Channel Estimation Algorithms For MIMO-OFDM UWB Systems

The performance of wireless communication systems is mainly impacted by wireless channel environment, and can be improved by coherent detection, equalization, space-time encoding/decoding technology from using channel information. Therefore, channel estimation is very important.High spectral efficiency, high transfer rate and high quality are the goal of wireless communication systems. As a combination of OFDM with space-time coding technique,MIMO-OFDM technology can effectively enhance the system capacity, which has attracted wide attentions. At present, the UWB technology is one of the hottest research fields due to its higher transport rate by occupying a very wide bandwidth (7.5GHz). But channel estimation in those broadband communication technologys face new challenge. This paper focuses on the channel estimation and channel order detection for MB-OFDM UWB systems and MIMO-OFDM systems respectively.For the MB-OFDM UWB system, the main work of author can be summarized as follows:(1) An improved channel estimation method with low complexity is proposed. To reduce the high computational complexity of the MMSE algorithm, we transfer the inversion problem into linear equations which can be easily solved by elementary transformation. Compared to the original MMSE method, the proposed algorithm not only has lower computational complexity, but also possess the same performance.(2) Based on the classic time-domain LS channel estimation, we propose the one-diagonal decomposition method and the three-diagonal decomposition method to reduce the computational complexity. The methods are based on the diagonal dominant property of the auto-correlation of the training m sequence. The new methods has much lower computational complexity with small performance loss,For the MIMO-OFDM system, the main work can be summarized as follows:(1) Using cyclostationarity properties and Periodic pre-coding, a novel One-Cycle blind channel estimation algorithm is proposed. Furthermore, from analyzing the detailed distribution properties of auto-correlation function, a reduced dimensional method based on the One-Cycle algorithm with reduced computational complexity is also proposed.(2) Based on cyclostationarity properties and information AIC criteria, a novel blind channel order detection method is proposed. The detection performance of the proposed method is...