This dissertation summarizes achievements and ongoing designs of Field-Programmable Gate Array (FPGA) accelerators for Distributed Coherent Mesh Beamforming (DCMB). The goal of the distributed coherent network beamforming program is to create a network of distributed beams. The radios that make up this network must be small in size, weight, power, and cost while being able to overcome long transmission distances and interference. Due to the limitations, a solid communication link can be developed, using high speed to significantly increase signal strength and reduce interference. Two slots were developed to calculate the beamformer for the target platforms. One route is purely FPGA-based. Another option is a hybrid approach that uses the FPGA to do some of the initial calculations and the rest on the Central Processing Unit (CPU). Overall latency was significantly reduced when performing FPGA calculations. DCMB has become a technology for improving wireless communication systems, providing adaptability and efficiency in dynamic environments. This dissertation presents an in-depth study of DCMB with specific innovations in accelerator design and overall controller architecture. I investigate the design and implementation of dedicated accelerators adapted for DCMB tasks, including Finite Impulse Response (FIR) filtering, matrix multiplication, QR decomposition, and compensation on FPGA platforms. These accelerators are specially optimized for real-time processing and better performance on DCMB systems. Compared to soft-core processors, my research shows that hardware accelerators provide significantly faster processing speeds, enabling fast execution and reduced latency in communication systems. In addition, I discuss the design and integration of a general controller that optimizes the operation of accelerators and coordinates the beamforming process between distributed nodes. Through experiments with analytical and simulation tools, my study highlights the superiority of hardware accelerators over soft-core processors for high-speed calculation tasks in DCMB systems.

Traditional imaging systems such as the human eye and optical cameras capture the scene ahead of them called the line of sight (LoS) objects. These imaging systems are limited by their lack of field of view (FoV). Information about the non-line of sight (NLoS) objects is lost due to the objects in the LoS. They are either opaque or absorb all the incident energy, allowing for no information about the NLoS scene to be transmitted back to the detector. Amongst the popular methods used for NLoS imaging, acoustic imaging [1] offers low resolutions and suffers from interference from environmental factors. Optical methods like time-of-flight (ToF) imaging perform poorly due to shorter wavelengths leading to more scattering and absorption by occluding objects in the scene. NLoS imaging with electromagnetic (EM) rays is preferred over traditional methods because of its allowance for higher spatial resolution. It is subject to lesser interference by atmospheric factors (wind, temperature gradients.)Most everyday surfaces offer diffuse and specular reflection due to their material properties. They behave as lossy mirrors enabling propagation paths between a Terahertz (THz) Imaging System and the NLoS objects. THz waves (300 GHz – 10 THz) are the least explored if not exploited band of frequencies in the EM spectrum. A THz NLoS Imaging system is a Radar (Radio Detection and Ranging) that works by recording the backscatter information received from sending out EM signals into free space where the EM signals undergo multiple bounces off different objects in the scene. Due to the inherent nature of the radars, the return information is perceived in a way that the NLoS objects are improperly depicted when reconstructed. A correction algorithm to account for this misplacement in the reconstruction of NLoS images is proposed and its implementation is discussed in detail as a part of this work. The reconstruction algorithm processes the obtained raw THz image and performs multiple stages of classification between LoS and NLoS objects using ray casting [2]. Then the information about line-of-sight objects is fed to a line detection mechanism to detect and model the detected surfaces as mirrors. Mirror folding [3] is performed starting from the farthest generations for the objects in non-line of sight. This algorithm has been evaluated with simulated images of objects behind a single wall and two walls. With the help of a scanning THz imaging system, measurements were collected in a controlled environment, and this data was fed into the implemented algorithm for testing.

Humanpresence detection is essential for a various number of applications including defense and healthcare. Accurate measurements of distances, relative velocities of humans, and other objects can be made with radars. They are largely impervious to external factors like the impact of smoke, dust, or rain. They are also capable of working in varied intensity of light in indoor environments. This report explores the analyzing of real data captured and the application of different detection algorithms. Adaptive thresholding suppresses stationary backgrounds while maintaining detection thresholds to keep false alarm rates low. Using different approaches of Constant False Alarm Rate (CFAR) namely Cell averaging, Smallest of Cell averaging,Greatest of Cell Averaging and Order Statistic, this report aims to show its performance in detecting humans in an indoor environment using real time data collected. The objective of this project is to explain the signal processing chain of presence detection using a small scale RADAR

The fifth generation (5G) of cellular communication is migrating towards higher frequenciesto cater to the demand for higher data rate applications. However, in higher frequency ranges, like mmWave and terahertz, physical blockage poses a significant challenge to the large-scale deployment of this new technology. Reconfigurable Intelligent Surfaces (RISs) have shown promising potential in extending the signal coverage and overcoming signal blockages in wireless communications. However, RIS integration in networks requires high coordination between network notes, resulting in barriers to the wide adoption of RISs and similar IoT devices. To this end, this work introduces a practical study of integrating a remotely controlled RIS in an Open RAN (ORAN) compliant 5G private network with minimal software stack modifications. This thesis proposes using cloud technologies and ORAN features, such as the Radio Intelligent Controller (RIC) and eXternal Applications (xApps), to coordinate the RIS transparently with a 5G base station operation. The proposed framework has been integrated into a proof-of-concept hardware prototype with a 5.8 GHz RIS. Experimental results demonstrate that the framework can control the beam steering in the RIS accurately within the network. The proposed framework shows promising potential for near real-time RIS beamforming control with minimal power consumption overhead.

With the significant advancements of wireless communication systems that aim to meet exponentially increasing data rate demands, two promising concepts have appeared: (i) Cell-free massive MIMO, which entails the joint transmission and processing of the signals allowing the removal of classical cell boundaries, and (ii) integrated sensing and communication (ISAC), unifying communication and sensing in a single framework. This dissertation aims to take steps toward overcoming the key challenges in each concept and eventually merge them for efficient future communication and sensing networks.Cell-free massive MIMO is a distributed MIMO concept that eliminates classical cell boundaries and provides a robust performance. A significant challenge in realizing the cell-free massive MIMO in practice is its deployment complexity. In particular, connecting its many distributed access points with the central processing unit through wired fronthaul is an expensive and time-consuming approach. To eliminate this problem and enhance scalability, in this dissertation, a cell-free massive MIMO architecture adopting a wireless fronthaul is proposed, and the optimization of achievable rates for the end-to-end system is carried out. The evaluation has shown the strong potential of employing wireless fronthaul in cell-free massive MIMO systems. ISAC merges radar and communication systems, allowing effective sharing of resources, including bandwidth and hardware. The ISAC framework also enables sensing to aid communications, which shows a significant potential in mobile communication applications. Specifically, radar sensing data can address challenges like beamforming overhead and blockages associated with higher frequency, large antenna arrays, and narrow beams. To that end, this dissertation develops radar-aided beamforming and blockage prediction approaches using low-cost radar devices and evaluates them in real-world systems to verify their potential. At the intersection of these two paradigms, the integration of sensing into cell-free massive MIMO systems emerges as an intriguing prospect for future technologies. This integration, however, presents the challenge of considering both sensing and communication objectives within a distributed system. With the motivation of overcoming this challenge, this dissertation investigates diverse beamforming and power allocation solutions. Comprehensive evaluations have shown that the incorporation of sensing objectives into joint beamforming designs offers substantial capabilities for next-generation wireless communication and sensing systems.

Millimeter-wave (mmWave) and sub-terahertz (sub-THz) systems aim to utilize the large bandwidth available at these frequencies. This has the potential to enable several future applications that require high data rates, such as autonomous vehicles and digital twins. These systems, however, have several challenges that need to be addressed to realize their gains in practice. First, they need to deploy large antenna arrays and use narrow beams to guarantee sufficient receive power. Adjusting the narrow beams of the large antenna arrays incurs massive beam training overhead. Second, the sensitivity to blockages is a key challenge for mmWave and THz networks. Since these networks mainly rely on line-of-sight (LOS) links, sudden link blockages highly threaten the reliability of the networks. Further, when the LOS link is blocked, the network typically needs to hand off the user to another LOS basestation, which may incur critical time latency, especially if a search over a large codebook of narrow beams is needed. A promising way to tackle both these challenges lies in leveraging additional side information such as visual, LiDAR, radar, and position data. These sensors provide rich information about the wireless environment, which can be utilized for fast beam and blockage prediction. This dissertation presents a machine-learning framework for sensing-aided beam and blockage prediction. In particular, for beam prediction, this work proposes to utilize visual and positional data to predict the optimal beam indices. For the first time, this work investigates the sensing-aided beam prediction task in a real-world vehicle-to-infrastructure and drone communication scenario. Similarly, for blockage prediction, this dissertation proposes a multi-modal wireless communication solution that utilizes bimodal machine learning to perform proactive blockage prediction and user hand-off. Evaluations on both real-world and synthetic datasets illustrate the promising performance of the proposed solutions and highlight their potential for next-generation communication and sensing systems.

Reconfigurable metasurfaces (RMSs) are promising solutions for beamforming and sensing applications including 5G and beyond wireless communications, satellite and radar systems, and biomarker sensing. In this work, three distinct RMS architectures – reconfigurable intelligent surfaces (RISs), meta-transmission lines (meta-TLs), and substrate integrated waveguide leaky-wave antennas (SIW-LWAs) are developed and characterized. The ever-increasing demand for higher data rates and lower latencies has propelled the telecommunications industry to adopt higher frequencies for 5G and beyond wireless communications. However, this transition to higher frequencies introduces challenges in terms of signal coverage and path loss. Many base stations would be necessary to ensure signal fidelity in such a setting, making bulky phased array-based solutions impractical. Consequently, to meet the unique needs of 5G and beyond wireless communication networks, this work proposes the use of RISs characterized by low-profile, low-RF losses, low-power consumption, and high-gain capabilities, making them excellent candidates for future wireless communication applications. Specifically, RISs at sub-6GHz, mmWave and sub-THz frequencies are analyzed to demonstrate their ability to improve signal strength and coverage. Further, a linear meta-TL wave space is designed to achieve miniaturization of true-time delay beamforming structures such as Rotman lenses which are traditionally bulky. To address this challenge, a modified lumped element TL model is proposed. A meta-TL is created by including the mutual coupling effects and can be used to slow down the electromagnetic signal and realize miniaturized lenses. A proof-of-concept 1D meta-TL is developed to demonstrate about 90% size reduction and 40% bandwidth improvement. Furthermore, a conformable antenna design for radio frequency-based tracking of hand gestures is also detailed. SIW-LWA is employed as the radiating element to couple RF signals into the human hand. The antenna is envisaged to be integrated in a wristband topology and capture the changes in the electric field caused by various movements of the hand. The scattering parameters are used to track the changes in the wrist anatomy. Sensor characterization showed significant sensitivity suppression due to lossy multi-dielectric nature tissues in the wrist. However, the sensor demonstrates good coupling of electromagnetic energy making it suitable for on-body wireless communications and magnetic resonance imaging applications.

Predicting nonlinear dynamical systems has been a long-standing challenge in science. This field is currently witnessing a revolution with the advent of machine learning methods. Concurrently, the analysis of dynamics in various nonlinear complex systems continues to be crucial. Guided by these directions, I conduct the following studies. Predicting critical transitions and transient states in nonlinear dynamics is a complex problem. I developed a solution called parameter-aware reservoir computing, which uses machine learning to track how system dynamics change with a driving parameter. I show that the transition point can be accurately predicted while trained in a sustained functioning regime before the transition. Notably, it can also predict if the system will enter a transient state, the distribution of transient lifetimes, and their average before a final collapse, which are crucial for management. I introduce a machine-learning-based digital twin for monitoring and predicting the evolution of externally driven nonlinear dynamical systems, where reservoir computing is exploited. Extensive tests on various models, encompassing optics, ecology, and climate, verify the approach’s effectiveness. The digital twins can extrapolate unknown system dynamics, continually forecast and monitor under non-stationary external driving, infer hidden variables, adapt to different driving waveforms, and extrapolate bifurcation behaviors across varying system sizes. Integrating engineered gene circuits into host cells poses a significant challenge in synthetic biology due to circuit-host interactions, such as growth feedback. I conducted systematic studies on hundreds of circuit structures exhibiting various functionalities, and identified a comprehensive categorization of growth-induced failures. I discerned three dynamical mechanisms behind these circuit failures. Moreover, my comprehensive computations reveal a scaling law between the circuit robustness and the intensity of growth feedback. A class of circuits with optimal robustness is also identified. Chimera states, a phenomenon of symmetry-breaking in oscillator networks, traditionally have transient lifetimes that grow exponentially with system size. However, my research on high-dimensional oscillators leads to the discovery of ’short-lived’ chimera states. Their lifetime increases logarithmically with system size and decreases logarithmically with random perturbations, indicating a unique fragility. To understand these states, I use a transverse stability analysis supported by simulations.

In this dissertation, I implement and demonstrate a distributed coherent mesh beamforming system, for wireless communications, that provides increased range, data rate, and robustness to interference. By using one or multiple distributed, locally-coherent meshes as antenna arrays, I develop an approach that realizes a performance improvement, related to the number of mesh elements, in signal-to-noise ratio over a traditional single-antenna to single-antenna link without interference. I further demonstrate that in the presence of interference, the signal-to-interference-plus-noise ratio improvement is significantly greater for a wide range of environments. I also discuss key performance bounds that drive system design decisions as well as techniques for robust distributed adaptive beamformer construction. I develop and implement an over-the-air distributed time and frequency synchronization algorithm to enable distributed coherence on software-defined radios. Finally, I implement the distributed coherent mesh beamforming system over-the-air on a network of software-defined radios and demonstrate both simulated and experimental results both with and without interference that achieve performance approaching the theoretical bounds.

With the rapid development of reflect-arrays and software-defined meta-surfaces, reconfigurable intelligent surfaces (RISs) have been envisioned as promising technologies for next-generation wireless communication and sensing systems. These surfaces comprise massive numbers of nearly-passive elements that interact with the incident signals in a smart way to improve the performance of such systems. In RIS-aided communication systems, designing this smart interaction, however, requires acquiring large-dimensional channel knowledge between the RIS and the transmitter/receiver. Acquiring this knowledge is one of the most crucial challenges in RISs as it is associated with large computational and hardware complexity. For RIS-aided sensing systems, it is interesting to first investigate scene depth perception based on millimeter wave (mmWave) multiple-input multiple-output (MIMO) sensing. While mmWave MIMO sensing systems address some critical limitations suffered by optical sensors, realizing these systems possess several key challenges: communication-constrained sensing framework design, beam codebook design, and scene depth estimation challenges. Given the high spatial resolution provided by the RISs, RIS-aided mmWave sensing systems have the potential to improve the scene depth perception, while imposing some key challenges too. In this dissertation, for RIS-aided communication systems, efficient RIS interaction design solutions are proposed by leveraging tools from compressive sensing and deep learning. The achievable rates of these solutions approach the upper bound, which assumes perfect channel knowledge, with negligible training overhead. For RIS-aided sensing systems, a mmWave MIMO based sensing framework is first developed for building accurate depth maps under the constraints imposed by the communication transceivers. Then, a scene depth estimation framework based on RIS-aided sensing is developed for building high-resolution accurate depth maps. Numerical simulations illustrate the promising performance of the proposed solutions, highlighting their potential for next-generation communication and sensing systems.