Design and implementation of adaptive OFDM in a software defined radio framework

This work aims at bringing information theoretic concepts for wireless resource management closer to the practical application by identifying and addressing the important challenges that arise from the implementation of an adaptive OFDM communication system. This comprises the presence of a real radio frequency (RF) channel, as well as the critical influence of the SNR estimation and hardware imperfections on the system performance. The contributions in this work are of theoretical and practical nature. Because the signal-to-noise ratio (SNR) provides the measure of signal quality often used as an input parameter for many resource allocation algorithms, the theoretical contribution of this work focuses largely on proposing an efficient algorithm for the SNR estimation in wireless OFDM systems. Furthermore, this work contains the study of the rate adaptive resource allocation algorithms in a synchronization mismatch scenario with the proposed SNR estimator. This discussion provides information about performance degradation of a real implementable transceiver with respect to an ideal transmission system with perfectly estimated parameters. The main practical contribution of this work is the design and implementation of an SDR-based reconfigurable framework for testing the capacity-achieving adaptive OFDM transmission in a wireless RF environment. The implementation comprises a complete reconfigurable OFDM physical layer transceiver with a large set of reconfigurable parameters, which are normally static in real systems. This framework enables adaptivity of individual subcarriers such that practical evaluation of resource allocation algorithms can be performed.