Graduate Student Research Posters

Applied Electromagnetics and Acoustics

Poster #1

Multi-physics modeling of thermoacoustic pulse generation and propagation during pulsed microwave ablation of tissue

Audrey Evans (Advisors: Profs. Susan Hagness & Chu Ma)

Microwave-induced thermoacoustic (TA) imaging is a potential alternative to conventional real-time imaging methods for monitoring microwave ablation (MWA). In this study, we develop a multi-physics model for the generation and propagation of microwave-induced TA signals during pulsed MWA. Our model couples electromagnetics, heat transfer, and acoustics physics. We compare simulation and experimental results for a pulsed MWA system wherein a coaxial MWA antenna is used to heat water. The simulated and experimentally measured TA signals for this configuration are in good qualitative agreement. This multi-physics modeling tool is valuable for understanding the fundamentals of TA signal generation and propagation from within an evolving ablation zone.

Advances in Electroporation for Future Gene Therapy Application

Yizhou Yao (Advisors: Profs. Susan Hagness & John Booske)

Electroporation is a dynamic phenomenon in which pores form in the cell membrane due to an externally applied pulsed electric field. Reversible electroporation is a highly efficient technique for permeabilizing cell membranes and introducing exogenous substances into the cell, including drugs, DNA, and other normally membrane impermeable substances. Experimental results show its great potential to increase the cell transduction rate in gene therapy.

Far-field Acoustic Subwavelength Imaging with Structured Illumination

Jinuan Lin (Advisor: Prof. Chu Ma)

In the past decade, subwavelength imaging technologies have attracted a lot of interest for breaking the diffraction limit. While localization-based methods are able to image isolated subwavelength objects, differentiating objects with subwavelength distances in the far field is still challenging. To solve this problem, it has been proposed in optical microscopy to exploit structured or blind structured illumination, where the spatial frequency mixing between the objects and the illumination converts evanescent waves to propagating waves that can reach far field. Here, we propose an acoustic analogy and demonstrate a framework, both theoretically and experimentally, for performing far-field acoustic subwavelength imaging using structured illumination generated by scattering media with subwavelength features. A scattering medium is placed behind subwavelength objects. The excitation wave is first diffracted by the scattering medium before it further passes through the objects and reaches the receiver in the far field. By utilizing a compressive sensing reconstruction algorithm, the image of the objects can be reconstructed with multiple measurements that are obtained by shifting or rotating the scattering medium. The proposed framework has great potential in medical imaging, non-destructive testing, and underwater acoustic imaging for improving imaging resolution of objects far away from the sensors.

Acoustic Topological Insulators: Design, Fabrication and Characterization

Michael Wang (Advisor: Prof. Chu Ma)

With the development of metamaterials, acoustic topological insulators (TIs) have aroused great academic interest in the field of biology, noise canceling, computation, etc. As a novel structure, a honeycomb lattice, also named phononic ‘graphene’, has been introduced to guide sound waves traveling along a designated direction. Our simulation has shown that honeycomb lattice made of pure metallic or core-shell metal-plastic pillars forms acoustic double Dirac cone edge state by tuning the radius of pillars due to the large impedance contrast of these materials. The edge state helps realize MHz waveguide under water. Phononic ‘graphene’ Tis can be extended to explore the application of ultrasonic quantum behavior in quantum computing field.

Communications, Networks, Privacy, and Security

Secure Pairing Methods for Ubiquitous IoT Devices

Kyuin Lee (Advisor: Prof. Younghyun Kim)

The emergence of the Internet of Things (IoT) and pervasive computing challenges in securely and conveniently connecting devices with limited user interfaces. In particular, discovering and bootstrapping a wireless connection (e.g., Wi-Fi and Bluetooth Low Energy) between two devices that share no prior knowledge, commonly known as pairing, often requires users to go through cumbersome tasks of manually discovering the target device and entering a long passkey. To address this challenge, we propose two usable communication methods designed for mobile and stationary IoT devices. SyncVibe, which is aimed for mobile devices equipped with the vibration motor, leverages the inherent nature of close-proximity transmission of mechanical vibration. By simply keeping two devices in direct contact, the user can bootstrap a secure, high-bandwidth wireless connection without manual pairing procedures. For stationary IoT devices constantly connected to home’s power source, we present VoltKey, a method that transparently and periodically generates network authentication keys for devices, leveraging temporally and spatially unique noise contexts observed in 60 Hz commercial powerline infrastructure. VoltKey is based on the observation that all trusted devices connected to user’s electrical outlets observe noise signal that is temporally and spatially unique to a small set of space, preventing malicious devices without any physical outlet access from pairing with legitimate devices to exchange data. In our experiments under various realistic environments, we demonstrate that both methods can correctly establish a key pair among legitimate devices with over 93% success rate, proving itself as a suitable communication channel for short data transmission for mobile and stationary IoT devices.

Preech: A System for Privacy-Preserving Speech Transcription

Shimaa Ahmed (Advisor: Prof. Kassem Fawaz)

New Advances in machine learning have made Automated Speech Recognition (ASR) systems practical and more scalable. These systems, however, pose serious privacy threats as speech is a rich source of sensitive acoustic and textual information. Although offline and open-source ASR eliminates the privacy risks, its transcription performance is inferior to that of cloud-based ASR systems, especially for real-world use cases. In this work, we propose Preech, an end-to-end speech transcription system which lies at an intermediate point in the privacy-utility spectrum. It protects the acoustic features of the speakers’ voices and protects the privacy of the textual content at an improved performance relative to offline ASR. Additionally, Preech provides several control knobs to allow customizable utility-usability-privacy trade-off. It relies on cloud-based services to transcribe a speech file after applying a series of privacy-preserving operations on the user’s side.

Enhancing Transport Layer Performance in Millimeter Wave Access Networks

Zongshen Wu (Advisor: Prof. Parmesh Ramanathan)

Millimeter wave (mmWave) access networks have the potential to meet the high-throughput and low-latency needs of immersive applications. However, due to the highly directional nature of the mmWave beams and their susceptibility to human-induced blockage, the associated wireless links are vulnerable to large channel fluctuations. These fluctuations result in disproportionately adverse effects on performance of transport layer protocols such as Transmission Control Protocol (TCP). Furthermore, we show in this work, that TCP continues significantly underperforming even when dual connectivity is used to combat these effects. To overcome this challenge, we propose a network layer solution, COded Taking And Giving (COTAG) scheme to sustain low-latency and high-throughput end-to-end TCP performance. In particular, COTAG creates network encoded packets at the network gateway and each mmWave access point (AP) aiming to adaptively take the spare bandwidth on each link for transmission. Further, if one link does not have enough bandwidth, COTAG actively abandons the transmission opportunity via conditionally dropping packets. Consequently, COTAG actively adapts to changes in mmWave link quality and enhances the TCP performance without jeopardizing the latency of immersive connect delivery. To evaluate the effectiveness of the proposed COTAG, we conduct experiments and simulations using off-the-shelf APs and NS-2. The evaluation results show COTAG significantly improves end-to-end TCP performance.

Kalεido: Real-Time Privacy Control for Eye-Tracking Systems

Jingjie Li (Advisor: Prof. Younghyun Kim)

Recent advances in sensing and computing technologies have led to the rise of eye-tracking platforms. Ranging from mobiles to high-end mixed reality headsets, a wide spectrum of interactive systems now employs eye-tracking. However, eye gaze data is a rich source of sensitive information that can reveal an individual’s physiological and psychological traits. Prior approaches to protecting eye-tracking data suffer from two major drawbacks: they are either incompatible with the current eye-tracking ecosystem or provide no formal privacy guarantee. In this paper, we propose Kalεido, an eye-tracking data processing system that (1) provides a formal privacy guarantee, (2) integrates seamlessly with existing eye-tracking ecosystems, and (3) operates in real-time. Kalεido acts as an intermediary protection layer in the software stack of eye-tracking systems. We conduct a comprehensive user study and trace-based analysis to evaluate Kalεido. Our user study shows that the users enjoy a satisfactory level of utility from Kalεido. Additionally, we present empirical evidence of Kalεido’s effectiveness in thwarting real-world attacks on eye-tracking data.

Computer Systems and Architecture

Poster #10

Low-Power Architecture via Approximate Computing and Unary Computing

Di Wu (Advisor: Prof. Joshua San Miguel)

With the flourishing of machine learning workloads on the edge, the demands for low power computer architectures has beening growing more than ever. However, conventional binary architectures, which process multiple parallel data, do not scale well as the precision increases. As an alternative, my research applies different computing schemes, including approximate computing and unary computing to address the scaling probelm in conventional binary architectures to boost the hardware efficiency.

WiChronos: Energy-Efficient Modulation for Long-Range, Large-Scale Wireless Networks

Yaman Sangar (Advisor: Prof. Bhuvana Krishnaswamy)

Power efficient wireless communication has become a bottleneck for long range and large scale deployments. We propose and prototype WiChronos, an energy efficient modulation technique for long range wireless communication in a large scale network. Using off-the-shelf (OTS) components, we demonstrate that WiChronos achieves an impressive 60% improvement in battery life compared to state-of-the-art LPWAN technologies at distances over 800 meters. We also show that more than 1000 WiChronos senders co-exist with less than 5% probability of collisions in low traffic conditions.

Scheduling of Iterative Computing Hardware Units for Accuracy and Energy Efficiency

Setareh Behroozi (Advisor: Prof. Younghyun Kim)

Iterative computing, where the output accuracy gradually improves over multiple iterations, enables dynamic reconfiguration of energy-quality trade-offs by adjusting the latency (i.e., number of iterations). In order to take full advantage of the dynamic reconfigurability of iterative computing hardware, an efficient method for determining the optimal latency is crucial. In this poster, we introduce an integer linear programming (ILP)-based scheduling method to determine the optimal latency of iterative computing hardware. We consider the input-dependence of output accuracy of approximate hardware using data-driven error modeling for accurate quality estimation. The proposed method finds optimal or near-optimal latency with a significant speedup compared to exhaustive search and decision tree-based optimization.

Design and Management of Domain-Specific System-on-Chip

A. Alper Goksoy (Advisor: Prof. Umit Ogras)

Heterogeneous systems-on-chip (SoCs) are highly favorable computing platforms due to their superior performance and energy efficiency potential compared to homogeneous architectures. They can be further tailored to a specific domain of applications by incorporating processing elements (PEs) that accelerate frequently used kernels in these applications. However, this potential is contingent upon optimizing the SoC for the target domain and utilizing its resources effectively at runtime. To this end, system-level design – including scheduling, power-thermal management algorithms and design space exploration studies – plays a crucial role. This poster presents a system-level domain-specific SoC simulation (DS3) framework to address this need. DS3 enables both design space exploration and dynamic resource management for power-performance optimization of domain applications. We showcase Imitation Learning based scheduler along with constraint programming, heuristic and list schedulers for the baseline comparisons using DS3 on real-world applications from wireless communications and radar processing domain. DS3, as well as the reference applications, is shared as open-source software to stimulate research in this area.

Health Monitoring using IoT devices

Sizhe An (Advisor: Prof. Umit Ogras)

There is a growing number of devices that target health and activity monitoring. These devices can enable continuous data collection and analysis outside clinical environments without interfering with daily activities. Hence, they can enable a wide range or smart health applications. However, non-invasive devices like radar devices also attract lots of research interests recently. The user doesn’t have to wear or put on a wearable device, instead, they will be exposed to a radar device. We want to combine the two aspects: wearable, and radar for accurate and reliable data that can be used for diagnosis and decision making. Our goal is to use multiple kinds of IoT devices, do data fusion and finally apply them to different heal applications.

Camouflage – Towards Secure Synthetic Trace Generation

Asmita Pal (Advisor: Prof. Joshua San Miguel)

Traces incorporate architectural behaviors such that they can be used for evaluating branch predictors, prefetchers and even cache replacement policies. Applications are evolving fast these days and incorporating traces from more real workloads becomes necessary to enable more efficient design enhancements. Unfortunately, traces are often kept proprietary since they can give out sensitive information about user inputs to an application. Even when they are released, developers try to obfuscate all information that can make it vulnerable. Such program structure agnostic obfuscation leads to elimination of crucial program behavior. To address these challenges, we propose Camouflage to generate synthetic memory address traces with quantifiable guarantees on information leakage.

Energy Systems

Finding Optimal Power System Frequencies

David Sehloff (Advisors: Profs. Line Road & Giri Venkataramanan)

Developments in grid-scale power electronics have removed the necessity that power systems operate at a single fixed frequency, unlocking substantial advantages. This work quantifies these separate advantages of capacity, flexibility, and control, using a variable frequency circuit model and an AC optimal power flow formulation and solver, demonstrating value for power systems planning and operation.

Analysis and Simulation of High-Speed Integrated Motor Compressor Concept using FSPM topology

Leyue Zhang (Advisor: Prof. Bulent Sarlioglu)

The goal of this project is to continue to investigate and design our integrated motor compressor concept for high-speed operation. By shaping the salient rotor poles of the flux-switching PM machine as airfoils of axial-flow compressor rotor, we can integrate the compressor with the electric machine to achieve higher system efficiency and compactness. So far, the low-speed version of this concept has been designed and investigated. The high-speed machine has its challenges, such as structural strength, high-frequency AC loss, and iron loss. In the design process, both electromagnetic and thermodynamic performances need to be taken into consideration. Especially the blade shape, solidity, blade angles, and angle of attack need to be carefully designed. In this research, first, requirements for the high-speed integrated motor-compressor is being developed. Then, the needs and challenges in the machine design process are also investigated. Analytical analysis and finite element analysis are performed to design the high-speed integrated motor compressor. The future work includes optimizing the preliminary design to achieve better performance, prototyping, and testing the machine.

Design of Wide Bandgap-based Integrated Motor Drives

Renato Amorim Torres (Advisor: Prof. Bulent Sarlioglu)

Integrated motor drives (IMDs) can deliver substantial economic benefits by replacing low-efficiency fixed-speed motors with compact adjustable-speed drives that combine the power electronics and motor into the same physical structure. Major advances in the development of new types of power semiconductor switches made from wide-bandgap (WBG) semiconductor materials such as silicon carbide and gallium nitride open intriguing opportunities for the development of IMDs. These WBG power switches are capable of operating efficiently at high switching frequencies (>100 kHz) and high junction temperatures (>150⁰C) allowing major size reduction of the power electronics. This poster investigates the combination of new WBG power devices for use in integrated motor drive applications. Different design aspects are analyzed and addressed to develop a high performance 3kW IMD.

100 kW Traction Motor Drive System Design with WBG Current Source Inverter

Sangwhee Lee (Advisor: Prof. Bulent Sarlioglu)

A WBG device based, current source inverter machine drive system is proposed for EV traction application

Rethinking the Control of Bearingless Motors

Nathan Petersen (Advisor: Prof. Eric Severson)

Applying physics-based control techniques to magnetically levitated motor systems enables more insight into the fundamental operation of these machines.

Towards electrostatic levitation of rotating machines

Michael Mayberry (Advisors: Profs. Eric Severson & Daniel Ludois)

Contact-less suspension of macro-scale objects has traditionally been reserved almost exclusively for magnetic systems containing copper, steel laminations, and permanent magnets. Electrostatic systems, on the other hand, have been left out primarily due to their much lower force density in atmosphere. However, the ability to levitate non-ferromagnetic materials (e.g., conductors, semiconductors, dielectrics), and recent developments in high-torque density electrostatic motors may allow for practical electrostatic suspensions at meaningful scales. To transport electrostatic forces out of the micro-scale and potentially into the power domain, this work proposes a unique electrostatic bearing that is explored as the dual of a magnetic bearing. The bearing being developed uses a two stator/one rotor that will require no copper coils, electric steel, or rare-earth magnets. To lower the force density disparity between electrostatic and magnetic systems, a vacuum environment is used, that allows for kV order voltages to be applied across small air gaps without breakdown.

Machine Learning, Signal Processing, and Information Theory

Coded-InvNet for Resilient Prediction Serving Systems

Tuan Dinh (Advisor: Prof. Kangwook Lee)

Inspired by a new coded computation algorithm for invertible functions, we propose CodedInvNet, a new approach to design resilient prediction serving systems that can gracefully handle stragglers or node failures. CodedInvNet leverages recent findings in the deep learning literature such as invertible neural networks, Manifold Mixup, and domain translation algorithms, identifying interesting research directions that span across machine learning and systems. Our experimental results show that CodedInvNet can outperform existing approaches, especially when the compute resource overhead is as low as 10%.

Approximate Logic Minimization for Energy-Efficient Combinational Neural Networks

Tianen Chen (Advisor: Prof. Younghyun Kim)

In recent years, neural networks (NNs) are finding their application under more and more severe energy constraints such as inside a sensor. For extremely low-power applications, combinational NNs that consist of purely combinational logic are emerging as a promising solution to remove power-hungry memory accesses. However, it has been explored only for small NNs due to the large size of the resulting logic circuits. In this paper, to push NN-based data processing further into the extreme, we present a novel method for implementing energyefficient combinational NNs through iteratively performing inputaware stochastic logic minimization. We prove the efficacy of the proposed method on the CIFAR-10 dataset, maintaining a competitive accuracy while successfully replacing layers of a VGG-style network with pure combinational logic. Experimental results suggest our novel input-aware stochastic logic minimization method can remove all memory accesses while only using a fraction of the energy.

Breaking Fair Classification with Optimal Poisoning Attacks

Changhun Jo (Advisors: Profs. Kangwook Lee & Sebastien Roch)

Minimizing risk with fairness constraints is one of popular approaches to learning a fair classifier. Recent works showed that this approach yields an unfair classifier if the training set is corrupted. In this work, we characterize the minimum amount of data corruption required for a successful attack. We find a lower bound on this quantity and propose an upper bound by constructing an explicit data poisoning algorithm. Our bounds are tight when the target model is the unconstrained risk minimizer.

GenMix: Generative Model-Based Mixup

Jy-yong Sohn (Advisors: Profs. Kangwook Lee & Dimitris Papailiopoulos)

Mixup is a data augmentation method that augments the training set with convex combinations of samples from two different classes. While mixup improves the prediction performance on various tasks, it fails dramatically on certain data distributions. Inspired by the fact that such failures can be prevented by considering the overall data distribution, we propose GenMix, a new mixup framework that solves this problem with the aid of generative models. We empirically show that GenMix improves the robust accuracy of existing mixup methods by a large margin in strong attack scenarios, without compromising overall generalization performance.

Learning Conditional GANs via Reprogramming Unconditional GANs

Daewon Seo (Advisor: Prof. Kangwook Lee)

Neural reprogramming is the process of repurposing a neural network via input preprocessing. Inspired by this paradigm, we propose RepGAN, a novel framework for designing and training conditional GANs (cGANs). RepGAN first learns an unconditional GAN, and then, for each class, it repurposes the learned generator into a classconditional generator, obtaining a conditional GAN as a whole. This separation approach allows for systematic semi-supervised learning as well as robust training to label noise and class imbalance. Our analysis and experimental results show that RepGAN can learn unbiased class-conditional distributions where existing methods fail to do so.

Optics and Photonics

Tunable metal oxides enabled by focused ion beam implantation

Hongyan Mei (Advisor: Prof. Mikhail Kats)

We demonstrate a convenient method to modify the material properties of two metal oxide systems, zinc oxide and vanadium dioxide, by directly irradiating with gallium ions using a commercial focused-ion beam (FIB) system. The carrier concentration of FIB-doped Ga:ZnO can be modulated by tuning the doping level and adjusting the post-annealing treatment, leading to the tuning of carrier concentration. We also demonstrate how focused ion beam irradiation can be used to engineer the thermally driven insulator-metal transition (IMT) of VO2 by inducing a specific amount of structural defects. These results manifest the potential of using FIB implantation to realize the tunable flat optics such as metasurfaces.

Ultrafast pulse generation in the mid-infrared via modulated emissivity

Yuzhe Xiao (Advisor: Prof. Mikhail Kats)

We demonstrate that mid-infrared pulses can be generated by fast emissivity modulation of semiconductors. Ultrafast visible-frequency pulses were used to pump intrinsic unpatterned silicon and gallium arsenide, resulting in nanosecond-scale thermally emitted pulses.

Passive frequency conversion of ultraviolet images into the visible

Jad Salman (Advisor: Prof. Mikhail Kats)

We demonstrate a passive down-conversion imaging system that converts broadband ultraviolet light to narrow-band green light while preserving the directionality of rays, and thus enabling direct down-conversion imaging. At the same time our system has high transparency in the visible, enabling superimposed visible and ultraviolet imaging. The frequency conversion is performed by a subwavelength-thickness transparent downconverter based on highly efficient CsPbBr3 nanocrystals incorporated into the focal plane of a simple telescope or relay-lens geometry. This demonstration sets the stage for the incorporation of other high-efficiency perovskite nanocrystal materials to enable passive multi-frequency conversion imaging systems.

Temperature-independent spectral radiation using a hysteresis-free solid-to-solid phase transition

Jonathan King (Advisor: Prof. Mikhail Kats)

Thermal emission is the process by which all objects at nonzero temperatures emit light and is well described by the Planck, Kirchhoff, and Stefan–Boltzmann laws. For most solids, the thermally emitted power increases monotonically with temperature in a one-to-one relationship that enables applications such as infrared imaging and noncontact thermometry. Here, we demonstrated ultrathin thermal emitters that violate this one-to-one relationship via the use of samarium nickel oxide (SmNiO3), a strongly correlated quantum material that undergoes a fully reversible, temperature- driven solid-state phase transition. The smooth and hysteresis-free nature of this unique insulator-to-metal phase transition enabled us to engineer the temperature dependence of emissivity to precisely cancel out the intrinsic blackbody profile described by the Stefan– Boltzmann law, for both heating and cooling. Our design results in temperature-independent thermally emitted power within the long-wave atmospheric transparency window (wavelengths of 8 to 14 μm), across a broad temperature range of ∼30 °C, centered around ∼120 °C. The ability to decouple temperature and thermal emission opens a gateway for controlling the visibility of objects to infrared cameras and, more broadly, opportunities for quantum materials in controlling heat transfer.

Tunable infrared optics based on phase-transition materials

Chenghao Wan (Advisor: Prof. Mikhail Kats)

This poster summarizes our group’s work on realizing tunable optical devices in the infrared spectral range using phase-transition materials—e.g., vanadium dioxide (VO2) and samarium nickelate (SmNiO3)—whose optical properties are tunable via external stimuli such as temperature, incident intensity, and strain. By incorporating the phase-transition materials into nanophotonic geometries such as metasurfaces or thin-film assemblies, we have demonstrated a variety of tunable functional devices including optical limiters, nonlinear optical isolators, switchable filters, and thermal emitters. In the poster, we also highlight our capabilities in the experimental characterization of these dynamic optical materials, the design and optimization of optical devices using multiphysics simulations, and the fabrication and testing of these devices.

Optical forces for nanocraft propulsion

Demeng Feng (Advisor: Prof. Mikhail Kats)

Optical forces originate from momentum exchange between incident photons and materials. The understanding and manipulation of optical forces has applications in biology, optomechanical devices, and light sails. Our group’s previous work demonstrated that light sails designed to be propelled by lasers can be made mechanically stable by structuring the sail with dielectric optical metasurfaces; this work required the evaluation the microscopic optical forces exerted on parts of the sail illuminated by a high-power laser. Since then, we have explored best practices of calculating optical forces and force densities, which is an unsettled field in the literature. Recently, we evaluated the optical force on a nanostructured gold plate via several methods based on Lorentz and Einstein-Laub formalisms. We noticed that the abnormal force-density distribution near material boundaries could lead to potential errors, so careful modeling is needed. The ability to calculate optical forces precisely will be required for the design of light sails and other applications that rely on optical forces.

TDB

April Yu (Advisor: Prof. Mikhail Kats)

TDB

Optimization and Control

Poster #37

Efficient Active Machine Learning for Crowdsourcing

Scott Sievert (Advisor: Prof. Rob Nowak)

Social scientists are often interesting in having crowdsourcing users judge the similarity of different items. The easiest query to ask these crowdsourcing users is, for example, “In terms of facial emotion, is face X more similar to face A or B?” One difficulty that arises is there are many queries to ask these crowdsourcing users. There are existing methods in the field of “active machine learning,” but they’re limited by the expensive search required. We have created a system to efficiently deploy these algorithms, and provide experimental results that illustrate the use of our software.

Solid-state Electronics and Quantum Technologies

Flexible Capacitive and Optical Sensors for Medical Applications

Jayer Fernandes (Advisor: Prof. Hongrui Jiang)

There is a strong need for better sensing systems to improve the diagnostic process, in order to screen diseases early and reduce the incidents of false positives. This entails the development of sensors that can improve resident training and improve our understanding of disease mechanisms. The first part of this work looks at the development of flexible capacitive force sensors for the quantification and standardization of the forces and hand motions used during Clinical Breast Examinations, with the aim of improving techniques learned during resident training. The sensor design comprises of an overlapping geometry, along with a patterned dielectric layer, allowing us to easily distinguish between normal and shear forces, as well as discern the direction of the shear force. The second part of this work looks at the development of flexible nanophotonic gratings for the measurement of intraocular strain at the choroid, with the long term aim of aiding in understanding the mechanisms underlying glaucoma. A proof of concept device was fabricated and characterized with an ex-vivo setup, demonstrating it’s ability to provide quantifiable visual information about the change in strain