Resource adaptation in time-varying frequency-selective channels

The expansion of wireless services has necessitated support of higher mobility. This results in time-varying fading channels. In this scenario, the transmitter can obtain the channel state information (CSI) at the beginning of each transmission. Due to channel variations, the CSI becomes outdated at a later part of the transmission. This causes performance degradation to CSI based transmission schemes in general, and challenges for power allocation in cognitive radio (CR) environments. Such issues were overlooked in the literature. In this dissertation, we propose resource adaptation schemes to address these issues. Particularly, we investigate the power allocation for underlay CR and resource structure adaptation for orthogonal frequency division multiple access (OFDMA) system in time-varying frequency-selective fading channels.;In CR networks, the power allocation strategies in time-varying channels have not been developed yet. We propose several interference control and power allocation strategies with different criteria for underlay CRs in time-varying frequency-selective fading channels and extend the transmission schemes to multiple secondary systems. The transmission schemes are evaluated in terms of interference statistics at the primary receiver and the ergodic capacity of the secondary user link. Simulation results show how the proposed schemes with different criteria shape the interference at the primary receiver and the capacity of the secondary user. The results also illustrate better interference control capability of the proposed approaches over the existing scheme.;User channelization, where the frequency and time resources are structured into several user channels, can substantially help to capture the diversity, but it has not been well investigated. In this dissertation, we develop theoretical formulations of optimal resource structures under different metrics and analyze them in frequency-selective time-varying channels. We propose a simplified system model which maintains the main features, and based on which, we develop a new practical approach of user resource structuring which captures better diversity for OFDMA systems in multiuser time-varying frequency-selective channels. In contrast to the existing resource structures is fixed throughout the transmission frame, the proposed resource structures evolve within the frame to enjoy better diversity gain. Simulations validate that the proposed practical approach can obtain a near-optimal structure design, and provide significant advantage over the existing structures.