Detection and channel estimation for MIMO-OFDM wireless communications

Orthogonal frequency division multiplexing (OFDM) transforms a frequency selective channel into a large set of individual frequency non-selective narrowband channels, which is suited for a multiple-input multiple-output (MIMO) structure that re quires a frequency non-selective characteristic at each channel when the transmission rate is high enough to make the whole channel frequency selective. Therefore, a MIMO system employing OFDM, denoted MIMO-OFDM, is able to achieve high spectral efficiency. However, the adoption of multiple antenna elements at the transmitter for spatial transmission results in a superposition of multiple transmitted signals at the receiver weighted by their corresponding multipath channels and makes the reception more difficult. This imposes a real challenge on how to design a practical system that can offer a true spectral efficiency improvement.; In this dissertation, we couple spatial transmission technology with data detection and channel estimation algorithm design to improve spectral efficiency at the physical layer for MIMO-OFDM wireless communications. In the MIMO-OFDM structure, either a space-time coded OFDM signal is sent over two antenna elements (AEs) exploiting spatial diversity, or an independent OFDM signal is sent over each of AEs utilizing spatial multiplexing. A data detection algorithm, QRD-M algorithm, is proposed to obtain a spectrally-efficient and computationally-efficient solution. The QRD-M algorithm combines advantages of both the QR decomposition (QRD) and the M-algorithm to attain a comparable performance with the maximum likelihood (ML) algorithm but reduces the computational complexity. To further improve the computational efficiency, an adaptive complexity QRD-M (AC-QRD-M) algorithm is proposed in which a different number of searching branches are assigned to individual subcarriers. The idea is to perform adaptive tree search by distinguishing strong and weak subcarriers.; Coupled tightly with spatial transmission and data detection, accurate channel estimation is indispensable to reliable communication. With the assumption of a static channel over a packet, a space-time block coded algorithm operates on two training symbols to estimate channel coefficients, and an EM-based algorithm works on one training symbol to iteratively estimate the channels in the ML sense. In a scenario of fast-fading channels, a decision-directed Kalman filter (KF) solution is proposed for joint estimation of channel coefficients, which is robust even at large normalized Doppler spreads. Iterative channel estimation through a non-decision-directed approach using Kalman filters is studied to further improve the estimation.; Signal models and algorithms for spatial transmission, data detection, and channel estimation are derived, developed and simulated. Based on the results, a performance improvement is obtained through the cooperation of spatial transmission, data detection and channel estimation.