Wirelessly Powered Communications: From Signal Optimization to Antenna Design
Time: Wed 2022-06-15 13.30
Location: Kollegiesalen, Brinellvägen 8, Stockholm
Video link: ink for online defense
Subject area: Electrical Engineering
Doctoral student: Boules Atef Mouris , Teknisk informationsvetenskap
Opponent: Professor Aggelos Bletsas, Technical University of Crete
Supervisor: Associate Professor Ragnar Thobaben, Teknisk informationsvetenskap
Future internet-of-things (IoT) and beyond 5G communication systems are envisioned to offer large-scale wireless connectivity where the different components of life, society and industry are connected in a smart yet sustainable way. The need for continuous battery charging and/or replacement is a bottleneck for sustainability in these systems. As the number of battery-powered wireless devices grows, it is associated with an increase in both the maintenance costs and the impact on the environment. Wireless power transfer (WPT) is a promising solution to enable self-sustainable operation and limit battery usage in the enormous amount of devices that the future wireless systems will bring.
WPT co-exists by nature with other well-established communication systems. However, WPT signals are usually transmitted at a higher power level than information signals to overcome propagation losses and provide sufficient power to the receiver. Therefore, when designing a WPT system, it is essential to consider and minimize its impact on the co-existing and co-located communication systems. Moreover, in order to enable wirelessly powered communication (WPC) nodes, efficient WPT is not enough on its own but it is also important to minimize the power consumption of the node and optimize its energy usage. This thesis investigates the above described issues from both a theoretical and an implementation perspectives. It is divided into two parts; the first part focuses on signals and system level optimization with the goal of achieving wirelessly powered sensing nodes, the second part concerns enabling simultaneous wireless information and power transfer (SWIPT) by exploring novel designs of antennas and microwave components.
In the first part of the thesis, we first study the optimization of multi-tone signals to maximize the efficiency of WPT. We discuss and consider different practical non-linear energy harvester models in the problem formulation. Taking into account the in-band co-existing communication links, we provide the optimal weights for the multi-tone signals that maximizes the efficiency of WPT while minimizing the interference. The performance gains obtained using our optimization methods are highlighted through comparisons with other solutions existing in the literature. Furthermore, we present a low-complexity algorithm for designing the multi-tone signal in order to enable practical implementation. Secondly, we study the use of analog joint source-channel coding (AJSCC) in low-power sensing schemes. We propose a novel low-complexity dimension reduction mapping that is used to compress multiple sensor readings into one signal, and thus, limits the power consumption at the sensing node. We provide a comprehensive analysis of the distortion performance of our proposed mapping. We also show that energy scheduling can be utilized to improve the distortion performance of the compression mapping. Moreover, we discuss the practical circuit implementation of our proposed mapping and explain that it provides a very good distortion performance compared to the other AJSCC benchmarks despite having a much lower complexity circuit implementation. The findings of the first part of the thesis are valuable within the context of efficient and practical usage of WPT to energize a low-power IoT sensing node.
Motivated by the need for high isolation between co-located SWIPT antennas, the second part of the thesis first presents a SWIPT antenna design utilizing differential feeding in addition to an electromagnetic bandgap (EBG) structure to minimize mutual coupling between the antennas dedicated for power transmission and information exchange. Second, it investigates exploiting glide symmetry in designing EBG structures and microwave filters. We demonstrate that glide symmetry can increase the operational bandwidth of mushroom-type EBG structures without any additional manufacturing costs. A detailed equivalent circuit model is derived in order to explain this bandwidth increment. Full-wave simulations as well as experimental results are presented to verify the benefits of the glide-symmetric versions compared to the conventional structures without glide symmetry. As an alternative to the use of mushroom-type EBGs, a detailed study on the application of glide symmetry to defected ground structures (DGSs) is also conducted. We show that glide-symmetric DGSs can provide a higher rejection level as well as a higher rejection bandwidth compared to their conventional versions without symmetry. The improvement in the rejection level and bandwidth of both mushroom-type EBGs and DGSs is also explained to be useful in common-mode rejection filters. Finally, we show that fully planar EBG structures can utilize glide symmetry for size reduction and providing an increased level of isolation between microstrip patch antennas. The results of the second part of the thesis enable a new class of hardware designs that are useful for the practical realization of SWIPT systems.