Research On Key Technologies Of MIMO Systems In Cellular Networks

With the rapid development of wireless communication technologies, the number of users who enjoy various applications of wireless communication is growing exponentially, and the demand for high data rate become far more urgent than before. However, the radio frequency spectrum resources are extremely limited, how to further improve the the spectral efficiency under limited frequency spectrum resources is an urgent problem to be solved. Multiple-input multiple-output (MIMO) technology has become a hot topic in the field of wireless communications. MIMO technology can significantly improve the data rate and link reliability, which has great theoretical significance and practical value, and has become one of the most important ways for the next generation of wireless communications to increase the channel capacity. MIMO technology can be categorized into closed-loop and open-loop MIMO. In closed-loop MIMO downlink, the bandwidth of the control channel by which the feedback is conveyed is severely limited, which restricts the MIMO precoding to obtain the full precoding gain with full channel state information (CSI). In the case of limited feedback, how to achieve MIMO precoding gain is an important research problem. In open-loop MIMO downlink, the transmitter can not obtain the CSI. However, the transmitter can also enhance the signal strength or obtain the spatial multiplexing gain by using open-loop MIMO technology which has also become a hot research topic. In this paper, we focus on the closed-loop and open-loop MIMO in cellular networks to expand the research. The main research results and contributions are as follows:(1) In closed-loop single user MIMO downlink, in view of codebook-based limited feedback precoding, a precoding scheme using alternating codebook is proposed. This scheme can virtually increase the size of codebook to further reduce the quantization errors without increasing feedback overhead. The alternating codebook scheme has better BER performance than conventional precoding scheme. At1km/h, alternating codebook scheme with4feedback bits can approach the performance of conventional precoding with10feedback bits. When the moving speed is3km/h, the alternating codebook scheme improves precoding performance from conventional precoding with4feedback bits to conventional precoding with8feedback bits. When the moving speed is15km/h, the gain achieved with the alternating codebook scheme is less but still significantly better than conventional precoding with4feedback bits. The alternating codebook gain diminishes at60km/h, and falls back to the same performance as conventional precoding with4bits. (2) In closed-loop single user MIMO downlink, in view of differential codebook-based limited feedback precoding, we propose to use the codebook set which incorporates codewords optimized for different channel variation speed. Through alternating these codebooks, not only the codebook size is virtually expanded, but also the codewords optimized for different speed become available at the same feedback time, avoiding the drawback of the original differential codebook which is always optimized for a single channel variation speed. From the analysis and simulation results, the proposed scheme can not only keep the robust channel tracking ability of differential precoding at low channel variation speed, but also improve the performance of differential precoding at medium to high speed, which can make the differential precoding adapt to different channel variation speed. This approach can make the limited feedback differential precoding more practical. At1km/h, the alternating differential codebook precoding performs even better than the original differential precoding with the codebook optimized for1km/h. At15km/h, the alternating codebook differential precoding still outperforms the original differential precoding optimized for1km/h and30km/h. For30km/h, the alternating codebook performance is better than the original differential precoding with the codebook optimized for1km/h and is closed to that with the codebook optimized solely for30km/h.(3) In closed-loop multi-user MIMO downlink, we apply the alternating codebook scheme into multi-user limited feedback precoding. The SDMA with orthogonal beamforming used by the3GPP-LTE standard is used as downlink transmission scheme. We further optimize the algorithm and evaluate the performance impact on multi-user MIMO systems in the cellular environment and verify its effectiveness and performance enhancement. At lkm/h, the performance of proposed scheme with4feedback bits approaches the performance of conventional precoding with10feedback bits, gaining almost6bits. At3km/h, the performance of proposed scheme with4feedback bits is close to that of conventional precoding with8bits.(4) In open-loop MIMO downlink, in view of MIMO transmission scheme in a single frequency broadcast nework, we first analyse the factors which affect the performance of open-loop MIMO spatial multiplexing, and propose a distributed MIMO spatial multiplexing scheme. We then apply the proposed scheme into LTE MBMS. We review the standardized transmission scheme in LTE MBMS, and the co-site MIMO spatial multiplexing is applied into LTE MBMS. Facing the drawback of co-site MIMO technology, we propose two distributed multi-layer transmission schemes:one is the scheduled distributed multi-layer transmission scheme, and the other is random selection based distributed multi-layer transmission scheme. Both of them can overcome the drawbacks of co-site MIMO, such as antenna correlation, LOS transmission, etc. With5MHz bandwidth, the proposed two schemes both have40%performance gain over standardized transmission in LTE MBMS, and24%performance gain over co-site MIMO spatial multiplexing scheme.