pith. sign in

arxiv: 2606.13516 · v1 · pith:UWZNKNV3new · submitted 2026-06-11 · 💻 cs.NE

Adaptive-Frequency Resonate-and-Fire Neurons for Spectral Estimation of Streaming Radar Signals

Stefano Chiavazza , Sen Yuan , Marc Geilen , Francesco Fioranelli , Federico Corradi This is my paper

Pith reviewed 2026-06-27 04:43 UTC · model grok-4.3

classification 💻 cs.NE
keywords FMCW radarresonate-and-fire neuronsfrequency estimationneuromorphic signal processingadaptive neuronsmulti-target trackingstreaming data
0
0 comments X

The pith

Adaptive resonate-and-fire neurons estimate FMCW radar target frequencies sample-by-sample with memory scaling only by target count.

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

FMCW radar systems normally apply FFT to blocks of samples to extract range and velocity, which demands storing large data segments. The paper replaces this with adaptive resonate-and-fire neurons that tune their internal frequency to match the strongest components of each incoming sample. Because processing is strictly sample-by-sample, memory grows only with the number of active targets rather than the length of the observation window. A feedback loop is added so that separate neurons can settle on different frequencies when several targets are present. Demonstrations on both simulated signals and measured radar returns show that the neurons track multiple targets while using substantially less memory than conventional block methods.

Core claim

The adaptive-frequency resonate-and-fire neuron, expressed as a discrete-time dynamical system, continuously adjusts its oscillation frequency to the dominant frequency components present in the streaming radar input, thereby directly recovering target range and velocity without first computing a full frequency spectrum. The model processes each sample individually, and an additional feedback mechanism lets multiple neurons converge on distinct frequency components when several targets are present.

What carries the argument

Adaptive-frequency resonate-and-fire (ARF) neurons formulated as a discrete-time dynamical system whose internal frequency is adjusted to lock onto input signal components.

If this is right

  • Memory requirements scale with the number of tracked targets rather than signal length.
  • A feedback mechanism enables multiple neurons to lock on distinct frequency components.
  • The method tracks multiple targets successfully on both simulated and experimental radar data.
  • Memory usage becomes proportional only to the number of tracked targets, fitting resource-constrained and edge radar applications.

Where Pith is reading between the lines

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

  • The sample-by-sample operation could map directly onto low-power neuromorphic hardware for continuous radar monitoring.
  • The same frequency-locking principle might be tested on other streaming signals such as audio or communications waveforms.
  • If the locking remains stable across varying signal-to-noise ratios, latency in multi-target scenes could drop compared with block FFT pipelines.

Load-bearing premise

The introduced feedback mechanism lets multiple neurons lock onto distinct frequencies in multi-target scenes without loss of tracking accuracy or stability.

What would settle it

Apply the model to measured multi-target FMCW signals containing two or more closely spaced frequencies and check whether the neurons maintain separate, accurate locks or whether they merge and tracking error rises above that of an FFT reference.

Figures

Figures reproduced from arXiv: 2606.13516 by Federico Corradi, Francesco Fioranelli, Marc Geilen, Sen Yuan, Stefano Chiavazza.

Figure 1
Figure 1. Figure 1: Working principle of the adaptive resonate-and-fire (ARF) network proposed for this work. [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Mean-field feedback connections scheme. The output signals of all the neurons are summed [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: System dynamics during the simulation. The red and yellow lines represent the real and [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Processing pipeline for concurrent range and velocity estimation using two layers of adap [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Complete processing pipeline for an example with 4 moving targets, showing the connection [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Frequency dynamics for a single neuron processing a single-tone input signal with a Hanning [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Single-neuron performance metrics evaluated across 50,000 simulations with random target [PITH_FULL_IMAGE:figures/full_fig_p010_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Frequency dynamics for multiple neurons. (a) Simulation of two neurons settling onto a [PITH_FULL_IMAGE:figures/full_fig_p011_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Two-target performance metrics using a two-tone input signal. (a) Total convergence error [PITH_FULL_IMAGE:figures/full_fig_p012_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Spike output generation and reconstruction performance metrics. (a) Total number of [PITH_FULL_IMAGE:figures/full_fig_p012_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Experimental hardware setups used for the data collection at TU Delft. (a) Single corner [PITH_FULL_IMAGE:figures/full_fig_p013_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Frequency convergence dynamics evaluated on recorded hardware data. Metric Avg. Value FFT Peak 3.649 m Exact Freq. 3.640 m Recon. Freq. 3.602 m Spike Count 30 [PITH_FULL_IMAGE:figures/full_fig_p013_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Multi-target performance on experimentally recorded data. (a) Top panel shows the [PITH_FULL_IMAGE:figures/full_fig_p017_13.png] view at source ↗
read the original abstract

Frequency Modulated Continuous Wave (FMCW) radar systems traditionally rely on Fourier-based methods, such as the Fast Fourier Transform (FFT), to estimate target range and velocity. While computationally efficient, these approaches require storing and processing large blocks of data, which can become a bottleneck in memory-constrained or low-latency applications. In this work, we propose a neuromorphic-inspired signal processing method based on adaptive resonate-and-fire (ARF) neurons formulated as a discrete-time dynamical system. Each neuron dynamically adjusts its internal frequency to match dominant frequency components of the input radar signal, enabling direct estimation of target ranges and velocities without computing the full frequency spectrum. The proposed model operates in a sample-by-sample fashion, resulting in memory requirements that scale with the number of tracked targets rather than the signal length. A feedback mechanism is also introduced to enable multiple neurons to lock on distinct frequency components in multi-target cases. Results on simulated and experimental data demonstrate that the method can successfully track multiple targets. Compared to conventional FFT-based approaches, the proposed method offers reduced memory usage proportional only to the number of tracked targets, making it suitable for resource-constrained and edge-based radar applications.

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

3 major / 1 minor

Summary. The paper proposes adaptive resonate-and-fire (ARF) neurons, formulated as a discrete-time dynamical system, for spectral estimation in streaming FMCW radar signals. Each neuron adjusts its internal frequency to match dominant components, enabling direct range and velocity estimation without computing the full FFT spectrum. The model operates sample-by-sample with memory scaling proportional to the number of tracked targets rather than signal length; a feedback mechanism is introduced to allow multiple neurons to lock onto distinct frequencies in multi-target scenarios. The abstract asserts successful tracking on simulated and experimental data and reduced memory usage suitable for edge applications.

Significance. If the central claims hold, the work could offer a neuromorphic alternative to block-based FFT methods with substantially lower memory footprint for real-time, resource-constrained radar processing. The sample-by-sample operation and explicit scaling with target count represent a potentially useful conceptual contribution for streaming applications.

major comments (3)
  1. [Abstract] Abstract: the assertion that 'results on simulated and experimental data demonstrate that the method can successfully track multiple targets' supplies no quantitative error metrics, baseline comparisons to FFT, or details on how the adaptation rule and feedback were validated, leaving the central performance claim without visible supporting evidence.
  2. [Abstract] Abstract (feedback mechanism description): no stability bounds, Lyapunov analysis, or minimum frequency-separation condition is provided for the coupled neuron system, which is load-bearing for the claim that multiple neurons reliably lock on distinct components without duplication or drift.
  3. [Abstract] Abstract (memory-scaling claim): the statement that memory 'scales with the number of tracked targets rather than the signal length' rests on the unverified assumption that each ARF neuron maintains fully independent state updated sample-by-sample and that feedback couples them only enough to prevent duplication; no explicit state-update equations or dynamical analysis confirm this independence.
minor comments (1)
  1. [Abstract] The abstract would benefit from a concise statement of the specific radar parameters (chirp rate, sampling frequency, number of targets) used in the reported experiments.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major comment point-by-point below, indicating planned revisions where the manuscript will be updated.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the assertion that 'results on simulated and experimental data demonstrate that the method can successfully track multiple targets' supplies no quantitative error metrics, baseline comparisons to FFT, or details on how the adaptation rule and feedback were validated, leaving the central performance claim without visible supporting evidence.

    Authors: The abstract provides a high-level summary of the claims. The full manuscript contains quantitative error metrics (range/velocity RMSE), direct FFT baseline comparisons, and validation details for the adaptation rule and feedback in Sections 4 (simulations) and 5 (experimental data). To improve visibility of the supporting evidence, we will revise the abstract to include a concise reference to the achieved metrics and comparisons. revision: yes

  2. Referee: [Abstract] Abstract (feedback mechanism description): no stability bounds, Lyapunov analysis, or minimum frequency-separation condition is provided for the coupled neuron system, which is load-bearing for the claim that multiple neurons reliably lock on distinct components without duplication or drift.

    Authors: The feedback mechanism and its empirical behavior are detailed in Section 3 with supporting multi-target simulations. While a full Lyapunov analysis is not included, we provide empirical evidence of stable locking for frequency separations above an observed threshold. We agree a theoretical treatment would be valuable and will add a discussion of the empirical minimum frequency-separation condition and stability observations in the revised manuscript. revision: partial

  3. Referee: [Abstract] Abstract (memory-scaling claim): the statement that memory 'scales with the number of tracked targets rather than the signal length' rests on the unverified assumption that each ARF neuron maintains fully independent state updated sample-by-sample and that feedback couples them only enough to prevent duplication; no explicit state-update equations or dynamical analysis confirm this independence.

    Authors: Explicit per-neuron state-update equations appear in Section 2.1 (Equations 1–4), showing independent frequency, phase, and amplitude states updated sample-by-sample, with feedback acting only as a frequency-adjustment term to avoid duplication. Memory complexity is therefore O(N) for N neurons. We will add an explicit paragraph in Section 2 clarifying state independence and the resulting memory scaling to address the concern. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained

full rationale

The paper introduces a new discrete-time dynamical system for ARF neurons with adaptive frequency and a feedback mechanism. The central claims (sample-by-sample operation, memory scaling with tracked targets rather than signal length, and multi-neuron locking on distinct frequencies) are presented as direct consequences of the proposed formulation and its application to radar signals. No equations, derivations, or self-citations are shown that reduce these claims to fitted parameters, self-definitions, or prior author results by construction. The comparison to FFT is external and falsifiable. This matches the default case of an honest non-finding.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 1 invented entities

Abstract-only review; no explicit free parameters, axioms, or invented entities beyond the core neuron model are stated. The adaptation rule and feedback mechanism are introduced without derivation details or external validation.

invented entities (1)
  • adaptive resonate-and-fire (ARF) neuron no independent evidence
    purpose: Dynamically adjusts internal frequency to match dominant components of the input radar signal for direct range-velocity estimation
    Core modeling construct introduced in the abstract as a discrete-time dynamical system.

pith-pipeline@v0.9.1-grok · 5748 in / 1130 out tokens · 29937 ms · 2026-06-27T04:43:49.589428+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

35 extracted references · 31 canonical work pages

  1. [1]

    Perception and sensing for autonomous vehicles under adverse weather conditions: A survey

    You Zhang, Alexander Carballo, Hanting Yang, and Kazuya Takeda. “Perception and sensing for autonomous vehicles under adverse weather conditions: A survey”. In:ISPRS Journal of Photogrammetry and Remote Sensing196 (Feb. 2023), pp. 146–177. doi:10.1016/j.isprsjprs.2022.12.021

    work page doi:10.1016/j.isprsjprs.2022.12.021 2023

  2. [2]

    Automotive Radar Signal Processing: Research Directions and Practical Challenges

    Florian Engels, Philipp Heidenreich, Markus Wintermantel, Lukas St¨ acker, Muhammad Al Kadi, and Abdelhak M. Zoubir. “Automotive Radar Signal Processing: Research Directions and Practical Challenges”. In:IEEE Journal of Selected Topics in Signal Processing15.4 (2021), pp. 865–878. doi:10.1109/JSTSP.2021.3063666

    work page doi:10.1109/jstsp.2021.3063666 2021

  3. [3]

    1.1 Computing’s energy problem (and what we can do about it)

    Mark Horowitz. “1.1 Computing’s energy problem (and what we can do about it)”. In:2014 IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC). Feb. 2014, pp. 10–14.doi:10.1109/ISSCC.2014.6757323

    work page doi:10.1109/isscc.2014.6757323 2014

  4. [4]

    Ordered statistic CFAR technique - an overview

    Hermann Rohling. “Ordered statistic CFAR technique - an overview”. In:2011 12th International Radar Symposium (IRS). Sept. 2011, pp. 631–638

    2011

  5. [5]

    Memory Bandwidth Efficient Two-Dimensional Fast Fourier Transform Algorithm and Implementation for Large Problem Sizes

    Berkin Akin, Peter A. Milder, Franz Franchetti, and James C. Hoe. “Memory Bandwidth Efficient Two-Dimensional Fast Fourier Transform Algorithm and Implementation for Large Problem Sizes”. In:2012 IEEE 20th International Symposium on Field-Programmable Custom Computing Machines. 2012 IEEE 20th International Symposium on Field-Programmable Custom Computing ...

    work page doi:10.1109/fccm.2012.40 2012

  6. [6]

    Hw/Sw Co-Design technique for 2D fast fourier transform algorithm on Zynq SoC

    Yassin Kortli, Souhir Gabsi, Maher Jridi, Ayman Alfalou, and Mohamed Atri. “Hw/Sw Co-Design technique for 2D fast fourier transform algorithm on Zynq SoC”. In:Integration82 (Jan. 1, 2022), pp. 78–88.doi:10.1016/j.vlsi.2021.09.005

    work page doi:10.1016/j.vlsi.2021.09.005 2022

  7. [7]

    Energy-Efficient Sparse FFT and Compressed Transpose Memory for mmWave FMCW Radar Sensor System

    Tourangbam Harishore Singh, Po-Tsang Huang, Kung-Shuo Kao, Chih-Shiang Cheng, Kuei-Ann Wen, and Li-Chun Wang. “Energy-Efficient Sparse FFT and Compressed Transpose Memory for mmWave FMCW Radar Sensor System”. In:IEEE Transactions on Instrumentation and Measurement73 (2024), pp. 1–11. doi:10.1109/TIM.2024.3385823

    work page doi:10.1109/tim.2024.3385823 2024

  8. [8]

    FourierPIM: High-throughput in-memory Fast Fourier Transform and polynomial multiplication

    Orian Leitersdorf, Yahav Boneh, Gonen Gazit, Ronny Ronen, and Shahar Kvatinsky. “FourierPIM: High-throughput in-memory Fast Fourier Transform and polynomial multiplication”. In:Memories - Materials, Devices, Circuits and Systems4 (July 1, 2023), p. 100034. doi:10.1016/j.memori.2023.100034

    work page doi:10.1016/j.memori.2023.100034 2023

  9. [9]

    Advancing Neuromorphic Computing With Loihi: A Survey of Results and Outlook

    Mike Davies, Andreas Wild, Garrick Orchard, Yulia Sandamirskaya, Gabriel A. Fonseca Guerra, Prasad Joshi, Philipp Plank, and Sumedh R. Risbud. “Advancing Neuromorphic Computing With Loihi: A Survey of Results and Outlook”. In:Proceedings of the IEEE109.5 (May 2021), pp. 911–934. doi:10.1109/JPROC.2021.3067593

    work page doi:10.1109/jproc.2021.3067593 2021

  10. [10]

    and Liu, S.-C

    Giacomo Indiveri and Shih-Chii Liu. “Memory and Information Processing in Neuromorphic Systems”. In:Proceedings of the IEEE103.8 (Aug. 2015), pp. 1379–1397. doi:10.1109/JPROC.2015.2444094

    work page doi:10.1109/jproc.2015.2444094 2015

  11. [11]

    Spiking Neural Network for Fourier Transform and Object Detection for Automotive Radar

    Javier L´ opez-Randulfe, Tobias Duswald, Zhenshan Bing, and Alois Knoll. “Spiking Neural Network for Fourier Transform and Object Detection for Automotive Radar”. In:Frontiers in Neurorobotics15 (June 7, 2021).doi:10.3389/fnbot.2021.688344

    work page doi:10.3389/fnbot.2021.688344 2021

  12. [12]

    Time-Coded Spiking Fourier Transform in Neuromorphic Hardware

    Javier L´ opez-Randulfe, Nico Reeb, Negin Karimi, Chen Liu, Hector A. Gonzalez, Robin Dietrich, Bernhard Vogginger, Christian Mayr, and Alois Knoll. “Time-Coded Spiking Fourier Transform in Neuromorphic Hardware”. In:IEEE Transactions on Computers71.11 (Nov. 2022), pp. 2792–2802. doi:10.1109/TC.2022.3162708

    work page doi:10.1109/tc.2022.3162708 2022

  13. [13]

    SNN-FT: Temporal-Coded Spiking Neural Networks for Fourier Transform

    Shu Wang, Hongzhi Zheng, Yu Chen, Ammar Belatreche, Guoqi Wang, Yi Jin, Jibin Wu, Ming Zhang, Yang Yang, and Hong Li. “SNN-FT: Temporal-Coded Spiking Neural Networks for Fourier Transform”. In:IEEE Transactions on Neural Networks and Learning Systems(2025), pp. 1–13. doi:10.1109/TNNLS.2025.3622893. 18

    work page doi:10.1109/tnnls.2025.3622893 2025

  14. [14]

    Low-Latency Spike-Based Range and Velocity Estimation of FMCW Radar Signals

    Stefano Chiavazza, Sen Yuan, Francesco Fioranelli, and Federico Corradi. “Low-Latency Spike-Based Range and Velocity Estimation of FMCW Radar Signals”. In:2025 22nd European Radar Conference (EuRAD). 2025 22nd European Radar Conference (EuRAD). Sept. 2025, pp. 335–338. doi:10.23919/EuRAD65285.2025.11234227. [15]Resonate-and-fire neurons - ScienceDirect. u...

    work page doi:10.23919/eurad65285.2025.11234227 2025

  15. [15]

    Efficient Neuromorphic Signal Processing with Loihi 2

    Garrick Orchard, E. Paxon Frady, Daniel Ben Dayan Rubin, Sophia Sanborn, Sumit Bam Shrestha, Friedrich T. Sommer, and Mike Davies. “Efficient Neuromorphic Signal Processing with Loihi 2”. In:2021 IEEE Workshop on Signal Processing Systems (SiPS). 2021 IEEE Workshop on Signal Processing Systems (SiPS). Oct. 2021, pp. 254–259. doi:10.1109/SiPS52927.2021.00053

    work page doi:10.1109/sips52927.2021.00053 2021

  16. [16]

    Resonate-and-Fire Neurons for Radar Interference Detection

    Julian Hille, Daniel Auge, Cyprian Grassmann, and Alois Knoll. “Resonate-and-Fire Neurons for Radar Interference Detection”. In:Proceedings of the International Conference on Neuromorphic Systems 2022. ICONS: International Conference on Neuromorphic Systems. Knoxville TN USA: ACM, July 27, 2022, pp. 1–4.doi:10.1145/3546790.3546816

    work page doi:10.1145/3546790.3546816 2022

  17. [17]

    May 2024

    Ahmed Shaaban, Zaid Chaaban, Manfred Strobel, Wolfgang Furtner, Robert Weigel, and Fabian Lurz.Resonate-and-Fire Spiking Neurons for Target Detection and Hand Gesture Recognition: A Hybrid Approach. May 2024. arXiv:2405.19351. url:https://arxiv.org/abs/2405.19351

    arXiv 2024

  18. [18]

    NeuroRadar: A Neuromorphic Radar Sensor for Low-Power IoT Systems

    Kai Zheng, Kun Qian, Timothy Woodford, and Xinyu Zhang. “NeuroRadar: A Neuromorphic Radar Sensor for Low-Power IoT Systems”. In:Proceedings of the 21st ACM Conference on Embedded Networked Sensor Systems (SenSys). ACM, Nov. 2023, pp. 223–236.doi:10.1145/3625687.3625788

    work page doi:10.1145/3625687.3625788 2023

  19. [19]

    Range and angle estimation with spiking neural resonators for FMCW radar

    Nico Reeb, Javier L´ opez-Randulfe, Robin Dietrich, and Alois C. Knoll. “Range and angle estimation with spiking neural resonators for FMCW radar”. In:Neuromorphic Computing and Engineering5.2 (May 2025), p. 024009. doi:10.1088/2634-4386/adcf46

    work page doi:10.1088/2634-4386/adcf46 2025

  20. [20]

    Energy-efficient radar detection with spiking neural resonators via activity-gated sparsity on Intel Loihi 2

    Nico Reeb, Moritz Grabmann, Javier L´ opez-Randulfe, Robin Dietrich, and Alois C. Knoll. “Energy-efficient radar detection with spiking neural resonators via activity-gated sparsity on Intel Loihi 2”. In:Neuromorphic Computing and Engineering(2026). In press. doi:10.1088/2634-4386/ae629d

    work page doi:10.1088/2634-4386/ae629d 2026

  21. [21]

    Neuromorphic computing for radar and radio systems: a survey

    Hanna Hamrell, Thomas Sj¨ ogren, and Hannes Ovr´ en. “Neuromorphic computing for radar and radio systems: a survey”. In:Neuromorphic Computing and Engineering6.2 (May 2026), p. 022001. doi:10.1088/2634-4386/ae6369

    work page doi:10.1088/2634-4386/ae6369 2026

  22. [22]

    Dynamic Hebbian learning in adaptive frequency oscillators

    Ludovic Righetti, Jonas Buchli, and Auke Jan Ijspeert. “Dynamic Hebbian learning in adaptive frequency oscillators”. In:Physica D: Nonlinear Phenomena216.2 (Apr. 15, 2006), pp. 269–281. doi:10.1016/j.physd.2006.02.009

    work page doi:10.1016/j.physd.2006.02.009 2006

  23. [23]

    Frequency analysis with coupled nonlinear oscillators

    Jonas Buchli, Ludovic Righetti, and Auke Jan Ijspeert. “Frequency analysis with coupled nonlinear oscillators”. In:Physica D: Nonlinear Phenomena237.13 (Aug. 1, 2008), pp. 1705–1718. doi:10.1016/j.physd.2008.01.014

    work page doi:10.1016/j.physd.2008.01.014 2008

  24. [24]

    Simple and Practical Algorithm for Sparse Fourier Transform

    Haitham Hassanieh, Piotr Indyk, Dina Katabi, and Eric Price. “Simple and Practical Algorithm for Sparse Fourier Transform”. In:Proceedings of the Twenty-Third Annual ACM-SIAM Symposium on Discrete Algorithms. Proceedings of the Twenty-Third Annual ACM-SIAM Symposium on Discrete Algorithms. Society for Industrial and Applied Mathematics, Jan. 17, 2012, pp....

    work page doi:10.1137/1.9781611973099.93 2012

  25. [25]

    Adaptive Frequency Model for Phase-Frequency Synchronization in Large Populations of Globally Coupled Nonlinear Oscillators

    J. A. Acebr´ on and R. Spigler. “Adaptive Frequency Model for Phase-Frequency Synchronization in Large Populations of Globally Coupled Nonlinear Oscillators”. In:Physical Review Letters81.11 (Sept. 14, 1998), pp. 2229–2232. doi:10.1103/PhysRevLett.81.2229. 19

    work page doi:10.1103/physrevlett.81.2229 1998

  26. [26]

    An adaptive model for synchrony in the firefly Pteroptyx malaccae

    B. Ermentrout. “An adaptive model for synchrony in the firefly Pteroptyx malaccae”. In:Journal of Mathematical Biology29.6 (June 1, 1991), pp. 571–585. doi:10.1007/BF00164052

    work page doi:10.1007/bf00164052 1991

  27. [27]

    The Duffing adaptive oscillator

    Edmon Perkins. “The Duffing adaptive oscillator”. In:Nonlinear Dynamics113.4 (Feb. 1, 2025), pp. 2987–3000. doi:10.1007/s11071-024-10417-0

    work page doi:10.1007/s11071-024-10417-0 2025

  28. [28]

    Vibrational resonance in coupled self-learning Duffing oscillators and its application in noisy radio frequency signal processing

    Jianhua Yang, Litai Lou, Shangyuan Li, Zhongqiu Wang, and Miguel A F Sanju´ an. “Vibrational resonance in coupled self-learning Duffing oscillators and its application in noisy radio frequency signal processing”. In:Physics Letters A571 (Mar. 2026), p. 131311. doi:10.1016/j.physleta.2025.131311

    work page doi:10.1016/j.physleta.2025.131311 2026

  29. [29]

    Self-learning physical reservoir computer

    Md Raf E Ul Shougat, XiaoFu Li, and Edmon Perkins. “Self-learning physical reservoir computer”. In:Physical Review E109.6 (June 5, 2024), p. 064205.doi:10.1103/PhysRevE.109.064205

    work page doi:10.1103/physreve.109.064205 2024

  30. [30]

    On the VLSI Implementation of Adaptive-Frequency Hopf Oscillator

    Arash Ahmadi, Eduardo Mangieri, Koushik Maharatna, Srinandan Dasmahapatra, and Mark Zwolinski. “On the VLSI Implementation of Adaptive-Frequency Hopf Oscillator”. In:IEEE Transactions on Circuits and Systems I: Regular Papers58.5 (May 2011), pp. 1076–1088.doi:10.1109/TCSI.2010.2092070

    work page doi:10.1109/tcsi.2010.2092070 2011

  31. [31]

    Analog VLSI implementation of resonate-and-fire neuron

    Kenji Nakada, Tetsuya Asai, and Hatsuo Hayashi. “Analog VLSI implementation of resonate-and-fire neuron”. In:International Journal of Neural Systems16.6 (Dec. 2006), pp. 445–456. doi:10.1142/S0129065706000846

    work page doi:10.1142/s0129065706000846 2006

  32. [32]

    Direct Signal Encoding With Analog Resonate-and-Fire Neurons

    Hannah M. Lehmann, Julian Hille, Cyprian Grassmann, and Vadim Issakov. “Direct Signal Encoding With Analog Resonate-and-Fire Neurons”. In:IEEE Access11 (2023), pp. 50052–50063.doi:10.1109/ACCESS.2023.3278098

    work page doi:10.1109/access.2023.3278098 2023

  33. [33]

    A Linear Implementation of an Analog Resonate-and-Fire Neuron

    Angqi Liu, Filippo Moro, Sebastian Billaudelle, and Melika Payvand. A Linear Implementation of an Analog Resonate-and-Fire Neuron. Nov. 2025. arXiv:2511.12297.url:https://arxiv.org/abs/2511.12297

    Pith/arXiv arXiv 2025

  34. [34]

    MIMO Radar for Advanced Driver-Assistance Systems and Autonomous Driving: Advantages and Challenges

    Shunqiao Sun, Athina P. Petropulu, and H. Vincent Poor. “MIMO Radar for Advanced Driver-Assistance Systems and Autonomous Driving: Advantages and Challenges”. In:IEEE Signal Processing Magazine37.4 (2020), pp. 98–117. doi:10.1109/MSP.2020.2978507

    work page doi:10.1109/msp.2020.2978507 2020

  35. [35]

    Texas Instruments Inc.Design guide: TIDEP-01012—Imaging radar using cascaded mmWave sensor reference design (REV. A). 2019. url:https://www.ti.com/lit/ug/tiduen5a/tiduen5a.pdf. 20

    2019