Suhas Diggavi, Ph.D., December 1998
Principal advisor: Prof. Thomas M. Cover
Associate advisor: Prof. A. Paulraj
- Prof. Thomas Kailath
- Prof. D. Cox (Committee president).
Channel time-variation (or fading) is a major impairment in digital wireless communications. This occurs due to the mobility of the user or of objects in the propagation environment. The limited availability of spectral bandwidth necessitates the use of resource-sharing schemes between multiple users. As the transmission medium is shared between the users, this leads to interference between different users. In this dissertation we examine aspects of reliable communication under such impairments.
Spectral re-use introduces co-channel interference between users sharing the same frequency channels. The co-channel interference can be modeled as additive non-Gaussian noise whose covariance matrix is estimated. To study the effect of this impairment, we find the worst noise processes in the sense of mutual information, for given covariance constraints. Under some conditions on the signal and noise covariance matrices, we show the robustness of Gaussian signaling. We show that robust signal design is equivalent to finding the class of worst noise covariance matrices and designing for it. We also demonstrate the solution to the game-theoretic problem under a banded matrix constraint (specified up to a certain covariance lag) on the noise covariance matrix. In this case, we show that under certain conditions (sufficient input power) the worst channel noise has maximum entropy.
The use of multiple-antenna spatial diversity is emerging as a promising architecture for transmission over time-varying (or fading) channels. Recent results indicate significant gains in reliable data-rate by using transmitter and receiver antenna diversity. We derive the mutual information and cut-off rates for these channels. We then show that the capacity grows at least linearly with the number of antennas, not only when the number of antennas becomes large but also when the signal-to-noise ratio becomes large. In the presence of Inter-Symbol Interference (ISI) the use of multicarrier schemes has been proposed. Orthogonal Frequency Division Multiplexing (OFDM) is a popular multicarrier scheme based on the Fourier decomposition. We use OFDM as an example to study the achievable rate of multicarrier schemes on fading ISI channels. Using this we examine the trade-off between complexity and overhead.
Finally, we use the insights gained from our theoretical analysis to propose a robust receiver algorithm suitable for fast time-varying ISI channels in the presence of undesired co-channel interference. Most earlier schemes use decision-directed adaptation for suppressing the interference and these lead to severe error-propagation in time-varying channels. We propose a new scheme where we maintain estimates of the channel response and the noise covariance, conditioned on candidate data sequences. We use a colored Gaussian decoding metric, based on the estimated noise covariance matrix, to detect the signal while suppressing the interference. We maintain several candidate data sequences and their corresponding channel (and noise covariance) estimates, to develop a joint channel-data estimation (JCDE) interference suppression scheme. We also describe an estimation algorithm which incorporates knowledge of the channel structure to significantly improve performance. We study the performance of this scheme in realistic channel environments both through analysis and simulation.