Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

Disaggregated Deep Learning via In-Physics Computing at Radio Frequency

T0 review · 3 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Neural-network math falls to 6 femtojoules per MAC via a radio mixer.

desk verdict The SDR demonstration is real and the math holds up, but the 6.0 fJ/MAC headline is a client-only component estimate that excludes the central broadcast power actually driving the mixer, so the energy claim is not demonstrated as stated. read the letter →

arxiv 2504.17752 v1 pith:RIA2I7ZM submitted 2025-04-24 cs.ET cs.LGeess.SPphysics.app-ph

Zhihui Gao , Sri Krishna Vadlamani , Kfir Sulimany , Dirk Englund , Tingjun Chen This is my paper · ORCID
classification cs.ETcs.LGeess.SPphysics.app-ph
keywords in-physicscomputingradio-frequencyover-the-airmodelbroadcastpassivefrequencymixerenergy-efficientdeeplearninginferencedisaggregatededgematrix-vectormultiplication
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

WISE proposes that deep-learning inference on edge devices does not need the model stored locally or multiplied digitally: a central radio broadcasts the weights as a radio waveform, and each client multiplies that waveform with its own data in a passive frequency mixer, so the matrix-vector product $y = Wx$ emerges from the physics of frequency mixing. The paper argues that this disaggregated, in-physics design removes the dominant memory-access and multiply-accumulate costs from the client, leaving energy dominated by waveform generation, sampling, and decoding. On a software-defined radio testbed with over-the-air broadcasts, WISE reports 95.7% MNIST classification accuracy at 6.0 fJ per real-valued MAC (165.8 TOPS/W), and argues that per-MAC energy improves as the problem size grows, approaching a thermodynamic limit set by thermal noise. Battery-powered cameras, drones, and IoT nodes could then run useful models without storing them, at energy costs two to three orders of magnitude below digital ASICs.

What carries the argument

The load-bearing object is the hybrid convolution theorem for OFDM waveforms: when two time-domain OFDM waveforms are multiplied, their frequency-domain symbol spectra linearly convolve; with W's entries placed on subcarriers and x's entries on every M-th subcarrier, the middle M subcarriers of the convolution equal $y = Wx$. A passive double-balanced diode mixer performs the time-domain multiplication, and channel state information is folded into a precoder $V = W/H$ at the central radio so the client receives undistorted weights. A Zadoff-Chu phase sequence in the activation function spreads the power of the next layer's waveform evenly across the spectrum.

What would settle it

Measure the end-to-end energy of one WISE-R inference including the central radio's transmit power and the client's local-oscillator source, then divide by the number of real-valued MACs; if the total exceeds the quoted 6.0 fJ/MAC by orders of magnitude, the operational-power claim fails. A second check: replace the diode mixer with an ideal analog multiplier and repeat the N=4,096 inner-product test; 5-bit accuracy should improve in the high-SNR regime if thermal noise, not mixer switching, is the floor.

Watch

Extended reading notes

Core claim

The central claim is that a general complex-valued matrix-vector multiplication $y = Wx$ can be carried out in the analog radio-frequency domain. W and x are mapped onto orthogonal subcarriers of two OFDM waveforms; a passive mixer multiplies the time-domain waveforms, and the hybrid convolution theorem makes the spectrum of the product contain the entries of $Wx$ in a narrow band, which the client samples and decodes. The paper reports that this in-physics MVM achieves over 5-bit computing accuracy for inner products up to $N=32{,}768$, and that on MNIST and AudioMNIST the three-client WISE-R implementation reaches 95.7% and 97.2% classification accuracy at 6.0 fJ/MAC and 2.8 fJ/MAC, corresponding to 165.8 and 359.7 TOPS/W. The paper further claims that as $N$ grows, per-MAC energy approaches $e_{\mathrm{tdl}} = \mathrm{SNR}\, k_B T_0 / 4$, the thermal-noise limit of analog hardware, which can sit below the Landauer bound for the same bit accuracy.

Load-bearing premise

The headline energy numbers stand or fall on whether a client's energy really is just the reported waveform-generation, sampling, and decoding costs, with the broadcast, mixer drive, and activation costs negligible once amortized.

Editorial extensions

If this is right

  • A client can run a three-layer fully connected model for image or audio classification without storing the model, receiving freshly broadcast weights on demand.
  • Per-client energy per MAC drops as the layer's input dimension N grows, because ADC sampling and FFT decoding costs amortize; the paper measures 2.4 fJ/MAC at N=4,096 and 1.4 fJ/MAC at N=32,768 for inner products.
  • Computation throughput across U clients grows as $4UB/[(1+\alpha)(1+\beta)]$, so more bandwidth or more clients directly buys more MACs per second.
  • With ideal hardware the per-MAC energy approaches $\mathrm{SNR}\, k_B T_0/4$, a thermal-noise floor that can fall below the Landauer limit for 4-bit and 5-bit accuracy.
  • The same broadcast-and-precode structure extends in the paper's analysis to other MVM-based models, including convolutional neural networks and transformers.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper's headline energy accounting counts only the client side; a system-level fJ/MAC that also charges the central radio's broadcast power, the mixer's local-oscillator drive, and the activation functions will be higher unless those costs are amortized over many clients and inferences.
  • The same frequency-mixer mechanism could be applied to other analog inner-product workloads, such as fixed-code correlation or filter-bank processing, where the broadcast waveform is a known kernel rather than a trained weight matrix.
  • A clean hardware check of the accuracy model is to replace the diode mixer with an ideal analog multiplier: if 5-bit accuracy does not improve in the high-SNR regime, then mixer switching, not thermal noise, is the actual error floor.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The paper proposes WISE, an edge-computing architecture in which a central radio broadcasts frequency-encoded neural-network weights over the air and each client uses a passive RF mixer to compute the matrix-vector product y = Wx in the analog domain. The authors derive an OFDM-based frequency mapping that realizes the MVM as a convolution during frequency mixing, implement a three-client USRP X310 plus Mini-Circuits ZEM-4300+ testbed, and report inner-product RMSE and MNIST/AudioMNIST classification accuracy as functions of SNR. They report 95.7% MNIST accuracy at 25 dB SNR and claim an energy efficiency of 6.0 fJ/MAC, corresponding to 165.8 TOPS/W, computed from an energy model that combines thermal-noise-limited client waveform generation, ADC sampling, and digital FFT decoding.

Significance. The experimental MVM and classification results are externally benchmarked and are not fitted to the energy model; the accuracy degrades with SNR as expected, which lends credibility to the in-physics MVM mechanism. The disaggregated broadcast idea is a worthwhile architectural contribution, and the supplementary material is unusually detailed about the experimental setup. However, the headline energy-efficiency claim is not a measured operation power: it is assembled from literature ADC/ASIC energies and an explicitly client-only model that excludes the central broadcast energy. For the demonstrated three-client setup, the broadcast energy dominates by roughly four orders of magnitude, so the system-level efficiency gain over digital ASICs is not established and the claimed two-to-three order improvement should be substantially qualified.

major comments (3)
  1. [Methods Eq. (2); Supplementary Sections 7B, 12C, 15] The headline claim of 6.0 fJ/MAC and 165.8 TOPS/W is not a measured operation power. Eq. (2) (and Supp. Eq. S105) computes e_mvm from a component model that assumes e_adc = 1 pJ/sample and e_dig = 1 pJ/MAC, and Supplementary Section 7B explicitly excludes the central radio's broadcast energy. In the demonstrated MNIST setup the central radio transmits at about 9 dBm average power for roughly 17.7 ms per inference (Supplementary Sections 12C and 15), consuming about 140 uJ; with three clients this is about 44 pJ per client per real-valued MAC, four orders of magnitude larger than the reported 6 fJ/MAC. The reported numbers are therefore a lower-bound design estimate for an idealized WISE-R ASIC, not the operation power of the demonstrated SDR system. The abstract and Results should be revised to state this, and a system-level energy accounting should be provided as a function of the number of clients, together with the number of clients at which the per-client broadcast cost becomes negligible.
  2. [Methods Eq. (3); Supplementary Section 8F] The 'thermodynamic limit' e_tdl is defined as the N-to-infinity limit of the same energy model in Eq. (2), not as an independently derived physical bound. Claims that WISE 'approaches the thermodynamic limit' and 'surpasses the Landauer bound' are therefore statements about the model's own asymptotic, and they inherit all omissions of the model, including the broadcast energy and the assumed ADC/ASIC energies. This should be relabeled as a model-internal asymptotic limit, and the comparison with Landauer should either be derived from a first-principles accounting of the actual physical resources or removed.
  3. [Results, Fig. 4b; Methods, Eq. (2)] The experimentally reported accuracy values are accompanied by energy-efficiency numbers evaluated from Eq. (2), not from any power measurement of the SDR testbed. Because e3, the digital FFT decoding term at an assumed 1 pJ/MAC, is a substantial fraction of the reported 6.0 fJ/MAC for MNIST, the phrase 'experimental energy efficiency' in the Results section is misleading. The actual USRP X310 power draw is not reported, and the SDR itself does not operate at fJ/MAC scale; the values should be called projected efficiencies from the component model, with all assumed parameters stated in the main text.
minor comments (5)
  1. [Abstract] The abstract says 'ultra-low operation power of 6.0 fJ/MAC'; fJ/MAC is energy per operation, not power. Please use 'operation energy' or 'energy per MAC' consistently.
  2. [Fig. 2e caption] The Fig. 2e caption reports MNIST accuracies of 97.1% to 97.4% across the three clients, while the main text reports 95.7% at 25 dB SNR as the headline accuracy. The caption appears to correspond to the x-precoding scheme in Supplementary Section 13 rather than the W-precoding configuration used for the headline number; please reconcile this inconsistency.
  3. [Supplementary, Eq. (S87)] The Landauer-limit expression appears garbled: 'eLandauer = b2· ln 2·kT0' should be a clearly defined formula in terms of bit precision b. Please correct the typography and define the comparison precisely.
  4. [Fig. S6 caption] The caption contains a typo: 'Mino-Circuits ZEM-4300+' should be 'Mini-Circuits ZEM-4300+'.
  5. [Methods, Eq. (4)] The computation throughput is reported in OPS, but the quantity is real-valued MACs per second; please define OPS explicitly in the main text to avoid confusion with floating-point operations.

Circularity Check

2 steps flagged · score 6.0 of 10

Energy-efficiency headline and the 'thermodynamic limit' are outputs of the paper's own defining equations, while the MVM accuracy benchmarks are external and not circular.

  1. self definitional [Methods, Eq. (3); Introduction/Discussion]
    "With ideal hardware (η = 1, α = β = 0), emvm approaches its thermodynamic limit (TDL) as N→+∞, etdl := limN→+∞ emvm = SNR·kBT0/4. (3)"

    The paper defines 'TDL' as the large-N limit of its own energy model emvm in Eq. (2). The statement that WISE approaches the TDL is therefore true by construction, not by an independent physical derivation. The claimed 28.7 zJ/MAC etdl is simply Eq. (2) evaluated at η=1, α=β=0, with eadc and edig amortized to zero, so the 'thermodynamic limit' and 'surpassing the Landauer bound' conclusions rest on naming the limit of the paper's own formula rather than on an independently derived bound.

  2. other [Methods, Eq. (2); Results and Abstract (energy-efficiency claims)]
    "We evaluate the energy efficiency of WISE given by equation (2), where the SNR values are varied by adjusting the transmit power of x(t). ... we demonstrate that WISE achieves 95.7% image classification accuracy with ultra-low operation power of 6.0 fJ/MAC per client."

    The 6.0 fJ/MAC headline is not a wall-plug measurement; it is the value of Eq. (2) computed from assumed constants (eadc=1 pJ/sample, edig=1 pJ/MAC, η=1.48×10−4) and an empirical SNR. Presenting this model evaluation as 'demonstrated operation power' makes the central efficiency claim an output of its own defining equation. Separately, the 'per client' accounting explicitly excludes the central radio's broadcast power that drives the LO, so the reported number is a component-level model estimate renamed as an experimental demonstration. The accuracy benchmarks themselves are external and remain independent.

full rationale

The core MVM mathematics (frequency-domain encoding, subcarrier mapping, hybrid convolution theorem) is self-contained and does not reduce to its inputs, and the classification accuracies on MNIST and AudioMNIST are externally benchmarked against digital-computing accuracy, so those results are not circular. There is no load-bearing uniqueness theorem and no self-citation chain; the few self-citations (e.g., refs. 8, 35, 36) are not used to justify the central architecture. The circularity is confined to the energy-efficiency claims: the 'thermodynamic limit' is defined as the limit of the paper's own Eq. (2), and the fJ/MAC numbers are evaluations of that same equation presented as measured operation power. The exclusion of central-radio broadcast energy is a scope/correctness issue rather than a circular reduction, but even without it the headline efficiency is a definitional model output. Because the accuracy results are independent and not fitted, the paper is only partially circular, not fully self-referential.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The central energy claim rests on the assumed/chosen overhead coefficients alpha and beta, the measured loss eta, and the assumed ADC/digital MAC energies. No new physical entity is postulated; the WISE radio is a built prototype. The most fragile inputs are eta and the exclusion of broadcast and LO energy from the per-client metric.

free parameters (6)
  • alpha (zero-subcarrier overhead) = 0.33 (M'=6, Delta M=1)
    Chosen empirically from LPF roll-off measurements; inflates E1, E2, and E3, and therefore the headline fJ/MAC.
  • beta (cyclic prefix overhead) = 0.25
    Chosen for timing synchronization tolerance; affects the waveform duration and the energy terms.
  • M' (MVM decomposition output size) = 6
    Selected from a simulation sweep (Fig. S16) as the most energy-efficient decomposition level; changes e3 and alpha/beta.
  • eta (overall hardware efficiency) = 1.48e-4
    Product of measured TX efficiency 10%, mixer insertion loss 11.4 dB, and RX noise figure 16.9 dB; directly scales e1, the dominant term at high SNR.
  • eadc (ADC energy per sample) = 1 pJ/sample
    Assumed from an ADC survey reference; e2 scales linearly with this choice and contributes roughly 1 fJ/MAC on MNIST.
  • edig (digital MAC energy) = 1 pJ/MAC
    Assumed from ASIC references; e3 scales linearly with this choice and contributes roughly 3 fJ/MAC on MNIST.
assumptions (5)
  • domain assumption Passive diode-ring mixer output r_IF(t) proportional to sgn(r_LO(t)) times r_RF(t) behaves as ideal analog multiplication after filtering.
    Supplementary Section 12B states this switching approximation and notes residual noise; the computing accuracy and SNR analysis rely on it.
  • domain assumption Thermal noise at kBT0 = -174 dBm/Hz is the only noise floor, and the SNR captured in the narrowband output predicts computing RMSE.
    Methods Eq. (1)-(3) and Supplementary Sections 7B and 8F construct the entire energy model from this noise model.
  • domain assumption MMSE channel estimation with nearest-neighbor interpolation yields H* approximately equal to H, so W-precoding or x-precoding compensates the wireless channel.
    Supplementary Eqs. (S96)-(S97) and (S102); the accuracy of the broadcast model depends on this approximation.
  • domain assumption Complex-valued FC models trained with |y| and ZC-phase activation have the same task competence as the real-valued LeNet-300-100 baselines.
    Methods 'Dataset and Complex Model Architecture'; the comparison baseline is a digital model chosen by the authors, not the identical network.
  • standard math OFDM DFT/IDFT and the hybrid convolution theorem are valid for the chosen subcarrier mappings.
    Supplementary Sections 5-7; standard signal processing.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Disaggregated Deep Learning via In-Physics Computing at Radio Frequency." pith.science (2026). https://pith.science/paper/RIA2I7ZM

@misc{pith2026250417752,
  author       = {Pith},
  title        = {Pith review of: Disaggregated Deep Learning via In-Physics Computing at Radio Frequency},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/RIA2I7ZM}},
  note         = {Machine review of arXiv:2504.17752}
}
read the original abstract

Modern edge devices, such as cameras, drones, and Internet-of-Things nodes, rely on deep learning to enable a wide range of intelligent applications, including object recognition, environment perception, and autonomous navigation. However, deploying deep learning models directly on the often resource-constrained edge devices demands significant memory footprints and computational power for real-time inference using traditional digital computing architectures. In this paper, we present WISE, a novel computing architecture for wireless edge networks designed to overcome energy constraints in deep learning inference. WISE achieves this goal through two key innovations: disaggregated model access via wireless broadcasting and in-physics computation of general complex-valued matrix-vector multiplications directly at radio frequency. Using a software-defined radio platform with wirelessly broadcast model weights over the air, we demonstrate that WISE achieves 95.7% image classification accuracy with ultra-low operation power of 6.0 fJ/MAC per client, corresponding to a computation efficiency of 165.8 TOPS/W. This approach enables energy-efficient deep learning inference on wirelessly connected edge devices, achieving more than two orders of magnitude improvement in efficiency compared to traditional digital computing.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Machine Intelligence on Wireless Edge Networks

    cs.ET 2025-06 conditional novelty 6.0 of 10

    MIWEN broadcasts neural network weights as radio signals and computes inference by analog multiplication in a device's existing RF mixer, reaching near-digital MNIST accuracy inside an optimal energy window.

Reference graph

Works this paper leans on

57 extracted references · 51 canonical work pages · cited by 1 Pith paper

  1. [1]

    LeCun, Y

    Y. LeCun, Y. Bengio, G. Hinton, Deep learning. Nature 521(7553), 436–444 (2015)

    work page 2015

  2. [2]

    Y. Guo, Y. Liu, A. Oerlemans, S. Lao, S. Wu, M.S. Lew, Deep learning for visual understanding: A review. Neurocomputing 187, 27–48 (2016)

    work page 2016

  3. [3]

    Vinyals, I

    O. Vinyals, I. Babuschkin, W.M. Czarnecki, M. Mathieu, A. Dudzik, J. Chung, D.H. Choi, R. Pow- ell, T. Ewalds, P. Georgiev, et al., Grandmaster level in StarCraft II using multi-agent reinforcement learning. Nature 575(7782), 350–354 (2019)

    work page 2019

  4. [4]

    Rokhsaritalemi, A

    S. Rokhsaritalemi, A. Sadeghi-Niaraki, S.M. Choi, A review on mixed reality: Current trends, challenges and prospects. Applied Sciences 10(2), 636 (2020)

    work page 2020

  5. [5]

    Brown, B

    T. Brown, B. Mann, N. Ryder, M. Subbiah, J.D. Kaplan, P. Dhariwal, A. Neelakantan, P. Shyam, G. Sas- try, A. Askell, et al., Language models are few-shot learners , in Proc. Advances in Neural Information Processing Systems (NeurIPS) (2020)

    work page 2020

  6. [6]

    Touvron, L

    H. Touvron, L. Martin, K. Stone, P. Albert, A. Almahairi, Y. Babaei, N. Bashlykov, S. Batra, P. Bhar- gava, S. Bhosale, et al., Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288 (2023)

    arXiv 2023

  7. [7]

    Horowitz, Computing’s energy problem (and what we can do about it) , in Proc

    M. Horowitz, Computing’s energy problem (and what we can do about it) , in Proc. IEEE International Solid-State Circuits Conference (ISSCC) (2014)

    work page 2014

  8. [8]

    Sulimany, S.K

    K. Sulimany, S.K. Vadlamani, R. Hamerly, P. Iyengar, D. Englund, Quantum-secure multiparty deep learning. arXiv preprint arXiv:2408.05629 (2024)

    arXiv 2024

Show all 57 references
  1. [9]

    Landauer, Irreversibility and heat generation in the computing process

    R. Landauer, Irreversibility and heat generation in the computing process. IBM Journal of Research and Development 5(3), 183–191 (1961)

    1961

  2. [10]

    Miller, Attojoule optoelectronics for low-energy information processing and communications

    D.A. Miller, Attojoule optoelectronics for low-energy information processing and communications. IEEE Journal of Lightwave Technology 35(3), 346–396 (2017)

    2017

  3. [11]

    Hamerly, L

    R. Hamerly, L. Bernstein, A. Sludds, M. Soljaˇ ci´ c, D. Englund, Large-scale optical neural networks based on photoelectric multiplication. Physical Review X 9(2), 021032 (2019)

    2019

  4. [12]

    Shen, N.C

    Y. Shen, N.C. Harris, S. Skirlo, M. Prabhu, T. Baehr-Jones, M. Hochberg, X. Sun, S. Zhao, H. Larochelle, D. Englund, et al., Deep learning with coherent nanophotonic circuits. Nature Photonics 11(7), 441–446 (2017) 50

    2017

  5. [13]

    Y. Zuo, B. Li, Y. Zhao, Y. Jiang, Y.C. Chen, P. Chen, G.B. Jo, J. Liu, S. Du, All-optical neural network with nonlinear activation functions. Optica 6(9), 1132–1137 (2019)

    2019

  6. [14]

    X. Xu, M. Tan, B. Corcoran, J. Wu, A. Boes, T.G. Nguyen, S.T. Chu, B.E. Little, D.G. Hicks, R. Moran- dotti, et al., 11 TOPS photonic convolutional accelerator for optical neural networks. Nature589(7840), 44–51 (2021)

    2021

  7. [15]

    Zhang, M

    H. Zhang, M. Gu, X. Jiang, J. Thompson, H. Cai, S. Paesani, R. Santagati, A. Laing, Y. Zhang, M.H. Yung, et al., An optical neural chip for implementing complex-valued neural network. Nature Communications 12(1), 457 (2021)

    2021

  8. [16]

    Feldmann, N

    J. Feldmann, N. Youngblood, M. Karpov, H. Gehring, X. Li, M. Stappers, M. Le Gallo, X. Fu, A. Lukashchuk, A.S. Raja, et al., Parallel convolutional processing using an integrated photonic tensor core. Nature 589(7840), 52–58 (2021)

    2021

  9. [17]

    Wright, T

    L.G. Wright, T. Onodera, M.M. Stein, T. Wang, D.T. Schachter, Z. Hu, P.L. McMahon, Deep physical neural networks trained with backpropagation. Nature 601(7894), 549–555 (2022)

    2022

  10. [18]

    Wang, M.M

    T. Wang, M.M. Sohoni, L.G. Wright, M.M. Stein, S.Y. Ma, T. Onodera, M.G. Anderson, P.L. McMahon, Image sensing with multilayer nonlinear optical neural networks. Nature Photonics 17(5), 408–415 (2023)

    2023

  11. [19]

    Y. Chen, M. Nazhamaiti, H. Xu, Y. Meng, T. Zhou, G. Li, J. Fan, Q. Wei, J. Wu, F. Qiao, et al., All-analog photoelectronic chip for high-speed vision tasks. Nature 623(7985), 48–57 (2023)

    2023

  12. [20]

    S.Y. Ma, T. Wang, J. Laydevant, L.G. Wright, P.L. McMahon, Quantum-limited stochastic optical neural networks operating at a few quanta per activation. Nature Communications 16(1), 359 (2025)

    2025

  13. [21]

    Agarwal, T.T

    S. Agarwal, T.T. Quach, O. Parekh, A.H. Hsia, E.P. DeBenedictis, C.D. James, M.J. Marinella, J.B. Aimone, Energy scaling advantages of resistive memory crossbar based computation and its application to sparse coding. Frontiers in Neuroscience 9, 484 (2016)

    2016

  14. [22]

    Marinella, S

    M.J. Marinella, S. Agarwal, A. Hsia, I. Richter, R. Jacobs-Gedrim, J. Niroula, S.J. Plimpton, E. Ipek, C.D. James, Multiscale co-design analysis of energy, latency, area, and accuracy of a reram analog neural training accelerator. IEEE Journal on Emerging and Selected Topics i...

    2018

  15. [23]

    Ankit, I.E

    A. Ankit, I.E. Hajj, S.R. Chalamalasetti, G. Ndu, M. Foltin, R.S. Williams, P. Faraboschi, W.m.W. Hwu, J.P. Strachan, K. Roy, et al., PUMA: A programmable ultra-efficient memristor-based accelerator for machine learning inference , in Proc. ACM International Conference on Arch...

    2019

  16. [24]

    Sebastian, M

    A. Sebastian, M. Le Gallo, R. Khaddam-Aljameh, E. Eleftheriou, Memory devices and applications for in-memory computing. Nature Nanotechnology 15(7), 529–544 (2020)

    2020

  17. [25]

    Ambrogio, P

    S. Ambrogio, P. Narayanan, A. Okazaki, A. Fasoli, C. Mackin, K. Hosokawa, A. Nomura, T. Yasuda, A. Chen, A. Friz, et al., An analog-AI chip for energy-efficient speech recognition and transcription. 51 Nature 620(7975), 768–775 (2023)

    2023

  18. [26]

    S. Jung, H. Lee, S. Myung, H. Kim, S.K. Yoon, S.W. Kwon, Y. Ju, M. Kim, W. Yi, S. Han, et al., A crossbar array of magnetoresistive memory devices for in-memory computing. Nature 601(7892), 211–216 (2022)

    2022

  19. [27]

    C. Liu, Q. Ma, Z.J. Luo, Q.R. Hong, Q. Xiao, H.C. Zhang, L. Miao, W.M. Yu, Q. Cheng, L. Li, et al., A programmable diffractive deep neural network based on a digital-coding metasurface array. Nature Electronics 5(2), 113–122 (2022)

    2022

  20. [28]

    Sanchez, G

    S.G. Sanchez, G. Reus-Muns, C. Bocanegra, Y. Li, U. Muncuk, Y. Naderi, Y. Wang, S. Ioannidis, K.R. Chowdhury, AirNN: Over-the-air computation for neural networks via reconfigurable intelligent surfaces. IEEE/ACM Transactions on Networking 31(6), 2470–2482 (2022)

    2022

  21. [29]

    Reus-Muns, K

    G. Reus-Muns, K. Alemdar, S.G. Sanchez, D. Roy, K.R. Chowdhury, AirFC: Designing fully connected layers for neural networks with wireless signals , in Proc. ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Compu...

    2023

  22. [30]

    J. Tong, Z. An, X. Zhao, S. Liao, L. Yang, In-sensor machine learning: Radio frequency neural networks for wireless sensing, in Proc. ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc) (2024)

    2024

  23. [31]

    Cotrufo, S.B

    M. Cotrufo, S.B. Sulejman, L. Wesemann, M.A. Rahman, M. Bhaskaran, A. Roberts, A. Al` u, Recon- figurable image processing metasurfaces with phase-change materials. Nature Communications 15(1), 4483 (2024)

    2024

  24. [32]

    A. Ross, N. Leroux, A. De Riz, D. Markovi´ c, D. Sanz-Hern´ andez, J. Trastoy, P. Bortolotti, D. Querlioz, L. Martins, L. Benetti, et al., Multilayer spintronic neural networks with radiofrequency connections. Nature Nanotechnology 18(11), 1273–1280 (2023)

    2023

  25. [33]

    LeCun, L

    Y. LeCun, L. Bottou, Y. Bengio, P. Haffner, Gradient-based learning applied to document recognition. Proceedings of the IEEE 86(11), 2278–2324 (1998)

    1998

  26. [34]

    Chu, Polyphase codes with good periodic correlation properties

    D. Chu, Polyphase codes with good periodic correlation properties. IEEE Transactions on Information Theory 18(4), 531–532 (1972)

    1972

  27. [35]

    Sludds, S

    A. Sludds, S. Bandyopadhyay, Z. Chen, Z. Zhong, J. Cochrane, L. Bernstein, D. Bunandar, P.B. Dixon, S.A. Hamilton, M. Streshinsky, et al., Delocalized photonic deep learning on the internet’s edge. Science 378(6617), 270–276 (2022)

    2022

  28. [36]

    Davis III, Z

    R. Davis III, Z. Chen, R. Hamerly, D. Englund, RF-photonic deep learning processor with shannon- limited data movement. arXiv preprint arXiv:2207.06883v2 (2024)

    2024 arXiv

  29. [37]

    J. Choi, Z. Wang, S. Venkataramani, P.I.J. Chuang, V. Srinivasan, K. Gopalakrishnan, PACT: Parameterized clipping activation for quantized neural networks. arXiv preprint arXiv:1805.06085 (2018) 52

    2018 arXiv

  30. [38]

    S. Garg, J. Lou, A. Jain, Z. Guo, B.J. Shastri, M. Nahmias, Dynamic precision analog computing for neural networks. IEEE Journal of Selected Topics in Quantum Electronics 29(2: Optical Computing), 1–12 (2022)

    2022

  31. [39]

    Abari, E

    O. Abari, E. Hamed, H. Hassanieh, A. Agarwal, D. Katabi, A.P. Chandrakasan, V. Stojanovic, A 0.75- million-point fourier-transform chip for frequency-sparse signals , in Proc. IEEE International Solid- State Circuits Conference (ISSCC) (2014)

    2014

  32. [40]

    Jouppi, C

    N.P. Jouppi, C. Young, N. Patil, D. Patterson, G. Agrawal, R. Bajwa, S. Bates, S. Bhatia, N. Boden, A. Borchers, et al., In-datacenter performance analysis of a tensor processing unit, in Proc. IEEE/ACM International Symposium on Computer Architecture (ISCA) (2017)

    2017

  33. [41]

    Becker, J

    S. Becker, J. Vielhaben, M. Ackermann, K.R. M¨ uller, S. Lapuschkin, W. Samek, AudioMNIST: exploring explainable artificial intelligence for audio analysis on a simple benchmark. Journal of the Franklin Institute 361(1), 418–428 (2024)

    2024

  34. [42]

    Vaswani, N

    A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A.N. Gomez, L. Kaiser, I. Polosukhin, Attention is all you need. Advances in Neural Information Processing Systems 30 (2017)

    2017

  35. [43]

    Zhang, A.T

    H. Zhang, A.T. Narayanan, H. Herdian, B. Liu, Y. Wang, A. Shirane, K. Okada, 0.2 mW 70 Fs RMS- jitter injection-locked PLL using de-sensitized SSPD-based injecting-time self-alignment achieving 270 dB FoM and−66 dBc reference spur , in Proc. Symposium on VLSI Technology and Ci...

    2019

  36. [44]

    H. Choi, S. Cho, A 7.5 GHz subharmonic injection-locked clock multiplier with a 62.5 MHz reference, −259.7 dB FoMJ, and −56.6 dBc reference spur , in Proc. IEEE International Solid-State Circuits Conference (ISSCC) (2024)

    2024

  37. [45]

    Rashed, A

    K. Rashed, A. Undavalli, S. Chakrabartty, A. Nagulu, A. Natarajan, A scalable and instantaneously wideband RF correlator based on margin computing. IEEE Journal of Solid-State Circuits 59(11), 3612–3626 (2024)

    2024

  38. [46]

    Ghobadi, R

    M. Ghobadi, R. Mahajan, A. Phanishayee, N. Devanur, J. Kulkarni, G. Ranade, P.A. Blanche, H. Raste- garfar, M. Glick, D. Kilper, Projector: Agile reconfigurable data center interconnect , in Proc. ACM SIGCOMM Conference (SIGCOMM) (2016)

    2016

  39. [47]

    Shepard, H

    C. Shepard, H. Yu, N. Anand, E. Li, T. Marzetta, R. Yang, L. Zhong, Argos: Practical many-antenna base stations, in Proc. ACM International Conference on Mobile Computing and Networking (MobiCom) (2012)

    2012

  40. [48]

    Kingma, J

    D.P. Kingma, J. Ba, Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  41. [49]

    Coaxial frequency mixer, 300–4300 MHz

    Mini-Circuits. Coaxial frequency mixer, 300–4300 MHz. https://www.minicircuits.com/pdfs/ ZEM-4300+.pdf

  42. [50]

    https://search.itu.int/history/HistoryDigitalCollectionDocLibrary/ 1.44.48.en.101.pdf (2020) 53

    Radio regulations (edition of 2020). https://search.itu.int/history/HistoryDigitalCollectionDocLibrary/ 1.44.48.en.101.pdf (2020) 53

    2020

  43. [51]

    S. Imai, H. Sato, K. Mukai, H. Okabe, A load-variation-tolerant Doherty power amplifier with dual- adaptive-bias scheme for 5G handsets , in Proc. IEEE International Solid-State Circuits Conference (ISSCC) (2024)

    2024

  44. [52]

    B. Murmann. ADC performance survey (1997-2024). [Online]. Available: https://github.com/ bmurmann/ADC-survey

    1997

  45. [53]

    Oppenheim, Discrete-time signal processing (Pearson Education India, 1999)

    A.V. Oppenheim, Discrete-time signal processing (Pearson Education India, 1999)

    1999

  46. [54]

    Friis, A note on a simple transmission formula

    H.T. Friis, A note on a simple transmission formula. Proceedings of the IRE 34(5), 254–256 (1946)

    1946

  47. [55]

    Bloessl, M

    B. Bloessl, M. Segata, C. Sommer, F. Dressler, An IEEE 802.11a/g/p OFDM receiver for GNU Radio , in Proc. 2nd Workshop on Software Radio Implementation Forum (SRIF) (2013)

    2013

  48. [56]

    Z. Gao, Y. Chen, T. Chen, Swirls: Sniffing Wi-Fi using radios with low sampling rates , in Proc. ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc) (2023)

    2023

  49. [57]

    Montgomery, E.A

    D.C. Montgomery, E.A. Peck, G.G. Vining, Introduction to linear regression analysis (John Wiley & Sons, 2021) 54

    2021

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.