimized with the aim ofmaximizing the sum-rate. Based on these proposals, multi-mode limited feedbackscheme which could switch feedback mode and select the optimal feedback bitallocation scheme is put forward. And this scheme can reach higher sum-rate than thetraditional scheme and improve the efficiency of feedback and spectrum utilization.Given the the relationship between the signal fading and distance and thesimplicity of acquiring location information, finally, we present a adaptive limitedfeedback scheme based on user’s location information in which the optimal feedbackbit allocation schemes corresponding to each mode could maximize the sum-rateunder different user’s location. This scheme realizes how to secure the optimalfeedback bit allocation scheme of each mode and switch mode by the user’s location,which improves the accuracy of adaptive switch about mode and the optimal feedbackbit allocation schemes, expands the performance advantages of multi-mode limitedfeedback scheme and reduce the system complexity and enhance the practicability ofthe feedback scheme.