pith. sign in

arxiv: 2604.28148 · v2 · submitted 2026-04-30 · 💻 cs.RO · eess.IV· physics.ins-det

Design and Characteristics of a Thin-Film ThermoMesh for the Efficient Embedded Sensing of a Spatio-Temporally Sparse Heat Source

Sajjad Boorghan Farahan , Ahmed Alajlouni , Jingzhou Zhao This is my paper

Pith reviewed 2026-05-07 07:50 UTC · model grok-4.3

classification 💻 cs.RO eess.IVphysics.ins-det
keywords ThermoMeshthermoelectric meshsparse heat sourcenonlinear interlayerthermal sensingembedded sensorVO2negative temperature coefficient
0
0 comments X

The pith

Nonlinear interlayers in a thermoelectric mesh raise minimum sensitivity for single sparse heat sources by up to 14,500 times.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces ThermoMesh, a passive thin-film device that places thermoelectric junctions across a grid and routes signals through resistive interlayers to locate isolated heat sources from boundary voltage readings. It establishes that a simple linear interlayer already makes sensitivity more uniform and lifts the weakest spots by roughly ten times in a 16 by 16 grid. Replacing the interlayer with a strongly nonlinear material produces far larger gains at bigger scales: a high-temperature ceramic layer multiplies the minimum sensitivity by about 14,500 times in a 200 by 200 grid, while a vanadium-dioxide layer that crosses its metal-insulator transition multiplies it by 24 times. These gains translate into accurate recovery of both location and temperature from noisy boundary data, opening the door to low-power embedded thermal sensing in environments where cameras cannot be used.

Core claim

Numerical modeling demonstrates that linear resistive interlayers flatten the sensitivity map and improve its minimum value by a factor of approximately ten for 16 by 16 meshes. Nonlinear temperature-dependent interlayers produce much larger improvements at scale: a ceramic negative-temperature-coefficient layer operating between 973 K and 1273 K yields roughly 14,500 times higher minimum sensitivity than the linear case at 200 by 200 scale, while a VO2 interlayer across its metal-insulator transition between 298 K and 373 K yields a 24-fold improvement. When synthetic single-sparse heat-source data with 40 dB boundary noise is inverted, the VO2 design recovers the source location in 98 % of

What carries the argument

Thermoelectric junctions arranged in a mesh with an intervening linear or nonlinear temperature-dependent resistive interlayer that redistributes heat flow and performs in-sensor compression for 1-sparse events.

If this is right

  • A linear resistive interlayer flattens the sensitivity distribution and raises minimum sensitivity by roughly ten times in a 16 by 16 mesh.
  • A ceramic NTC interlayer operating at 973-1273 K produces approximately 14,500 times higher minimum sensitivity than the linear design at 200 by 200 scale.
  • A VO2 interlayer across its metal-insulator transition yields a 24-fold sensitivity gain and supports 98 percent localization accuracy with 0.23 K mean temperature error at 40 dB SNR.
  • The same architecture delivers a noise-equivalent temperature of 0.07 K for the VO2 case and 1.49 K for the NTC case while remaining fully passive.
  • In-sensor compression from the interlayer enables simultaneous sensing and data reduction for single-event heat sources in harsh environments.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the predicted sensitivity gains hold in hardware, the mesh could be embedded directly into robotic surfaces or industrial components for continuous hot-spot detection without external illumination or active cooling.
  • The same nonlinearity principle might be adapted to other conduction-based sparse-sensing problems, such as pressure or chemical-event localization, where boundary measurements are easier than full-field imaging.
  • Scaling the mesh size further becomes practical once the minimum sensitivity no longer collapses with grid dimension.
  • Real-time reconstruction algorithms could be simplified because the interlayer already compresses the signal before digitization.

Load-bearing premise

The numerical models of heat conduction, thermoelectric voltages, and material nonlinearities accurately represent physical behavior, and every heat source remains strictly one-sparse with no fabrication variations or boundary effects.

What would settle it

Fabricate a physical 16 by 16 or 200 by 200 ThermoMesh prototype, apply a known single heat source at controlled locations, record the boundary voltages, and check whether the measured sensitivity distribution and localization accuracy match the numerical predictions within the reported error bounds.

Figures

Figures reproduced from arXiv: 2604.28148 by Ahmed Alajlouni, Jingzhou Zhao, Sajjad Boorghan Farahan.

Figure 1
Figure 1. Figure 1: ThermoMesh physical architecture and interlayer variants. A mesh of orthogonal Chromel and Alumel lines forms an array of thermocouple junctions. The inset highlights three junction configurations considered in this work: (a) a direct metal–metal crossing without an added interlayer, (b) insertion of a temperature-independent resistive interlayer yielding linear behavior, and (c) incorporation of a tempera… view at source ↗
Figure 2
Figure 2. Figure 2: Ideal-switch equivalent circuit. where the interlayer resistance depends on temperature. 2.3. ThermoMesh with Linear Resistance Interlayer We begin from the baseline ThermoMesh containing only crossed ther￾mocouple wires and no resistive interlayer. Its 2D equivalent circuit, used to assemble the linear measurement model in Eq. (4), is shown in view at source ↗
Figure 3
Figure 3. Figure 3: Equivalent circuit used to represent the baseline ThermoMesh (no interlayer). view at source ↗
Figure 4
Figure 4. Figure 4: Equivalent circuit of ThermoMesh with a linear resistive interlayer. view at source ↗
Figure 5
Figure 5. Figure 5: Sensitivity maps for a 16×16 ThermoMesh. (a) Baseline configuration without a resistive interlayer, showing strong edge-dominated sensitivity and reduced re￾sponse near the center. (b) Corresponding map with a linear resistive interlayer, illustrating redistribution and flattening of sensitivity across the mesh. 19 view at source ↗
Figure 6
Figure 6. Figure 6: Minimum sensitivity versus linear interlayer resistance for a 16 view at source ↗
Figure 7
Figure 7. Figure 7: Minimum sensitivity versus linear interlayer resistance across mesh sizes. For view at source ↗
Figure 8
Figure 8. Figure 8: Equivalent circuit of ThermoMesh with a thermistor at each crossing (the view at source ↗
Figure 9
Figure 9. Figure 9: High-temperature case: minimum sensitivity vs. number of pixels using a ceramic view at source ↗
Figure 10
Figure 10. Figure 10: Low-temperature case: minimum sensitivity vs. number of pixels using a VO view at source ↗
Figure 11
Figure 11. Figure 11: Temperature-dependent behavior of nonlinear interlayers. (a–b) Minimum sensitivity vs. heated-junction temperature for vanadium-oxide (VO2) and ceramic-NTC interlayers. (c–d) Super-linearity exponent κ(∆T)|T =Tamb vs. hot-spot temperature rise for the same interlayers. same sweeps with fixed ambient temperature, is shown in view at source ↗
Figure 12
Figure 12. Figure 12: Windowed Poisson overlap model. Probabilities that the maximum number of simultaneous events within a measurement window is 0, 1, or ≥ 2. (a) Dependence on the window ratio K = τm/τe for a fixed Poisson mean s. (b) Dependence on the Poisson mean s for a fixed window ratio K. These curves quantify the likelihood of sparsity violation due to event overlap. 32 view at source ↗
Figure 13
Figure 13. Figure 13: Conceptual bolometer-style isolated-pixel variant used to support effectively 1-sparse frames. 4. Conclusions and Future Work This paper introduced ThermoMesh, a passive digital—analog thermo￾electric thin-film sensor that maps a spatio–temporally sparse temperature field T to boundary voltages V through Seebeck transduction and a resistive interlayer. Focusing on the single-event (1-sparse) regime, we de… view at source ↗
read the original abstract

This work presents ThermoMesh, a passive thin-film thermoelectric mesh sensor designed to detect and characterize spatio-temporally sparse heat sources through conduction-based thermal imaging. The device integrates thermoelectric junctions with linear or nonlinear interlayer resistive elements to perform simultaneous sensing and in-sensor compression. We focus on the single-event (1-sparse) operation and define four performance metrics: range, efficiency, sensitivity, and accuracy. Numerical modeling shows that a linear resistive interlayer flattens the sensitivity distribution and improves minimum sensitivity by approximately tenfold for a $16\times16$ mesh. Nonlinear temperature-dependent interlayers further enhance minimum sensitivity at scale: a ceramic negative-temperature-coefficient (NTC) layer over 973-1273K yields a $\sim14{,}500\times$ higher minimum sensitivity than the linear design at a $200\times200$ mesh, while a VO$_2$ interlayer modeled across its metal-insulator transition (MIT) over 298-373K yields a $\sim24\times$ improvement. Using synthetic 1-sparse datasets with white boundary-channel noise at a signal-to-noise ratio of 40dB, the VO$_2$ case achieved $98\%$ localization accuracy, a mean absolute temperature error of $0.23$K, and a noise-equivalent temperature (NET) of $0.07$K. For the ceramic-NTC case no localization errors were observed under the tested conditions, with a mean absolute temperature error of $1.83$K and a NET of $1.49$K. These results indicate that ThermoMesh could enable energy-efficient embedded thermal sensing in scenarios where conventional infrared imaging is limited, such as molten-droplet detection or hot-spot monitoring in harsh environments.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 3 minor

Summary. The paper introduces ThermoMesh, a passive thin-film thermoelectric mesh sensor for conduction-based detection and characterization of spatio-temporally sparse (1-sparse) heat sources. It combines thermoelectric junctions with linear or nonlinear temperature-dependent resistive interlayers (NTC ceramic or VO2 across its MIT) to achieve simultaneous sensing and in-sensor compression. Numerical modeling on synthetic 1-sparse datasets with 40 dB additive white boundary-channel noise reports that a linear interlayer improves minimum sensitivity by ~10x at 16x16 scale; an NTC interlayer yields ~14,500x higher minimum sensitivity than linear at 200x200 scale; a VO2 interlayer yields ~24x improvement and, on the synthetic data, achieves 98% localization accuracy, 0.23 K mean absolute temperature error, and 0.07 K NET (NTC case: zero localization errors, 1.83 K MAE, 1.49 K NET). The work positions the architecture for energy-efficient embedded thermal sensing in harsh environments where conventional IR imaging is impractical.

Significance. If the reported sensitivity gains and accuracy metrics hold under real thermal conditions, the nonlinear-interlayer approach offers a promising route to passive, compressed thermal sensing that could reduce power and hardware demands relative to IR cameras in robotics or industrial monitoring applications. The use of material phase transitions (VO2 MIT) and temperature-dependent resistivity to flatten sensitivity distributions is a conceptually interesting in-sensor processing technique. However, because all quantitative results derive from forward simulations of standard thermoelectric/heat-conduction physics on idealized synthetic data, the practical significance remains provisional until hardware validation is provided.

major comments (2)
  1. Abstract and numerical-modeling sections: all quantitative claims (10x, 14,500x, and 24x minimum-sensitivity improvements; 98% localization accuracy; 0.23 K and 1.83 K MAE; 0.07 K and 1.49 K NET) are generated exclusively from forward simulations of 1-sparse heat sources under an assumed 40 dB white-Gaussian boundary-channel noise model. No experimental hardware results, prototype measurements, or calibration against real thermal data are presented. This is load-bearing because any mismatch between the modeled conduction physics, the precise NTC/VO2 resistivity-vs-temperature curves (especially across the MIT), or the noise statistics and real-device behavior would directly rescale the reported sensitivity factors and downstream accuracy metrics.
  2. Synthetic-dataset construction (implicit in the performance-evaluation section): the localization and error figures rest on the strict assumption of perfect 1-sparsity with no boundary effects, fabrication variation, or crosstalk. The manuscript should quantify how violations of this assumption (e.g., two simultaneous sources or realistic sensor nonuniformity) degrade the reported 98% accuracy and sub-Kelvin errors; without such analysis the claims cannot be considered robust.
minor comments (3)
  1. Clarify whether the 40 dB white noise is applied uniformly to all boundary channels or only a subset, and provide the exact noise model equation used in the simulations.
  2. Supply the precise functional forms and literature references for the temperature-dependent resistivity of the NTC ceramic (973-1273 K) and VO2 (298-373 K) layers that were implemented in the model.
  3. The abstract states that the linear interlayer 'flattens the sensitivity distribution'; a brief quantitative illustration (e.g., a plot or table of sensitivity variance before/after) would strengthen this claim.

Simulated Author's Rebuttal

2 responses · 1 unresolved

We thank the referee for the detailed and constructive review. The comments correctly identify that the current results are simulation-based and rest on idealized assumptions. We address each major comment below and outline the revisions we will make to strengthen the manuscript. We agree that additional discussion of limitations and robustness analysis are warranted.

read point-by-point responses

  1. Referee: Abstract and numerical-modeling sections: all quantitative claims (10x, 14,500x, and 24x minimum-sensitivity improvements; 98% localization accuracy; 0.23 K and 1.83 K MAE; 0.07 K and 1.49 K NET) are generated exclusively from forward simulations of 1-sparse heat sources under an assumed 40 dB white-Gaussian boundary-channel noise model. No experimental hardware results, prototype measurements, or calibration against real thermal data are presented. This is load-bearing because any mismatch between the modeled conduction physics, the precise NTC/VO2 resistivity-vs-temperature curves (especially across the MIT), or the noise statistics and real-device behavior would directly rescale the reported sensitivity factors and downstream accuracy metrics.

    Authors: We acknowledge that all quantitative results derive from numerical forward simulations using standard thermoelectric and heat-conduction equations with material properties drawn from published literature. The work is intended as a numerical demonstration of the ThermoMesh concept rather than a hardware validation study. In the revised manuscript we will (1) explicitly state in the abstract and introduction that the reported metrics are simulation-derived, (2) add a dedicated Limitations section that discusses the idealized assumptions on material curves, noise statistics, and perfect 1-sparsity, and (3) note that real-device deviations could rescale the sensitivity gains. These changes will make the provisional nature of the claims clear to readers. revision: partial

  2. Referee: Synthetic-dataset construction (implicit in the performance-evaluation section): the localization and error figures rest on the strict assumption of perfect 1-sparsity with no boundary effects, fabrication variation, or crosstalk. The manuscript should quantify how violations of this assumption (e.g., two simultaneous sources or realistic sensor nonuniformity) degrade the reported 98% accuracy and sub-Kelvin errors; without such analysis the claims cannot be considered robust.

    Authors: We agree that robustness to violations of the 1-sparse assumption is essential for credible claims. In the revised version we will add new simulation results that (a) evaluate performance under 2-sparse heat sources and (b) incorporate realistic fabrication nonuniformity modeled as Gaussian perturbations on interlayer resistances. We will report the resulting degradation in localization accuracy, MAE, and NET for both the VO2 and NTC cases, thereby quantifying the sensitivity of the reported metrics to these practical deviations. revision: yes

standing simulated objections not resolved
  • We are currently unable to provide experimental hardware results, prototype measurements, or calibration against real thermal data, as the present work is limited to numerical modeling and simulation of the proposed architecture.

Circularity Check

0 steps flagged

No circularity: results are forward numerical simulations of standard thermoelectric and heat-conduction equations on synthetic data.

full rationale

The paper computes sensitivity gains, localization accuracy, and NET values exclusively as outputs of forward modeling that solves the standard heat equation coupled to temperature-dependent resistance and Seebeck effects. No equation reduces a reported metric (e.g., the 10× or 14,500× minimum-sensitivity figures) to a quantity that was fitted to that same metric. No self-citation supplies a load-bearing uniqueness theorem, no ansatz is imported from prior author work, and no parameter is fitted on a subset then renamed as a prediction on the same data. The derivation chain is therefore self-contained against external physical benchmarks and does not collapse to its inputs by construction.

Axiom & Free-Parameter Ledger

3 free parameters · 2 axioms · 1 invented entities

The central claims rest on standard physical models of thermoelectric voltage generation and heat conduction applied to a novel mesh geometry. Material properties for VO2 and NTC ceramics are invoked over specific temperature windows; these are treated as inputs rather than derived. No new physical entities are postulated beyond the proposed sensor layout itself.

free parameters (3)
  • mesh size
    16×16 and 200×200 grids chosen to demonstrate scaling of sensitivity improvements
  • signal-to-noise ratio
    Fixed at 40 dB for synthetic dataset generation
  • material transition temperature windows
    973-1273 K for NTC and 298-373 K for VO2 chosen to model nonlinear regimes
axioms (2)
  • standard math Heat flow obeys Fourier's law of conduction
    Invoked throughout the numerical modeling of thermal imaging
  • standard math Thermoelectric junctions produce voltage proportional to local temperature difference
    Basis for all sensing and compression calculations
invented entities (1)
  • ThermoMesh sensor architecture no independent evidence
    purpose: Passive mesh that performs simultaneous thermal sensing and in-sensor compression for 1-sparse sources
    New device concept introduced in the paper; no independent experimental evidence provided

pith-pipeline@v0.9.0 · 5633 in / 1840 out tokens · 80489 ms · 2026-05-07T07:50:27.584154+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

63 extracted references · 63 canonical work pages

  1. [1]

    G. Pope, M. Lerjen, S. M¨ ullener, S. Schl¨ apfer, T. Walti, J. Widmer, C. Studer, Light curtain localization via compressive sensing, in: 2013 IEEE International Conference on Acoustics, Speech and Signal Process- ing, IEEE, 2013, pp. 2558–2562

    work page 2013

  2. [2]

    Liang, J

    X. Liang, J. Tang, Y. Zhong, B. Gao, H. Qian, H. Wu, Physical reservoir computing with emerging electronics, Nature Electronics 7 (3) (2024) 193–206

    work page 2024

  3. [3]

    T. Wan, B. Shao, S. Ma, Y. Zhou, Q. Li, Y. Chai, In-sensor computing: materials, devices, and integration technologies, Advanced materials 35 (37) (2023) 2203830

    work page 2023

  4. [4]

    G. Zhao, Y. Peng, J. Zhu, X. Tang, Z. Yu, In-sensor computing ann capacitive sensors, arXiv preprint arXiv:2405.17295

    work page arXiv

  5. [5]

    Lustig, D

    M. Lustig, D. Donoho, J. M. Pauly, Sparse mri: The application of com- pressed sensing for rapid mr imaging, Magnetic Resonance in Medicine: An Official Journal of the International Society for Magnetic Resonance in Medicine 58 (6) (2007) 1182–1195

    work page 2007

  6. [6]

    Li, L.-C

    M.-Z. Li, L.-C. Guo, G.-L. Ding, K. Zhou, Z.-Y. Xiong, S.-T. Han, Y. Zhou, Inorganic perovskite quantum dot-based strain sensors for data storage and in-sensor computing, ACS Applied Materials & Interfaces 13 (26) (2021) 30861–30873. 40

    work page 2021

  7. [7]

    Akhtar, R

    H. Akhtar, R. Kakarala, Efficient capacitive touch sensing using struc- tured matrices, in: Computational Imaging XIII, Vol. 9401, SPIE, 2015, pp. 220–228

    work page 2015

  8. [8]

    C. Luo, M. A. Borkar, A. J. Redfern, J. H. McClellan, Compressive sensing for sparse touch detection on capacitive touch screens, IEEE Journal on Emerging and Selected Topics in Circuits and Systems 2 (3) (2012) 639–648

    work page 2012

  9. [9]

    Z. Cao, Y. Xu, S. Yu, Z. Huang, Y. Hu, W. Lin, H. Wang, Y. Luo, Y. Zheng, Z. Chen, et al., A programmable electronic skin with event- driven in-sensor touch differential and decision-making, Advanced Func- tional Materials 35 (2) (2025) 2412649

    work page 2025

  10. [10]

    Y. Li, Y. Yang, C. Wang, Y. Dai, X. Yan, D. Kong, Z. Liao, S. Wang, G.-J. Ruan, P. Wang, et al., Massively parallel in-sensor skinomorphic computing, Nature Communications

    work page

  11. [11]

    J. Lin, C. Ma, Blind-label subwavelength ultrasound imaging, Science Advances 11 (5) (2025) eado2826

    work page 2025

  12. [12]

    J. H. Ender, On compressive sensing applied to radar, Signal Processing 90 (5) (2010) 1402–1414

    work page 2010

  13. [13]

    Anitori, M

    L. Anitori, M. Otten, P. Hoogeboom, Detection performance of compres- sive sensing applied to radar, IEEE, 2011

    work page 2011

  14. [14]

    Huang, S

    H. Huang, S. Shi, J. Zha, Y. Xia, H. Wang, P. Yang, L. Zheng, S. Xu, W. Wang, Y. Ren, et al., In-sensor compressing via programmable opto- electronic sensors based on van der waals heterostructures for intelligent machine vision, Nature Communications 16 (1) (2025) 3836

    work page 2025

  15. [15]

    J. Zhao, F. Ye, Where thermomesh meets thermonet: A machine learning based sensor for heat source localization and peak temperature estimation, Sensors and Actuators A: Physical 292 (2019) 30–38. 41

    work page 2019

  16. [16]

    Silva, G

    C. Silva, G. Romano, Thermal analog computing: Application to matrix- vector multiplication with inverse-designed metastructures, Physical Review Applied 25 (1) (2026) 014073

    work page 2026

  17. [17]

    L. Liu, Y. Dou, J. Wang, Y. Zhao, W. Kong, C. Ma, D. He, H. Wang, H. Zhang, A. Chang, et al., Recent advances in flexible temperature sensors: Materials, mechanism, fabrication, and applications, Advanced Science 11 (36) (2024) 2405003

    work page 2024

  18. [18]

    J. A. Hidalgo-L´ opez, J. Romero-S´ anchez, R. Fern´ andez-Ramos, J. F. Mart´ ın-Canales, J. F. R´ ıos-G´ omez, A low-cost, high-accuracy tempera- ture sensor array, Measurement 125 (2018) 425–431

    work page 2018

  19. [19]

    Huang, Z.-K

    C.-C. Huang, Z.-K. Kao, Y.-C. Liao, Flexible miniaturized nickel oxide thermistor arrays via inkjet printing technology, ACS applied materials & interfaces 5 (24) (2013) 12954–12959

    work page 2013

  20. [20]

    Katerinopoulou, P

    D. Katerinopoulou, P. Zalar, J. Sweelssen, G. Kiriakidis, C. Rentrop, P. Groen, G. H. Gelinck, J. van den Brand, E. C. Smits, Large-area all-printed temperature sensing surfaces using novel composite thermistor materials, Advanced Electronic Materials 5 (2) (2019) 1800605

    work page 2019

  21. [21]

    Urban, A

    G. Urban, A. Jachimowicz, F. Kohl, H. Kuttner, F. Olcaytug, H. Kam- per, F. Pittner, E. Mann-Buxbaum, T. Schalkhammer, O. Prohaska, et al., High-resolution thin-film temperature sensor arrays for medical applications, Sensors and Actuators A: Physical 22 (1-3) (1990) 650–654

    work page 1990

  22. [22]

    B¨ ucher, R

    T. B¨ ucher, R. Huber, C. Eschenbaum, A. Mertens, U. Lemmer, H. Am- rouch, Printed temperature sensor array for high-resolution thermal mapping, Scientific reports 12 (1) (2022) 14231

    work page 2022

  23. [23]

    J. Liu, Z. Li, M. Sun, L. Zhou, X. Wu, Y. Lu, Y. Shao, C. Liu, N. Huang, B. Hu, et al., Flexible bioelectronic systems with large-scale temperature 42 sensor arrays for monitoring and treatments of localized wound inflam- mation, Proceedings of the National Academy of Sciences 121 (49) (2024) e2412423121

    work page 2024

  24. [24]

    Nakajima, T

    T. Nakajima, T. Tsuchiya, Ultrathin highly flexible featherweight ceramic temperature sensor arrays, ACS Applied Materials & Interfaces 12 (32) (2020) 36600–36608

    work page 2020

  25. [25]

    R. C. Webb, A. P. Bonifas, A. Behnaz, Y. Zhang, K. J. Yu, H. Cheng, M. Shi, Z. Bian, Z. Liu, Y.-S. Kim, et al., Ultrathin conformal devices for precise and continuous thermal characterization of human skin, Nature materials 12 (10) (2013) 938–944

    work page 2013

  26. [26]

    A. Daus, M. Jaikissoon, A. I. Khan, A. Kumar, R. W. Grady, K. C. Saraswat, E. Pop, Fast-response flexible temperature sensors with atomi- cally thin molybdenum disulfide, Nano Letters 22 (15) (2022) 6135–6140

    work page 2022

  27. [27]

    P. K. Yadav, I. Yadav, B. Ajitha, A. Rajasekar, S. Gupta, Y. A. K. Reddy, Advancements of uncooled infrared microbolometer materials: A review, Sensors and Actuators A: Physical 342 (2022) 113611

    work page 2022

  28. [28]

    W. Luo, S. Zhang, Y. Gao, C. Shen, Review of mechanisms and detection methods of internal short circuits in lithium-ion batteries, Ionics (2025) 1–20

    work page 2025

  29. [29]

    Huang, L

    L. Huang, L. Liu, L. Lu, X. Feng, X. Han, W. Li, M. Zhang, D. Li, X. Liu, D. U. Sauer, et al., A review of the internal short circuit mech- anism in lithium-ion batteries: Inducement, detection and prevention, International Journal of Energy Research 45 (11) (2021) 15797–15831

    work page 2021

  30. [30]

    V. Sala, A. Vandone, F. Mazzucato, M. Banfi, S. Baraldo, A. Valente, Ai-aided thermal imaging with multispectral camera for direct energy deposition, in: 2024 IEEE International Workshop on Metrology for Industry 4.0 & IoT (MetroInd4. 0 & IoT), IEEE, 2024, pp. 150–155. 43

    work page 2024

  31. [31]

    G. A. Johnson, M. M. Dolde, J. T. Zaugg, M. J. Quintana, P. C. Collins, Monitoring, modeling, and statistical analysis in metal additive manu- facturing: A review, Materials 17 (23) (2024) 5872

    work page 2024

  32. [32]

    De Panfilis, A

    S. De Panfilis, A. Filipponi, Nucleation rate of solidification probed by x-ray absorption temperature scans in undercooled liquid metals, Journal of Applied Physics 88 (1) (2000) 562–570

    work page 2000

  33. [33]

    Liu, H.-Y

    Y. Liu, H.-Y. Cheng, J. A. Malen, F. Xiong, Thermoelectric active cooling for transient hot spots in microprocessors, Nature Communications 15 (1) (2024) 4275

    work page 2024

  34. [34]

    K. Yang, X. Chen, T. Zhu, H. Wang, W. Luo, X. Zhu, B. Tao, H. Wang, High-sensitivity atomic layer thermopile heat-flux sensor and its applica- tion in hypersonic low-density wind tunnel tests, IEEE Transactions on Instrumentation and Measurement 73 (2023) 1–7

    work page 2023

  35. [35]

    Shadloo, A

    M. Shadloo, A. Hadjadj, Laminar-turbulent transition in supersonic boundary layers with surface heat transfer: a numerical study, Numerical Heat Transfer, Part A: Applications 72 (1) (2017) 40–53

    work page 2017

  36. [36]

    Simoens, J

    F. Simoens, J. Meilhan, Terahertz real-time imaging uncooled array based on antenna-and cavity-coupled bolometers, Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 372 (2012) (2014) 20130111

    work page 2012

  37. [37]

    B. A. Olshausen, D. J. Field, Sparse coding of sensory inputs, Current opinion in neurobiology 14 (4) (2004) 481–487

    work page 2004

  38. [38]

    Escure, M

    C. Escure, M. Vardelle, P. Fauchais, Experimental and theoretical study of the impact of alumina droplets on cold and hot substrates, Plasma Chemistry and Plasma Processing 23 (2) (2003) 185–221. 44

    work page 2003

  39. [39]

    C. Chen, C. Li, S. Min, Q. Guo, Z. Xia, D. Liu, Z. Ma, F. Xia, Ultrafast silicon nanomembrane microbolometer for long-wavelength infrared light detection, Nano Letters 21 (19) (2021) 8385–8392

    work page 2021

  40. [40]

    H. Choi, X. Li, Fabrication and application of micro thin film thermocou- ples for transient temperature measurement in nanosecond pulsed laser micromachining of nickel, Sensors and Actuators A: Physical 136 (1) (2007) 118–124

    work page 2007

  41. [41]

    Houivet, J

    D. Houivet, J. Bernard, J.-M. Haussonne, High temperature ntc ceramic resistors (ambient–1000 c), Journal of the European Ceramic Society 24 (6) (2004) 1237–1241

    work page 2004

  42. [42]

    X. Ma, X. Liu, H. Li, A. Zhang, M. Huang, Influence of oxygen flow rate on metal–insulator transition of vanadium oxide thin films grown by rf magnetron sputtering, Applied Physics A 123 (3) (2017) 162

    work page 2017

  43. [43]

    H. Cao, X. Yan, Y. Li, L. Stan, W. Chen, N. P. Guisinger, H. Zhou, D. D. Fong, Enhancing the metal–insulator transition in vo2 heterostructures with graphene interlayers, Applied Physics Letters 121 (8)

    work page

  44. [44]

    J. D. Kendall, A. A. Conklin, R. Pantone, J. C. Nino, S. Kumar, Scalable in-memory computing architectures for sparse matrix multiplication, in: 2022 International Electron Devices Meeting (IEDM), IEEE, 2022, pp. 21–6

    work page 2022

  45. [45]

    Lichtsteiner, C

    P. Lichtsteiner, C. Posch, T. Delbruck, A 128 × 128 120 db 15 µs latency asynchronous temporal contrast vision sensor, IEEE journal of solid-state circuits 43 (2) (2008) 566–576

    work page 2008

  46. [46]

    X. Yuan, D. J. Brady, A. K. Katsaggelos, Snapshot compressive imaging: Theory, algorithms, and applications, IEEE Signal Processing Magazine 38 (2) (2021) 65–88. 45

    work page 2021

  47. [47]

    Scellier, A fast algorithm to simulate nonlinear resistive networks, arXiv preprint arXiv:2402.11674

    B. Scellier, A fast algorithm to simulate nonlinear resistive networks, arXiv preprint arXiv:2402.11674

    work page arXiv

  48. [48]

    Scellier, S

    B. Scellier, S. Mishra, Universal approximation theorem for nonlinear resistive networks, Physical Review Applied 23 (4) (2025) 044009

    work page 2025

  49. [49]

    arXiv preprint arXiv:2006.01981 , year=

    J. Kendall, R. Pantone, K. Manickavasagam, Y. Bengio, B. Scellier, Training end-to-end analog neural networks with equilibrium propagation. arxiv 2020, arXiv preprint arXiv:2006.01981

    work page arXiv 2020

  50. [50]

    Aguirre, A

    F. Aguirre, A. Sebastian, M. Le Gallo, W. Song, T. Wang, J. J. Yang, W. Lu, M.-F. Chang, D. Ielmini, Y. Yang, et al., Hardware implementa- tion of memristor-based artificial neural networks, Nature communica- tions 15 (1) (2024) 1974

    work page 2024

  51. [51]

    R. Zhu, S. Lilak, A. Loeffler, J. Lizier, A. Stieg, J. Gimzewski, Z. Kuncic, Online dynamical learning and sequence memory with neuromorphic nanowire networks, Nature Communications 14 (1) (2023) 6697

    work page 2023

  52. [52]

    Z. Qi, L. Mi, H. Qian, W. Zheng, Y. Guo, Y. Chai, Physical reservoir com- puting based on nanoscale materials and devices, Advanced Functional Materials 33 (43) (2023) 2306149

    work page 2023

  53. [53]

    V. M. Patel, R. Chellappa, Sparse representations, compressive sensing and dictionaries for pattern recognition, in: The first Asian conference on pattern recognition, IEEE, 2011, pp. 325–329

    work page 2011

  54. [54]

    Leblebici, S.-M

    Y. Leblebici, S.-M. Kang, CMOS digital integrated circuits: analysis and design, McGraw-Hill New York, 1996

    work page 1996

  55. [55]

    Y. Xie, J. Huang, R. Willett, Change-point detection for high-dimensional time series with missing data, IEEE Journal of Selected Topics in Signal Processing 7 (1) (2012) 12–27. 46

    work page 2012

  56. [56]

    G. E. Karniadakis, I. G. Kevrekidis, L. Lu, P. Perdikaris, S. Wang, L. Yang, Physics-informed machine learning, Nature Reviews Physics 3 (6) (2021) 422–440

    work page 2021

  57. [57]

    Blumensath, Compressed sensing with nonlinear observations and related nonlinear optimization problems, IEEE Transactions on Informa- tion Theory 59 (6) (2013) 3466–3474

    T. Blumensath, Compressed sensing with nonlinear observations and related nonlinear optimization problems, IEEE Transactions on Informa- tion Theory 59 (6) (2013) 3466–3474

    work page 2013

  58. [58]

    Z. Liu, M. Cai, S. Hong, J. Shi, S. Xie, C. Liu, H. Du, J. D. Morin, G. Li, L. Wang, et al., Data-driven inverse design of flexible pressure sensors, Proceedings of the National Academy of Sciences 121 (28) (2024) e2320222121

    work page 2024

  59. [59]

    Kohli, C

    M. Kohli, C. Wuethrich, K. Brooks, B. Willing, M. Forster, P. Muralt, N. Setter, P. Ryser, Pyroelectric thin-film sensor array, Sensors and Actuators A: Physical 60 (1-3) (1997) 147–153

    work page 1997

  60. [60]

    Zhang, F

    H. Zhang, F. Ye, C. Ma, Physics-informed neural networks with architec- tural physics embedding for large-scale wave field reconstruction, arXiv preprint arXiv:2603.02231

    work page arXiv

  61. [61]

    T. Feng, Z. Zhou, P. Wang, Z. Liao, Y. Wang, H. Zhao, W. Zhang, W. Liu, Transverse thermoelectric materials: Recent advances and challenges, Next Energy 3 (2024) 100105

    work page 2024

  62. [62]

    M. Li, H. Wu, E. M. Avery, Z. Qin, D. P. Goronzy, H. D. Nguyen, T. Liu, P. S. Weiss, Y. Hu, Electrically gated molecular thermal switch, Science 382 (6670) (2023) 585–589

    work page 2023

  63. [63]

    G. Ding, H. Li, J. Zhao, K. Zhou, Y. Zhai, Z. Lv, M. Zhang, Y. Yan, S.-T. Han, Y. Zhou, Nanomaterials for flexible neuromorphics, Chemical reviews 124 (22) (2024) 12738–12843. 47

    work page 2024