Research On Channel Test And Pre-Coding For MIMO Wireless Systems

Multiple Input Multiple Output (MIMO) systems have been shown to improve data capacity through spatial multiplexing. Due to the potential improvement in system performance, many wireless systems, including the IEEE 802.11n wireless LAN, IEEE 802.16e-basedMobile WiMAX? Wave 2 and the Long-Term Evolution (LTE) mobile wireless system, have recently adopted the use of MIMO technology. Research has shown that, the performance of MIMO systems is greatly affected by the channel property, especially the spatial correlation.In order to design brand-new communication systems, it is necessary to have a deep understanding on the property of the wireless channel, both for the adopted radio frequency and the would-be candidates. Only with accurate information on spatial, temporal and frequency aspects of the wireless channel, can we design proper MIMO technologies. By testing the channel coefficients, we can analyze the spatial- correlation, condition number and channel capacity for wireless channel. Therefore the channel coefficients test is of great importance.On another side, to eliminate or reduce the co-channel interference (CCI) in MIMO system, pre-coding is utilized in base station for signal processing. Channel state information (CSI) is needed when performing pre-coding. For FDD system, the transmitter can get CSI through feedback, while for TDD systems, CSI can be obtained by channel reciprocity. When the transmitter gets the exact CSI, the pre-coding matrix should be considered to improve the system performance or approach the capacity limit. If the transmitter only knows imperfect CSI, the pre-coding design should take the error information into account to make the design a robust scheme.We propose a novel test program, which is based on the interconnection of Agilent's EDA software ADSTM and RF test instruments. Compared with other existing test program, the proposed one is more flexible and cost-saving. We can setup different frequency and signal bandwidth in this test bench. Using channel coefficient, we can calculate temporal, spatial and frequency correlation and determine the channel quality indication—condition number. Furthermore, the obtained channel matrix can be used to verify the proposed pre-coding algorithm in the following chapter.We also discuss the robust pre-coding scheme in this thesis. Geometric mean decomposition (GMD), which is dedicated to decomposing the channel into sub-channel with equal gain, combined with vector pre-coding can get better system performance. The transmitter obtains imperfect CSI due to estimation errors, feedback error etc., so we should consider the statistical information of channel error when designing pre-coding matrix and optimum perturbation vector. Simulation results show that the proposed pre-coding algorithm is more robust to imperfect CSI than conventional pre-coder.