Studies on High-speed Radio Transmission Using MIMO-OFDM (Multiple Input Multiple Output-Orthogonal Frequency Division Multiplexing) Systems

1.利用实验系统对信道估计和误码率性能进行了评估：利用数字信号处理器(DSP)板搭建了一个OFDM解调系统。通过实验验证了本课题组提出的OFDM信道估计方案。将最小平方误差方法得到的冲激响应在时间域内进行快速傅里叶变换，计算出信道。然后，我们对该系统的误码率性能进行了评估。此外，我们还研究了一种有序连续检测方法和纠错码传输。信道和频偏的联合估计：我们提出了一种仅使用4个训练符号的信道和频偏联合估计方案。仿真结果表明，该方法能够准确估计误码率，并能获得较好的误码率性能。减少信道状态信息(CSI)反馈的数量：在频分双工系统中，本征波束-空分复用(E-SDM)需要CSI的反馈。为了减少反馈的数量，我们提出了一种新的方案。该方案利用最小二乘法在时间域估计脉冲响应，并对脉冲响应进行反馈。与所有子载波信道的反馈相比，这减少了CSI反馈的量。MIMO传播评估：MIMO系统的性能取决于电磁波传播环境。我们在视线(LOS)和非视线(NLOS)环境中进行了传播测量。结果表明，LOS环境下的误码率性能优于NLOS环境下的误码率性能。然而，也有研究表明，在LOS环境中，性能高度依赖于MIMO配置。

1. Evaluation of the channel estimation and BER performance using an experimental system :An OFDM demodulation system was built up using DSP boards. Experimentally, we validated the OFDM channel estimation scheme proposed by our group. The channel was calculated by FFT applying to the impulse response obtained by the minimum square error method in the time domain. Then, we evaluated the bit error rate (BER) performance of the proposed OFDM system. Also, we examined an ordered successive detection method and transmission with error correcting codes.2. Joint estimation of channel and frequency offset :We proposed a joint estimation scheme of the channel and frequency offset using only 4 OFDM training symbols. It was shown that we can estimate them accurately and can obtain good BER performance.3. Reduction of the amount of channel state information (CSI) feedback :Eigenbeam-space division multiplexing (E-SDM) requires the feedback of CSI in frequency division duplex systems. In order to reduce the amount of the feedback, we proposed a new scheme. The scheme estimates the impulse response using the minimum square error method in the time domain, and feedback the impulse response. This reduces the amount of CSI feedback comparing with the feedback of all the subcarrier channel.4. Evaluation of MIMO propagation :The performance of MIMO systems depends on electromagnetic wave propagation environments. We conducted propagation measurement in line-of-sight (LOS) and nonline-of-sight (NLOS) environments. It was shown that the BER performance is better in LOS environments than that in NLOS ones. However, it was also shown that the performance depends highly on the MIMO configurations in LOS environments.