31825 research outputs found
Sort by
Energy-efficient and Accuracy-aware DNN Inference with IoT Device-edge Collaboration
No full textThis article was originally published in IEEE Transactions on Services Computing. The version of record is available at: https://doi.org/10.1109/TSC.2025.3536311.
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
This article will be embargoed until 01/30/2027.Due to the limited energy and computing resources of Internet of Things (IoT) devices, the collaboration of IoT devices and edge servers is considered to handle the complex deep neural network (DNN) inference tasks. However, the heterogeneity of IoT devices and the various accuracy requirements of inference tasks make it difficult to deploy all the DNN models in edge servers. Moreover, a large-scale data transmission is engaged in collaborative inference, resulting in an increased demand on spectrum resource and energy consumption. To address these issues, in this paper, we first design an accuracy-aware multi-branch DNN inference model and quantify the relationship between branch selection and inference accuracy. Then, based on the multi-branch DNN model, we aim to minimize the energy consumption of devices by jointly optimizing the selection of DNN branches and partition layers, as well as the computing and communication resources allocation. The proposed problem is a mixed-integer nonlinear programming problem. We propose a hierarchical approach to decompose the problem, and then solve it with a proportional integral derivative based searching algorithm. Experimental results demonstrate our proposed scheme has better inference performance and can reduce the total energy consumption up to 65.3%, compared to other collaboration schemesThis work was supported in part by the National Natural Science Foundation of China under Grants 62302450, 62122069 and 62071431, in part by the Project Supported by Zhejiang Provincial Natural Science Foundation of China under Grant LQ24F020037, and in part by the National Science Fund for Excellent Young Scholars under Grant 62422112
A Novel ISAR Imaging Algorithm for a Maneuvering Target Based on Generalized Second-Order Time-Scaled Transform
No full textThis article was originally published in IEEE Transactions on Geoscience and Remote Sensing. The version of record is available at: https://doi.org/10.1109/TGRS.2025.3540457.
© 2025 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
This article will be embargoed until 02/11/2027.It is well-known that for a maneuvering target, its ISAR imaging quality may significantly deteriorate by using the classical range-Doppler (RD) algorithm. To address this issue, this paper proposes a novel ISAR imaging algorithm based on the generalized second-order time-scaled transform (GSOTST). In the proposed method, the multicomponent cubic phase signal (CPS) modeling is adopted for the radar echo signal after translational motion compensation (TMC) in order to portray the phase change characteristics more accurately. First, the target signal is transformed into the slow-time-delay-time domain using a correlation kernel function (CKF). Subsequently, the non-stationary phase is eliminated in the slow-time-generalized delay-time frequency domain, and the GSOTST is utilized to decouple the temporal variables. Finally, the generalized Fourier transform is performed to transform the signal into the 2-D frequency domain, where the energy of the target signal is integrated into a well-focused 2-D peak, enabling the high-precision target parameter estimation as well as finely focused ISAR imaging. The experimental results from both simulation and real-measured data validate the effectiveness of the proposed algorithm.This work was supported in part by the National Natural Science Foundation Program of China under Grants 62171272, in part by the USCAST2023-29
The role of angiotensin-(1-7) in regulating vascular endothelial function in postmenopausal women
No full textWenner, Megan M.Postmenopausal women (PMW) exhibit reduced endothelial function compared to younger, premenopausal women (YW). This is mainly attributed to the loss of estrogen with menopause but the mechanisms underlying the decline in endothelial function remain unclear. Angiotensin-(1-7) [Ang-(1-7)] induces vasodilation through the Mas receptor (MasR) and has been shown to restore endothelium-dependent dilation in women with endothelial dysfunction. Animal studies suggest menopause and aging reduce vascular sensitivity to Ang-(1-7) which may cause a compensatory upregulation of MasR. However, the role of the Ang-(1-7)/MasR axis has not yet been studied in humans. The central hypothesis of this dissertation was that the Ang-(1-7)/MasR axis is a main regulator of vascular function in women and that there is dysfunction in this pathway that occurs during menopause which leads to the pathologies associated with the onset of CVD. We hypothesize that PMW would have decreased vascular sensitivity to Ang-(1-7), local administration of Ang-(1-7) would improve endothelial function, and PMW would show a compensatory-based upregulation of MasR on endothelial cells. Methods: Blood flow was measured using laser Doppler flowmetry. To assess vascular sensitivity, Ang-(1-7) was locally administered in escalating doses in the presence and absence of L-NAME via cutaneous microdialysis to elicit a dose-dependent response. Dose response curves were fit to a sigmoidal curve and the ED50, slope, area under the curve, and peak response were compared between groups. To assess endothelial function, local heating of the cutaneous circulation to 42°C – which elicits an endothelium-dependent dilation – was performed during microdialysis perfusions of lactated Ringers (control) or Ang-(1-7). All skin blood flow data are expressed as cutaneous vascular conductance as a percentage of the maximum dilation elicited by sodium nitroprusside perfusions with heating to 43°C (CVC%max). Separately, venous endothelial cells were collected from PMW and YW and stained for MasR using immunocytochemistry and are expressed as protein expression arbitrary units (A.U.) following normalization to a positive control. All data are presented as mean+SD and alpha was set to P0.05 for all variables). There was also no significant difference in NO-dependent dilation to Ang-(1-7) between YW and PMW (YW: 7.56+26.36 AUC vs. PMW: 17.40+25.75 AUC; p=0.48). PMW displayed a blunted endothelial function shown by a significantly attenuated response to local heating in the control site (YW: 91.26+4.87 CVC%max vs. PMW: 85.97+5.63 CVC%max; p=0.03), however, there was no impact of Ang-(1-7) on the response to local heating in either group (YW: 89.23+10.35 CVC%max vs. PMW 87.19+9.57 CVC%max; p=0.88). Lastly, there were no differences in endothelial MasR expression between groups (YW: 0.36+0.08 A.U. vs. PMW: 0.35+0.13 A.U.; p=0.77). Conclusion: The Ang-(1-7)/MasR axis does not appear to contribute to endothelial function in PMW as was hypothesized. These data are first to examine the impact of Ang-(1-7)/MasR axis on endothelial function in PMW and provide an important first step for future research examining the relationship between this pathway and the physiological changes that occur with menopause.University of Delaware, Department of Kinesiology and Applied PhysiologyPh.D
The magneto-mechanical coupling of multiphase magnetorheological elastomers
No full textThis article was originally published in Journal of Physics: Condensed Matter. The version of record is available at: https://doi.org/10.1088/1361-648X/adac23.
© 2025 The Author(s). Published by IOP Publishing Lt.
Original content from this work may be used under the terms of the Creative Commons Attribution 4.0 license (https://creativecommons.org/licenses/by/4.0/). Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.Magnetorheological elastomers (MREs) are soft magnetic composites that achieve tunable changes in stiffness and damping in the presence of a magnetic field. Rigid particle composite (RC) MREs have been studied for decades for their potential applications to automotive dampers and robotic systems. Recently, magnetic fluid composite (FC) MREs have been developed which utilize magnetic fluids as inclusions to elastomers. An investigation into how inclusion phase affects magneto-mechanical performance may greatly improve MRE design capabilities. Here we experimentally evaluate the impact of solid and liquid magnetic inclusions on MRE properties, construct a simple model that captures the performance of diverse MRE material architectures, and demonstrate the use of the model to create material design maps relating the material structure, zero-field properties, and applied field to the elastic modulus and specific loss. The magneto-mechanical response is evaluated for three material architectures: RC, FC, and hybrid composite MREs that use solid particles, magnetic fluids, and a combination of the two as inclusions respectively. The model is developed through magnetic and mechanical energy principles, which suggests that the phase of the magnetic inclusions impacts the change in energy density during deformation. We show that the magneto-mechanical coupling factor is dependent on the zero-field properties of the composites, which allows for the development of material design maps to inform the fabrication of MREs based on desired properties.We acknowledge support from the Office of Naval Research Young Investigator Program (YIP) (N000142112699)
Navigation function tuning using randomized algorithms
No full textTanner, Herbert G.This thesis proposes an approach to quantitatively optimize navigation function parameters for robot motion planning in sphere worlds, utilizing randomized algorithms with statistical learning theory. ☐ Our methodology employs a Monte Carlo simulation-based sampling protocol to tackle the intricacies of navigation parameter optimization. Importantly, we use randomized algorithms in conjunction with desired probability levels to determine the appropriate sample sizes. We also introduce two key navigation performance metrics: average maximum curvature difference along paths and average bounded area deviation along a path for assessing the impact of parameter k on navigation smoothness and convergence efficiency. Finally, we identify an optimal k value that balances navigation efficiency and trajectory smoothness under certain probability levels. ☐ Our work validates the efficacy of applying randomized algorithms and statistical learning to navigation function optimization under probabilistic constraints. Despite computational time limitations and inherent simulation uncertainties, this study advances the field by proposing a quantitatively rigorous, probability-based method for automatic navigation function tuning. ☐ This research represents the first data-driven attempt to quantitatively optimize navigation function parameters and achieves state-of-the-art performance. The findings of this thesis contribute to robot motion planning, paving the way for enhanced navigation in complex environments and marking a crucial step towards more adaptable, efficient, and provably correct motion planning methodologies.University of Delaware, Department of Mechanical EngineeringM.S.M.E
Swapping rice for alternative cereals can reduce climate-induced production losses and increase farmer incomes in India
No full textThis article was originally published in Nature Communications. The version of record is available at: https://doi.org/10.1038/s41467-025-57420-6.
© The Author(s) 2025.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.The rising homogeneity of global crop supply has increased vulnerability to climatic and economic disruptions. While substantial work has examined yield variations in relation to climate variability, little is known about the influence of harvested area on production stability. To investigate this, here we take the example of monsoon cereal production in India, which has steadily shifted towards climate-sensitive rice and away from alternative cereals (finger millet, maize, pearl millet, and sorghum). We find that variations in harvested area are significantly associated with current and past price fluctuations for all cereals except rice. This suggests that farmer decisions based on economic factors may exercise great influence in determining variations in harvested area. We also show that optimized allocations of harvested area can reduce climate-induced production loss by 11% or improve farmer net profit by 11% while maintaining calorie production and cropland area. Such improvements would be possible by reducing harvested areas dedicated to rice and increasing areas allocated to alternative cereals. Our findings show that strategies using harvested area to address cereal yield fluctuations and improve farm profits could complement ongoing efforts to improve alternative cereal yields and stabilize cereal production.K.F.D. acknowledges support by the University of Delaware’s General University Research fund. D.W. acknowledges support by the University of Delaware’s Doctoral Fellowship for Excellence
Hierarchical non-covalent interactions in bioinspired peptide-polymer hydrogels and composite networks
No full textKorley, LaShanda T.J.Nature uses physically assembled, hierarchical structures to achieve materials with impressive mechanical properties and responsive functions. Inspired by natural materials that harness interactions of polypeptides to dictate properties such as modulus, toughness, and morphology, this dissertation has implemented multiple strategies to control hierarchical assembly in polymeric hydrogels. Specifically, polypeptide motifs were incorporated into polymer architectures to drive assembly and performance in polymer hydrogels, targeting potential applications in injectable biomaterials and responsive networks. Non-covalent interactions between building blocks of peptide-polymer hydrogels were modulated via multiple pathways, resulting in hydrogels with tunable properties. ☐ First, we examine the impact of peptide motifs poly(e-carbobenzyloxy-L-lysine) and poly(b-benzyl-L-aspartate) in poly(ethylene glycol) (PEG)-based polyureas on polymer hydrogelation and performance. These peptide-polyurea hybrids demonstrated rapid gelation upon addition of water driven by hierarchical assembly of peptide segments in species containing α-helical secondary structures. The mechanical strength of these peptide-polyurea hydrogels could be controlled by altering peptide segment length, and was largely maintained over a wide temperature range from 10- 80 °C. Furthermore, these physically assembled hydrogels demonstrated impressive shear recovery properties, in which the peptide-polyurea network could be recovered within 10 s after shear disruption. This research demonstrated the utility of peptide-polyureas as a robust, dynamic hydrogel platform. ☐ Next, these initial findings were extended through the addition of hydrogen-bonding nanofillers to further modulate the non-covalent interactions of peptide-polyurea hydrogels. Cellulose nanocrystals (CNCs) were incorporated into peptide polyurea hydrogels at various loadings to tailor key hydrogel properties via matrix-filler interactions. The mechanical reinforcement of peptide-polyurea hydrogels was shown to be dependent on both CNC loading and peptide segment length, with storage modulus increasing up to 1825%. Inclusion of CNCs also resulted in temperature-driven stiffening transitions, as well as shifts in the conformations of peptidic domains (α-helices or β-sheets). Nanofiller-matrix interactions also were shown to facilitate network reformation under shear, highlighting the potential for these nanocomposite hydrogels to serve as high-performance injectable materials. This work demonstrates that peptide-CNC interactions can be harnessed to improve the performance of non-covalently assembled, peptide-polymer hydrogels. ☐ Finally, multi-chain, peptide coiled-coil assemblies (bundlemers) were covalently incorporated into a hydrogel network. Alkene reactive sites were placed at controlled positions of the bundlemer sequence, allowing for photopolymerization with PEG diacrylate, resulting in crosslinked bundlemer-PEG hydrogel networks. The sequence position of reactive handles, as well as solution conditions, impacted polymerization kinetics and hydrogel mechanics. Liquid crystalline (LC) assembly formed by association of bundler peptides was achieved. Additionally, the application of shear force during polymerization was shown to drive increased LC formation and alignment in the hybrid hydrogels. Furthermore, it was demonstrated that changes in pH drove disruption of bundlemer LCs, thus revealing the potential for bundlemer-polymer hybrids to display stimuli-responsive behavior. ☐ Overall, this dissertation demonstrates multiple strategies to control non-covalent interactions in peptide-polymer hybrid hydrogels. Through synthetic control over peptide-polymer architectures, tailored matrix-filler interactions, and processing of hydrogel materials, hierarchical self-assembly in these platforms can be harnessed to achieve improved performance of polymeric hydrogels.University of Delaware, Department of Materials Science and EngineeringPh.D
Kinematics, kinetics, and muscle activations during human locomotion over compliant terrains
No full textThis article was originally published in Scientific Data. The version of record is available at: https://doi.org/10.1038/s41597-025-04433-x.
© The Author(s) 2025.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.Walking on compliant terrains, like carpets, grass, and soil, presents a unique challenge, especially for individuals with mobility impairments. In contrast to rigid-ground walking, compliant surfaces alter movement dynamics and increase the risk of falls. Understanding and modeling gait control across such soft and deformable surfaces is thus crucial for maintaining daily mobility. However, access to the necessary equipment for modeling compliant surface walking is limited. Therefore, in this paper, we present the first publicly available biomechanics dataset of 20 individuals walking on terrains of varying compliance, using a unique robotic device, the Variable Stiffness Treadmill 2 (VST 2), designed to simulate walking on adjustable compliant terrain. VST 2 provides a consistent and reproducible environment for studying the biomechanics of walking on such surfaces within laboratory settings. The goal of this dataset is to provide insights into the muscular, kinematic, and kinetic adaptations that occur when humans walk on compliant terrain in order to design better controllers for prosthetic limbs, improve rehabilitation protocols, and develop adaptive assistive devices that can enhance mobility on compliant surfaces.This material is based upon work supported by the National Science Foundation under Grants No. 2020009, 2015786, 2025797, and 2018905 and work supported by the National Institutes of Health Grant No. 1R01HD111071-01
Detection prospects of ultralight scalars with quantum sensing experiments
No full textSafronova, MariannaWith the discovery of the Higgs boson, the Standard Model is complete. However, mysteries such as the nature of dark matter and dark energy remain. The supersymmetry-motivated model of dark matter, weakly interacting massive particles, have thus far evaded detection. This motivates the search for alternative models of dark matter and exploration of associated phenomena. Ultralight scalar dark matter (ULDM), which can be produced in the early universe via the misalignment mechanism, is one such class of theories, which is well motivated due to the ubiquity of scalars in theories beyond the Standard Model (BSM). The goal of this thesis is to explore detection prospects of BSM ultralight scalars as the local dark matter and as originating from astrophysical sources via quantum sensing experiments. We use detailed simulations to explore different quantum metrology algorithms, including dynamical decoupling, to study how to achieve highest sensitivity to ULDM with a nuclear clock experiment. We further consider boson stars composed of ULDM and their explosions called bosenovae, and investigate whether such exotic astrophysical phenomena are detectable using quantum sensors on earth and in space. Finally, we consider general BSM scalar and pseudo-scalar bursts originating from astrophysical sources and study whether they can be used to do multi-messenger astronomy beyond the Standard Model.University of Delaware, Department of Physics and AstronomyPh.D
A Versatile Materials Class for Solution-Processed Opticsand Photonics Based On Titanium Oxide Hydrates andPolyalcohols: A Perspective
No full text©2025 The Author(s).Advanced Materials published by Wiley-VCH GmbH. This is an open access article under the terms of the Creative Commons Attribution License, https://creativecommons.org/licenses/by/4.0/ which permits use distribution and reproduction in any medium, provided the original work is properlycited.
This article was originally published in Advanced Materials. The version of record is available at: https://doi.org/10.1002/adma.202507028The ability to propagate light within a structure comprising a controlled spatial distribution of the refractive index n prompted the telecommunications revolution of the 20th century. More recently, progress with exploiting the flow of light has led to a broad range of light- and heat-management tools, as well as novel quantum devices. This perspective discusses a new versatile class of optical materials based on molecular hybrids of metal oxide hydrates and commodity polymers, such as poly(vinyl alcohol). These fascinating, easy-to-produce materials are examined, and their processing into useful architectures such as photonic crystals is reviewed, with a focus on thin-film optics. Their potential in other areas is also assessed, for instance, for the fabrication of optical microcavities that allow the formation of exciton-polaritons, enabling studies on strong light-matter interactions. Generally, these molecular hybrids open future opportunities in applications like optics, photonics, quantum devices, catalysis, and beyond.V.Q.C. was supported by the National Science Foundation Science and Technology Center (STC) for Integration ofModernOptoelectronic Materials on Demand (IMOD) under award number DMR-2019444. She also had support by the Georgia Tech Quantum Alliance and SPIE Optics and Photonics Education Scholarships for graduate student support. S.B., P.S., and N.S. thank the UK’s Engineering and Physical Sciences Research Council (EPSRC) for funding via the Center for Doctoral Training in Plastic Electronics
Materials, PE-CDT 9EP/G037515/1. Finally, it is important to acknowledge the European Research Council. The Starting Independent Researcher Project under the grant agreement No. 279587 has provided the important monetary support and scientific freedom to start N.S. to work in an entirely new field