Practical implementation and real-world validation of reconfigurable intelligent surfaces

Reconfigurable intelligent surface (RIS)-empowered communications represent exciting prospects as one of the promising technologies capable of meeting the requirements of the sixth generation networks, such as low latency, reliability, and dense connectivity. The literature still lacks real-world experimental applications implementing scenarios that suggest applicability. However, validation of test cases and real-world experiments of RISs are imperative to their practical viability. To this end, this thesis presents the implementation and experimentation of the RIS in order to reveal the RIS's potential for future-generation networks in the possible use cases of RIS-assisted wireless communication networks, such as coverage enhancement, physical layer security, spectrum sensing, and wireless data link management. In Chapter 3, a physical demonstration of an RIS-assisted communication system is provided in an indoor environment in order to enhance coverage by increasing the received signal power. The performance of the RIS-assisted system is initially analyzed for a set of different locations of the receiver and observed around 10 dB improvement in the received signal power by careful RIS phase adjustments. Then, an efficient codebook design for RIS configurations is employed to adjust the RIS states on the move without feedback channels. The impact of an efficient grouping of RIS elements is investigated with the objective of reducing the training time needed to find the optimal RIS configuration. In the extensive experimental measurements, training time is reduced with the proposed grouping scheme from one-half to one-eighth by sacrificing only a few dBs in received signal power. In Chapter 4, the performance of the RIS is characterized according to the practical measurements to implement physical layer security (PLS), on which the transmitter (Alice), the intended user (Bob), and the eavesdropper (Eve) are located in an indoor environment. The measurement-based validation of the RIS deployment where the RIS is situated to maximize the secrecy capacity of the wireless communication system is still an insufficient stage although the theoretical framework of the PLS boosted by the RIS has been revealed in the literature. As of this point, it remains a challenge and an open question how the RIS can provide secure wireless communication in a practical measurement scenario by adjusting the RIS's phase shifts in such a way that the difference between the received signal power of Bob and Eve is increased to maximize the level of secrecy. The measurement campaign of the proposed system model of the RIS-aided PLS scheme demonstrates that the secrecy capacity can be significantly improved by introducing the RIS into the system with the assumption of the availability of Eve's power measurements. In Chapter 5, an experimental demonstration of an RIS-empowered spectrum sensing system is presented, in which the object detection approach by deep-learning (DL) methods is utilized to detect the signal's type and location in the frequency and time domain. The state-of-art DL-based detectors of Detectron2 and YOLOv7 are trained with the spectrograms of the synthesized signal dataset of 4G LTE and 5G NR, in which the signals are treated as objects on an image. Introducing the RIS into the proposed measurement setup significantly improves the performance of the detectors due to the direct correlation between the received signal power and the performance of the spectrum sensing process. In Chapter 6, the performance of a millimeter wave (mmWave) RIS prototype is validated with a vector network analyzer in an anechoic chamber as well as the evaluation of the considered RIS is performed with the fifth-generation (5G) waveform operating at the mmWave frequency band in a data center. In fact, link management in a data center is one of the possible playgrounds of RIS employment since rack cabinets cause severe blockage for the wireless communication system in the room. The measurements in the data center demonstrate that deploying the mmWave RIS into the data center provides improvements in received signal power.