I2S Masters/ Doctoral Theses

AMD’s latest generation of adaptive system-on-chip (SoC) devices, the Versal product family, offers enhanced processing capabilities that are attractive to researchers and system designers. However, these capabilities introduce a significant knowledge barrier, limiting the practical benefits of Versal devices compared to more mature platforms from AMD, Intel, and other industry vendors. This project addresses this challenge through two primary deliverables: a software framework and a comprehensive user manual targeting Versal development. The software framework, named RSL Versal Core, provides a framework for users unfamiliar with Versal devices by selectively abstracting away more complex design components. Using a small set of commands, users can synthesize a programmable logic (PL) design, compile a Linux operating system for the onboard Arm processor with PL communication support, and program supported development boards. Following initial setup, the framework also supports extended software and firmware development for specific project needs. The accompanying user manual documents both RSL Versal Core and broader Versal development concepts. It guides users through reproducing and customizing the framework outputs manually and introduces key architectural and design principles useful for effective Versal-based system development. Together, these deliverables enable new developers to rapidly gain proficiency with Versal platforms and enable implementation of digital signal processing (DSP) concepts.

Due to rapid advances in high-speed analog-to-digital conversion and software-defined architectures, modern radar systems increasingly shift signal generation and conditioning into the digital domain. These architectures enable high-fidelity signal capture and provide substantial flexibility in waveform synthesis and signal processing that was previously impractical in analog implementations. Despite these advances, however, achievable radar performance remains fundamentally constrained by the physical transmit hardware through which the signal is ultimately realized. Nonlinear amplification, finite bandwidth, and memory effects introduce distortion that creates a significant gap between idealized waveform design and the waveform that is physically radiated.

To address this limitation, this work proposes a system-aware radar waveform design framework that couples data-driven system identification with deterministic optimization to generate waveforms tailored to the underlying transmit hardware. A complex baseband memory polynomial model is developed to characterize nonlinear transmit-chain behavior using loopback measurements, where $\ell_1$-regularized LASSO estimation is employed to improve robustness against ill-conditioning and feature redundancy. Under this architecture, a generalized integrated sidelobe level (GISL) objective is reformulated using logarithmic scalarization to produce a numerically stable and Pareto-tunable optimization criterion capable of balancing output energy and sidelobe suppression. Additionally, efficient vectorized gradient expressions are derived using Wirtinger calculus and implemented using gradient-based descent and the limited-memory BFGS algorithm for practical high-dimensional waveform synthesis.

To validate the framework, a comprehensive hardware-in-the-loop testbench was developed supporting direct model identification and experimental evaluation of optimized waveform performance. Simulation and experimental results demonstrate that continuous-phase FM waveforms exhibit strong inherent robustness to nonlinear distortion, while phase-coded waveforms with large instantaneous phase discontinuities show significantly greater sensitivity to transmit-chain impairments. Across both waveform classes, the proposed framework achieves substantial improvements in output power efficiency and pulse compression performance relative to system-agnostic waveform design. These results demonstrate that transmitter constraints must be treated as fundamental design variables rather than secondary effects and establish system-aware optimization as a practical framework for next-generation radar waveform synthesis.

Due to the ever-growing demand for access to the finite resources of the electromagnetic spectrum, significant effort has been directed toward improving spectrum utilization. This has become a particular challenge in radar transmission design, where waveform diversity techniques have emerged as a promising solution despite the accompanying implementation complexity. Diverse signals are inherently non-repeating and pose unique challenges in comparison to traditional radar waveforms. Software defined radios (SDRs) allow for traditional RF components and signal processing to be implemented and controlled in software rather than hardware, providing a platform for testing experimental radar algorithms. This thesis presents a real-time parallel implementation of five previously developed distinct waveform-diverse radar signals for use in a coherent SDR system. The implemented waveforms include stochastic waveform generation (StoWGe), multi-user radar communication (MURC), phase-attached radar communication (PARC), pseudo-random optimized frequency modulation (PRO-FM), and waveform recycling. To enable real-time generation at maximum SDR data rates, these waveforms are implemented using digital synthesis techniques via GPU parallel processing. This approach alleviates CPU resource limitations by offloading computationally intensive waveform generation tasks to the GPU, enabling continuous high-throughput operation. A custom asynchronous transmit and receive architecture is developed to integrate these GPU-accelerated waveforms with UHD-based SDR hardware. The system leverages a multithreaded framework approach that can sustain coherent and synchronized radar operation. To validate the system, a series of loopback testing across all waveforms and a variety of parameters is completed to confirm the execution of the generate-transmit-receive chain.

The RF spectrum is a precious, finite resource with ever-increasing demand. Consequently, the mandate to be a "good spectral neighbor" is in direct conflict with the requirements for high-performance sensing where correlation error is fundamentally limited. As such, matched-filter radar performance is often sidelobe-limited with estimation error being constrained by the time-bandwidth (TB) of the collective emission. The methods developed here seek to bridge this gap between idealized radar performance and practical utility via waveform design.    

Estimation error becomes more complex when employing pulse-agility. In doing so, range-sidelobe modulation (RSM) spreads energy across Doppler, rendering traditional methods ineffective. To address this, the gradient-based complementary-FM framework was developed to produce complementary sidelobe cancellation (CSC) after coherently combining subsets within a pulse-agile emission. In contrast to the majority of complementary signals, explored via phase-coding, these Comp-FM waveform subsets achieve CSC while preserving hardware-compatibility since they are FM (though design distortion is never completely avoided). Although Comp-FM addressed practicality via hardware amenability, CSC was localized to zero-Doppler. This work expands the Comp-FM notion to a Doppler-generalized (DG) framework, extending the cancellation condition to an arbitrary span. The same framework can likewise be employed to jointly optimize an entire coherent processing interval (CPI) to minimize RSM within the radar point-spread-function (PSF), thereby generalizing the notion of complementarity and introducing the potential for cognitive operation if sufficient scattering knowledge is available a-priori.          

Sensing with a single emitter is limited by self-inflicted error alone (e.g., clutter, sidelobes), while MIMO systems must additionally contend with the cross-responses from emitters operating concurrently (e.g., simultaneously, spatially proximate, in a shared spectrum), further degrading radar sensitivity. Now, total correlation error is dictated by the overlapping TB (i.e., how coincident are the signals) and number of operating emitters, compounding difficulty to estimate if left unaddressed. As such, the determination of "orthogonal waveforms" comprises a large portion of MIMO literature, though remains a phenomenological misnomer for pulsed emissions. Here, the notion of complementary-FM is applied to a multi-emitter context in which transmitter-amenable quasi-orthogonal subsets, occupying the same spectral band, are produced via a similar gradient-based approach. To further practicalize these MIMO-Comp-FM waveform subsets, the same "DG" approach described above, addressing the otherwise-default Doppler-induced degradation of complementary signals, is applied. In doing so, Doppler-independent separability and complementarity greatly improves estimation sensitivity for multi-emitter systems. 

This MIMO-Comp-FM framework is developed for standard matched filter processing. Coupling this framework with a "DG" form of the previously explored MIMO-MiCRFt is also investigated, illustrating the added benefit of pairing optimized subsets with similarly calibrated processing. 

Each of these methods is developed to address unique and increasingly complex sources of estimation error. All approaches are initially developed and evaluated via simulated analysis where ground-truth is known. Then, despite hardware-induced distortion being unavoidable, the MIMO-Comp-FM framework is confirmed via loopback measurements to preserve the majority of CSC that was observed in simulation. Finally, open-air demonstration of each approach validates practical utility on a radar system.

Metagenomic sequencing has become a central paradigm for studying complex microbial communities and their interactions with the host, with emerging applications in clinical prediction and disease modeling. In this work, we first investigate two representative application scenarios: predicting immune checkpoint inhibitor response in non-small cell lung cancer using gut microbial signatures, and characterizing host–microbiome interactions in neonatal systems. The proposed reference-free neural network captures both compositional and functional signals without reliance on reference genomes, while the neonatal study demonstrates how environmental and genetic factors reshape microbial communities and how probiotic intervention can mitigate pathogen-induced immune activation.

These studies highlight both the promise and the inherent difficulty of metagenomic analysis: transforming raw sequencing data into clinically actionable insights remains an algorithmically fragmented and computationally intensive process. This challenge arises from two key limitations: the lack of a unified algorithmic foundation for sequence alignment and the absence of systematic approaches for selecting and organizing analytical tools. Motivated by these challenges, we present a unified computational framework for metagenomic analysis that integrates complementary algorithmic and systems-level solutions.

First, to resolve fragmentation at the alignment level, we develop the Versatile Alignment Toolkit (VAT), a unified algorithmic system for biological sequence alignment across diverse applications. VAT introduces an asymmetric multi-view k-mer indexing scheme that integrates multiple seeding strategies within a single architecture and enables dynamic seed-length adjustment via longest common prefix (LCP)–based inference without re-indexing. A flexible seed-chaining mechanism further supports diverse alignment scenarios, including collinear, rearranged, and split alignments. Combined with a hardware-efficient in-register bitonic sorting algorithm and dynamic index-loading strategy, VAT achieves high efficiency and broad applicability across read mapping, homology search, and whole-genome alignment. Second, to address the challenge of tool selection and pipeline construction, we develop SNAIL, a natural language processing system for automated recognition of bioinformatics tools from large-scale and rapidly growing scientific literature. By integrating XGBoost and Transformer-based models such as SciBERT, SNAIL enables structured extraction of analytical tools and supports automated, reproducible pipeline construction.

Together, this work establishes a unified framework that is grounded in real-world applications and addresses key bottlenecks in metagenomic analysis, enabling more efficient, scalable, and clinically actionable workflows.

Conventional computer vision relies on spatially resolved, human-interpretable images, which inherently expose sensitive information and raise privacy concerns. In this study, we explore an alternative paradigm based on lensless imaging, where scenes are captured as diffraction patterns governed by the point spread function (PSF). Although unintelligible to humans, these measurements encode structured, distributed information that remains useful for computational inference. 

We propose a unified framework for privacy-preserving vision that operates directly on lensless sensor measurements by leveraging their frequency-domain and phase-encoded properties. The framework is developed along two complementary directions. First, we enable reconstruction-free inference by exploiting the intrinsic obfuscation of lensless data. We show that semantic tasks such as classification can be performed directly on diffraction patterns using models tailored to non-local, phase-scrambled representations. We further design lensless-aware architectures and integrate them into practical pipelines, including a Swin Transformer-based steganographic framework (DiffHide) for secure and imperceptible information embedding. To assess robustness, we formalize adversarial threat models and develop defenses against learning-based reconstruction attacks, particularly GAN-driven inversion. Second, we investigate the limits of privacy by studying the reconstructability of lensless measurements without explicit knowledge of the forward model. We develop learning-based reconstruction methods that approximate the inverse mapping and analyze conditions under which sensitive information can be recovered. Our results demonstrate that lensless measurements enable effective vision tasks without reconstruction, while providing a principled framework to evaluate and mitigate privacy risks. 

Coherent wavelength-division multiplexed (WDM) optical fiber systems have become the primary transmission technology for high-capacity data networks, driven by the explosive bandwidth demand of cloud computing, streaming services, and large-scale artificial intelligence training infrastructure. This dissertation investigates two fundamental aspects of digital coherent fiber optic systems under the unifying theme of source and monitoring: the design of multi-wavelength optical sources compatible with high-order coherent detection, and the leveraging of fiber Kerr-effect nonlinearity at the coherent receiver to perform physical-layer link health monitoring and to assess inherent security vulnerabilities — both achieved through digital signal processing of the received complex optical field without dedicated hardware.

We begin by addressing the multi-wavelength transmitter challenge in WDM coherent systems. Existing quantum-dot, quantum-dash, and quantum-well based optical frequency comb (OFC) sources share a common limitation: individual comb line linewidths in the tens of MHz range caused by low output power levels of 1–20 mW, making them incompatible with high-order coherent detection. We demonstrate coherent system application of a single-section InGaAsP QW Fabry-Perot laser diode with greater than 120 mW optical power at the fiber pigtail and 36.14 GHz mode spacing. The high optical power per mode produces Lorentzian equivalent linewidths below 100 kHz — compatible with 16-QAM carrier phase recovery without optical phase locking. Experimental results obtained using a commercial Ciena WaveLogic-Ai coherent transceiver demonstrate 20-channel WDM transmission over 78.3 km of standard single-mode fiber with all channels below the HD-FEC threshold of 3.8 × 10⁻³ at 30 GBaud differential-coded 16-QAM, corresponding to an aggregate capacity of 2.15 Tb/s from a single laser device.

After investigating the QW Fabry-Perot laser as a multi-wavelength source for coherent WDM transmission, we leverage the coherent receiver DSP to exploit fiber Kerr-effect nonlinearity for longitudinal power profile estimation, enabling reconstruction of the signal power distribution P(z) along the full multi-span link without dedicated hardware or traffic interruption. We propose a modified enhanced regular perturbation (ERP) method that corrects two independent physical error sources of the standard RP1 least-squares baseline: the accumulated nonlinear phase rotation, and the dispersion-mediated phase-to-intensity conversion — a second bias source not addressed by prior methods. The RP1 method produces mean absolute error (MAE) that scales quadratically with span count, growing to 1.656 dB at 10 spans and 3 dBm. The modified ERP reduces this to 0.608 dB — an improvement that grows consistently with link length, confirming increasing advantage in the long-haul regime. Extension to WDM through an XPM-aware per-channel formulation achieves MAE of 0.113–0.419 dB across 150–500 km link lengths.

In addition to its role in enabling DSP-based longitudinal power profile estimation, the fiber Kerr-effect nonlinearity is shown to give rise to an inherent physical-layer security vulnerability in coherent WDM systems. We show that an eavesdropper co-tenanting a shared fiber — transmitting a continuous-wave probe at a wavelength adjacent to the legitimate signal — can capture the XPM-induced waveform at the fiber output and apply a bidirectional gated recurrent unit neural network, trained on split-step Fourier method simulation data, to reconstruct the transmitted symbol sequence without physical fiber access and without perturbing the legitimate signal. This eavesdropping mechanism is experimentally validated using a commercial Ciena WaveLogic-Ai coherent transceiver for ASK, BPSK, QPSK, and 16-QAM modulation formats at 4.26 GBaud and 8.56 GBaud over one- and two-span 75 km fiber systems, achieving zero symbol errors under high-OSNR conditions. Noise-aware training over OSNR from 20 to 60 dB maintains symbol error rate below 10⁻² for OSNR above 25–30 dB.

Together, these three contributions demonstrate that the coherent fiber optic system is a versatile physical instrument extending well beyond its role as a data transmission medium. The coherent receiver infrastructure — deployed for high-order modulation and data recovery — simultaneously enables the high-power OFC laser to serve as a practical multi-wavelength transmitter source, and provides the complex field measurement capability through which fiber Kerr-effect nonlinearity can be exploited constructively for distributed link monitoring and, as a direct consequence, reveals an inherent physical-layer security exposure in shared fiber infrastructure. This unified perspective on the coherent system as both a transmission platform and a general-purpose measurement instrument has direct relevance to the design of spectrally efficient, self-monitoring, and physically secure optical interconnects for next-generation AI computing networks.

Advances in sensing, communication, and intelligent control have transformed power systems into data-driven smart grids, where forecasting and intelligent decision-making are essential components. Modern smart grids include distributed energy resources (DERs), renewable generation, battery energy storage systems, and large numbers of grid-edge devices that continuously generate time-series data. At the same time, increasing renewable penetration introduces substantial uncertainty in generation, net load, and market operations, while communication networks impose bandwidth, latency, and reliability constraints on timely data delivery. This dissertation addresses how time-series forecasting, data compression, and task-oriented wireless communication can be jointly designed for smart grid applications.

First, we study weather-aware distributed energy management in prosumer-centric microgrids and show that incorporating day-ahead weather information into decision-making improves battery dispatch and reduces the impact of renewable uncertainty. Second, we introduce forecasting-aware energy management in both wholesale and retail electricity markets, highlighting how renewable generation forecasting affects pricing, scheduling, and uncertainty mitigation. Third, we develop and evaluate deep learning methods for renewable generation forecasting, showing that Transformer-based models outperform recurrent baselines such as RNN and LSTM for wind and solar prediction tasks.

Building on this forecasting foundation, we develop a communication-efficient forecasting framework in which high-dimensional smart grid measurements are compressed into low-dimensional latent representations before transmission. This framework is extended into a task-oriented communication system that jointly optimizes data relevance and information timeliness, so that the receiver obtains compressed updates that remain useful for downstream forecasting tasks. Finally, we extend this framework to a distributed multi-node uplink setting, where multiple grid sensors share a bandwidth-limited channel, and develop scheduling policy that improves both the timeliness and task-relevance of received updates.

Alzheimer’s Disease (AD) is a progressive, irreversible neurodegenerative disorder and the leading cause of dementia worldwide, affecting an estimated 55 million people globally. The window of opportunity for intervention is demonstrably narrow, making reliable early-stage detection a clinical and scientific imperative. While current diagnostic techniques such as neuroimaging and cerebrospinal fluid (CSF) biomarkers carry well-defined limitations in scalability, cost, and access equity, speech has emerged as a compelling non-invasive proxy for cognitive function evaluation.

This work presents a novel approach for using acoustic feature selection as a decision-making technique and implements it using deep reinforcement learning. Specifically, we use a Deep-Q-Network (DQN) agent to navigate a high dimensional feature space of over 6,000 acoustic features extracted using the openSMILE toolkit, dynamically constructing maximally discriminative and non-redundant features subsets. In order to capture the latent structural dependencies among

acoustic features which classifier and wrapper methods have difficulty to model, we introduce the Graph Convolutional Network (GCN) based correlation awareness feature representation layer that operates as an auxiliary input to the DQN state encoder. Post selection interpretability is reinforced through TF-IDF weighting and K-means clustering which together yield both feature level and cluster level explanations that are clinically actionable. The framework is evaluated across five classifiers, namely, support vector machines (SVM), logistic regression, XGBoost, random forest, and feedforward neural network. We use 10-fold stratified cross-validation on established benchmarks of datasets, including DementiaBank Pitt Corpus, Ivanova, and ADReSS challenge data. The proposed approach is benchmarked against state-of-the-art feature selection methods such as LASSO, Recursive feature selection, and mutual information selectors. This research contributes to three primary intellectual advances: (1) a graph augmented state representation that encodes inter-feature relational structure within a reinforcement learning agent, (2) a clinically interpretable pipeline that bridges the gap between algorithmic performance and translational utility, and (3) multilingual data approach for the reinforcement learning agent framework. This study has direct implications for equitable, low-cost and scalable AD screening in both clinical and community settings.

Federated learning (FL) has emerged as a promising distributed machine learning framework that enables multiple devices to collaboratively train models without sharing raw data, thereby preserving privacy and reducing the need for centralized data collection. However, deploying FL in practical wireless environments introduces two major challenges. First, the data generated across distributed devices are often heterogeneous and non-IID, which makes a single global model insufficient for many users. Second, learning performance in wireless systems is strongly affected by communication constraints such as interference, unreliable channels, and
dynamic resource availability. This PhD research aims to address these challenges by bridging FL methods and wireless networks.

In the first thrust, we develop personalized and adaptive FL methods given the underlying wireless link conditions. To this end, we propose channel-aware neighbor selection and similarity-aware aggregation in wireless device-to-device (D2D) learning environments. We further investigate the impacts of partial model update reception on FL performance. The overarching goal of the first thrust is to enhance FL performance under wireless constraints. Next, we investigate the opposite direction and raise the question: How can FL-based distributed optimization be used for the design of next-generation wireless systems? To this end, we investigate communication-aware participation optimization in vehicular networks, where wireless resource allocation affects the number of clients that can successfully contribute to FL. We further extend this direction to integrated sensing and communication (ISAC) systems,
where personalized FL (PFL) is used to support distributed beamforming optimization with joint sensing and communication objectives.

Overall, this research establishes a unified framework for bridging FL and wireless networks. As a future direction, this work will be extended to more realistic ISAC settings with dynamic spectrum access, where communication, sensing, scheduling, and learning performance must be considered jointly.