Reduced-Complexity Wireless Transceiver Architectures and Techniques for Space-Time Communications

The dissertation sheds light on the performance gains of multi-antenna systems when the antenna aspects and the associated signal processing and coding aspects are integrated together in a multidisciplinary approach, addressing a variety of challenging tasks pertaining to the joint design of smart wireless transceivers and communication techniques. These tasks are at the intersection of different scientific disciplines including signal processing, communications, antennas and propagation. Specifically, the thesis deals with reduced-complexity space-time wireless transceiver architectures and associated communication techniques for multi-input multi-output (MIMO) and cognitive radio (CR) systems as well as wireless sensor networks (WSNs). The low-complexity architectures are obtained by equipping the wireless transceiver with passive control ports which require the minimum amount of RF hardware while enhancing the communication and spectrum sensing performance under non-ideal propagation conditions. The combined signal and antenna models derived throughout the dissertation are generic with respect to their applicability to any kind of reconfigurable, adaptive, parasitic and tunable RF systems.
In the context of point-to-point MIMO communication, the passive control ports
are employed for counteracting the near-field impairments represented by the local coupling and impedance mismatch, and the far-field impairments expressed by the spatial correlation and the diversity branch power imbalance. The communication RF ports are on the other hand employed for signal multiplexing. By this way, it is possible to obtain the optimal trade-off between correlation, coupling, antenna efficiency and spatial multiplexing gain over different channel scenarios.
In a multiuser scenario, the control ports are employed for synthesizing weaklycorrelated virtual antennas out of single-radio user terminals, thus achieving the multiuser diversity gains with low user populations.
The thesis also addresses the CR design challenges in size and cost constrained devices. The concept of spatial-spectrum sensing is proposed and enabled via lowcomplexity agile and steerable single-radio reactance-assisted antennas. In this context, the control ports are employed for altering both the frequency and the spatial response of the antenna. The thesis further examines the spectrum sensing performance enhancement using such reconfigurable antenna systems, via switched diversity and analogue beamforming in clustering propagation environments. Last but not least, analogue interference suppression is enabled by emulating conventional baseband zero-forcing (ZF) transceivers using a single-feed parasitic array.
In the last part of the thesis we propose novel space-time diversity techniques, such as the modified-Alamouti code, tailored for single-radio array architectures mounted on energy-constrained wireless sensor nodes. The proposed systems demonstrate significant total energy savings within simple non-cooperative communication setups compared to traditional cooperative diversity solutions under different realistic propagation conditions.