OFDM (Orthogonal Frequency Division Multiplexing) is an enabling technology for broadband wireless communications due to its ability in mitigating the multi-path propagation effect. To realize ubiquitous broadband wireless communications, all possible means such as diversity should be adopted to meet the stringent requirements of various channel conditions. OFDM with coding and interleaving can effectively take advantage of the frequency domain diversity. In the absence of frequency domain diversity, space and/or time diversity can be employed to improve system performance. However, optimum space and time diversity schemes in OFDM systems will generally induce high system complexity....[ Read more ]

OFDM (Orthogonal Frequency Division Multiplexing) is an enabling technology for broadband wireless communications due to its ability in mitigating the multi-path propagation effect. To realize ubiquitous broadband wireless communications, all possible means such as diversity should be adopted to meet the stringent requirements of various channel conditions. OFDM with coding and interleaving can effectively take advantage of the frequency domain diversity. In the absence of frequency domain diversity, space and/or time diversity can be employed to improve system performance. However, optimum space and time diversity schemes in OFDM systems will generally induce high system complexity.

In this thesis, we investigate a general OFDM architecture, which can be used to take advantage of space and time diversity with reduced complexity. In the general architecture, space diversity is achieved using time domain processing. Specifically, with multiple receive antennas, the theory of orthogonal designs and eigen-analysis are used to trade off system complexity and performance. When multiple transmit and multiple receive antennas are employed, a symbol based space diversity technique is proposed to take advantage of both transmit and receive space diversity with low complexity. In the general architecture, time diversity is achieved using a proposed bit interleaved time-frequency coded modulation (BITFCM) scheme. It is shown that the time-varying characteristics of the channel can be beneficial to system performance.

In any OFDM system, carrier frequency offset (CFO) estimation is an important issue that needs to be considered. In this thesis, we propose to use one training OFDM symbol with null sub-carriers for CFO estimation, which is of low complexity and with a large estimation range. In the proposed scheme, null subcarriers are allocated with the aid of sequence theory. We then investigate a general CFO estimator for OFDM based on the maximum likelihood estimation criterion. It is shown that with the same estimator architecture, the CFO can be obtained using training OFDM symbols, pilot tones, null sub-carriers, or a combination of them. Particularly, by taking advantage of the channel side information, the performance of CFO estimation can be significantly improved.

Finally, we investigate the CFO compensation issue in a multi-user scenario (i.e., OFDMA). We propose a reduced complexity scheme to reduce the multi-user interference (MUI) based upon the use of interference cancellation technique. Both SIR (signal to interference power ratio) analysis and simulation results show that the proposed interference cancellation scheme can significantly improve system performance.