---
_id: '9828'
abstract:
- lang: eng
  text: Amplitude demodulation is a classical operation used in signal processing.
    For a long time, its effective applications in practice have been limited to narrowband
    signals. In this work, we generalize amplitude demodulation to wideband signals.
    We pose demodulation as a recovery problem of an oversampled corrupted signal
    and introduce special iterative schemes belonging to the family of alternating
    projection algorithms to solve it. Sensibly chosen structural assumptions on the
    demodulation outputs allow us to reveal the high inferential accuracy of the method
    over a rich set of relevant signals. This new approach surpasses current state-of-the-art
    demodulation techniques apt to wideband signals in computational efficiency by
    up to many orders of magnitude with no sacrifice in quality. Such performance
    opens the door for applications of the amplitude demodulation procedure in new
    contexts. In particular, the new method makes online and large-scale offline data
    processing feasible, including the calculation of modulator-carrier pairs in higher
    dimensions and poor sampling conditions, independent of the signal bandwidth.
    We illustrate the utility and specifics of applications of the new method in practice
    by using natural speech and synthetic signals.
acknowledgement: The author thanks his colleagues K. Huszár and G. Tkačik for valuable
  discussions and comments on the manuscript.
article_processing_charge: No
article_type: original
arxiv: 1
author:
- first_name: Mantas
  full_name: Gabrielaitis, Mantas
  id: 4D5B0CBC-F248-11E8-B48F-1D18A9856A87
  last_name: Gabrielaitis
  orcid: 0000-0002-7758-2016
citation:
  ama: Gabrielaitis M. Fast and accurate amplitude demodulation of wideband signals.
    <i>IEEE Transactions on Signal Processing</i>. 2021;69:4039-4054. doi:<a href="https://doi.org/10.1109/TSP.2021.3087899">10.1109/TSP.2021.3087899</a>
  apa: Gabrielaitis, M. (2021). Fast and accurate amplitude demodulation of wideband
    signals. <i>IEEE Transactions on Signal Processing</i>. Institute of Electrical
    and Electronics Engineers. <a href="https://doi.org/10.1109/TSP.2021.3087899">https://doi.org/10.1109/TSP.2021.3087899</a>
  chicago: Gabrielaitis, Mantas. “Fast and Accurate Amplitude Demodulation of Wideband
    Signals.” <i>IEEE Transactions on Signal Processing</i>. Institute of Electrical
    and Electronics Engineers, 2021. <a href="https://doi.org/10.1109/TSP.2021.3087899">https://doi.org/10.1109/TSP.2021.3087899</a>.
  ieee: M. Gabrielaitis, “Fast and accurate amplitude demodulation of wideband signals,”
    <i>IEEE Transactions on Signal Processing</i>, vol. 69. Institute of Electrical
    and Electronics Engineers, pp. 4039–4054, 2021.
  ista: Gabrielaitis M. 2021. Fast and accurate amplitude demodulation of wideband
    signals. IEEE Transactions on Signal Processing. 69, 4039–4054.
  mla: Gabrielaitis, Mantas. “Fast and Accurate Amplitude Demodulation of Wideband
    Signals.” <i>IEEE Transactions on Signal Processing</i>, vol. 69, Institute of
    Electrical and Electronics Engineers, 2021, pp. 4039–54, doi:<a href="https://doi.org/10.1109/TSP.2021.3087899">10.1109/TSP.2021.3087899</a>.
  short: M. Gabrielaitis, IEEE Transactions on Signal Processing 69 (2021) 4039–4054.
corr_author: '1'
date_created: 2021-08-08T22:01:31Z
date_published: 2021-06-09T00:00:00Z
date_updated: 2024-10-09T21:00:43Z
day: '09'
department:
- _id: GaTk
doi: 10.1109/TSP.2021.3087899
external_id:
  arxiv:
  - '2102.04832'
  isi:
  - '000682123900002'
intvolume: '        69'
isi: 1
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://arxiv.org/abs/2102.04832
month: '06'
oa: 1
oa_version: Preprint
page: 4039 - 4054
publication: IEEE Transactions on Signal Processing
publication_identifier:
  eissn:
  - 1941-0476
  issn:
  - 1053-587X
publication_status: published
publisher: Institute of Electrical and Electronics Engineers
quality_controlled: '1'
scopus_import: '1'
status: public
title: Fast and accurate amplitude demodulation of wideband signals
type: journal_article
user_id: 4359f0d1-fa6c-11eb-b949-802e58b17ae8
volume: 69
year: '2021'
...
---
_id: '8268'
abstract:
- lang: eng
  text: 'Modern scientific instruments produce vast amounts of data, which can overwhelm
    the processing ability of computer systems. Lossy compression of data is an intriguing
    solution, but comes with its own drawbacks, such as potential signal loss, and
    the need for careful optimization of the compression ratio. In this work, we focus
    on a setting where this problem is especially acute: compressive sensing frameworks
    for interferometry and medical imaging. We ask the following question: can the
    precision of the data representation be lowered for all inputs, with recovery
    guarantees and practical performance Our first contribution is a theoretical analysis
    of the normalized Iterative Hard Thresholding (IHT) algorithm when all input data,
    meaning both the measurement matrix and the observation vector are quantized aggressively.
    We present a variant of low precision normalized IHT that, under mild conditions,
    can still provide recovery guarantees. The second contribution is the application
    of our quantization framework to radio astronomy and magnetic resonance imaging.
    We show that lowering the precision of the data can significantly accelerate image
    recovery. We evaluate our approach on telescope data and samples of brain images
    using CPU and FPGA implementations achieving up to a 9x speedup with negligible
    loss of recovery quality.'
acknowledgement: The authors would like to thank Dr. Michiel Brentjens at the Netherlands
  Institute for Radio Astronomy (ASTRON) for providing radio interferometer data and
  Dr. Josip Marjanovic and Dr. Franciszek Hennel at the Magnetic Resonance Technology
  of ETH Zurich for providing their insights on the experiments. CZ and the DS3Lab
  gratefully acknowledge the support from the Swiss Data Science Center, Alibaba,
  Google Focused Research Awards, Huawei, MeteoSwiss, Oracle Labs, Swisscom, Zurich
  Insurance, Chinese Scholarship Council, and the Department of Computer Science at
  ETH Zurich.
article_processing_charge: No
article_type: original
arxiv: 1
author:
- first_name: Nezihe Merve
  full_name: Gurel, Nezihe Merve
  last_name: Gurel
- first_name: Kaan
  full_name: Kara, Kaan
  last_name: Kara
- first_name: Alen
  full_name: Stojanov, Alen
  last_name: Stojanov
- first_name: Tyler
  full_name: Smith, Tyler
  last_name: Smith
- first_name: Thomas
  full_name: Lemmin, Thomas
  last_name: Lemmin
- first_name: Dan-Adrian
  full_name: Alistarh, Dan-Adrian
  id: 4A899BFC-F248-11E8-B48F-1D18A9856A87
  last_name: Alistarh
  orcid: 0000-0003-3650-940X
- first_name: Markus
  full_name: Puschel, Markus
  last_name: Puschel
- first_name: Ce
  full_name: Zhang, Ce
  last_name: Zhang
citation:
  ama: 'Gurel NM, Kara K, Stojanov A, et al. Compressive sensing using iterative hard
    thresholding with low precision data representation: Theory and applications.
    <i>IEEE Transactions on Signal Processing</i>. 2020;68:4268-4282. doi:<a href="https://doi.org/10.1109/TSP.2020.3010355">10.1109/TSP.2020.3010355</a>'
  apa: 'Gurel, N. M., Kara, K., Stojanov, A., Smith, T., Lemmin, T., Alistarh, D.-A.,
    … Zhang, C. (2020). Compressive sensing using iterative hard thresholding with
    low precision data representation: Theory and applications. <i>IEEE Transactions
    on Signal Processing</i>. IEEE. <a href="https://doi.org/10.1109/TSP.2020.3010355">https://doi.org/10.1109/TSP.2020.3010355</a>'
  chicago: 'Gurel, Nezihe Merve, Kaan Kara, Alen Stojanov, Tyler Smith, Thomas Lemmin,
    Dan-Adrian Alistarh, Markus Puschel, and Ce Zhang. “Compressive Sensing Using
    Iterative Hard Thresholding with Low Precision Data Representation: Theory and
    Applications.” <i>IEEE Transactions on Signal Processing</i>. IEEE, 2020. <a href="https://doi.org/10.1109/TSP.2020.3010355">https://doi.org/10.1109/TSP.2020.3010355</a>.'
  ieee: 'N. M. Gurel <i>et al.</i>, “Compressive sensing using iterative hard thresholding
    with low precision data representation: Theory and applications,” <i>IEEE Transactions
    on Signal Processing</i>, vol. 68. IEEE, pp. 4268–4282, 2020.'
  ista: 'Gurel NM, Kara K, Stojanov A, Smith T, Lemmin T, Alistarh D-A, Puschel M,
    Zhang C. 2020. Compressive sensing using iterative hard thresholding with low
    precision data representation: Theory and applications. IEEE Transactions on Signal
    Processing. 68, 4268–4282.'
  mla: 'Gurel, Nezihe Merve, et al. “Compressive Sensing Using Iterative Hard Thresholding
    with Low Precision Data Representation: Theory and Applications.” <i>IEEE Transactions
    on Signal Processing</i>, vol. 68, IEEE, 2020, pp. 4268–82, doi:<a href="https://doi.org/10.1109/TSP.2020.3010355">10.1109/TSP.2020.3010355</a>.'
  short: N.M. Gurel, K. Kara, A. Stojanov, T. Smith, T. Lemmin, D.-A. Alistarh, M.
    Puschel, C. Zhang, IEEE Transactions on Signal Processing 68 (2020) 4268–4282.
date_created: 2020-08-16T22:00:56Z
date_published: 2020-07-20T00:00:00Z
date_updated: 2025-07-10T11:55:10Z
day: '20'
department:
- _id: DaAl
doi: 10.1109/TSP.2020.3010355
external_id:
  arxiv:
  - '1802.04907'
  isi:
  - '000562044500001'
intvolume: '        68'
isi: 1
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://arxiv.org/abs/1802.04907
month: '07'
oa: 1
oa_version: Preprint
page: 4268-4282
publication: IEEE Transactions on Signal Processing
publication_identifier:
  eissn:
  - 1941-0476
  issn:
  - 1053-587X
publication_status: published
publisher: IEEE
quality_controlled: '1'
scopus_import: '1'
status: public
title: 'Compressive sensing using iterative hard thresholding with low precision data
  representation: Theory and applications'
type: journal_article
user_id: 2DF688A6-F248-11E8-B48F-1D18A9856A87
volume: 68
year: '2020'
...
---
_id: '18278'
abstract:
- lang: eng
  text: Solving inverse problems with iterative algorithms is popular, especially
    for large data. Due to time constraints, the number of possible iterations is
    usually limited, potentially affecting the achievable accuracy. Given an error
    one is willing to tolerate, an important question is whether it is possible to
    modify the original iterations to obtain faster convergence to a minimizer achieving
    the allowed error without increasing the computational cost of each iteration
    considerably. Relying on recent recovery techniques developed for settings in
    which the desired signal belongs to some low-dimensional set, we show that using
    a coarse estimate of this set may lead to faster convergence at the cost of an
    additional reconstruction error related to the accuracy of the set approximation.
    Our theory ties to recent advances in sparse recovery, compressed sensing, and
    deep learning. Particularly, it may provide a possible explanation to the successful
    approximation of the ℓ 1 -minimization solution by neural networks with layers
    representing iterations, as practiced in the learned iterative shrinkage-thresholding
    algorithm.
article_processing_charge: No
arxiv: 1
author:
- first_name: Raja
  full_name: Giryes, Raja
  last_name: Giryes
- first_name: Yonina C.
  full_name: Eldar, Yonina C.
  last_name: Eldar
- first_name: Alexander
  full_name: Bronstein, Alexander
  id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
  last_name: Bronstein
  orcid: 0000-0001-9699-8730
- first_name: Guillermo
  full_name: Sapiro, Guillermo
  last_name: Sapiro
citation:
  ama: Giryes R, Eldar YC, Bronstein AM, Sapiro G. Tradeoffs between convergence speed
    and reconstruction accuracy in inverse problems. <i>IEEE Transactions on Signal
    Processing</i>. 2018;66(7):1676-1690. doi:<a href="https://doi.org/10.1109/tsp.2018.2791945">10.1109/tsp.2018.2791945</a>
  apa: Giryes, R., Eldar, Y. C., Bronstein, A. M., &#38; Sapiro, G. (2018). Tradeoffs
    between convergence speed and reconstruction accuracy in inverse problems. <i>IEEE
    Transactions on Signal Processing</i>. IEEE. <a href="https://doi.org/10.1109/tsp.2018.2791945">https://doi.org/10.1109/tsp.2018.2791945</a>
  chicago: Giryes, Raja, Yonina C. Eldar, Alex M. Bronstein, and Guillermo Sapiro.
    “Tradeoffs between Convergence Speed and Reconstruction Accuracy in Inverse Problems.”
    <i>IEEE Transactions on Signal Processing</i>. IEEE, 2018. <a href="https://doi.org/10.1109/tsp.2018.2791945">https://doi.org/10.1109/tsp.2018.2791945</a>.
  ieee: R. Giryes, Y. C. Eldar, A. M. Bronstein, and G. Sapiro, “Tradeoffs between
    convergence speed and reconstruction accuracy in inverse problems,” <i>IEEE Transactions
    on Signal Processing</i>, vol. 66, no. 7. IEEE, pp. 1676–1690, 2018.
  ista: Giryes R, Eldar YC, Bronstein AM, Sapiro G. 2018. Tradeoffs between convergence
    speed and reconstruction accuracy in inverse problems. IEEE Transactions on Signal
    Processing. 66(7), 1676–1690.
  mla: Giryes, Raja, et al. “Tradeoffs between Convergence Speed and Reconstruction
    Accuracy in Inverse Problems.” <i>IEEE Transactions on Signal Processing</i>,
    vol. 66, no. 7, IEEE, 2018, pp. 1676–90, doi:<a href="https://doi.org/10.1109/tsp.2018.2791945">10.1109/tsp.2018.2791945</a>.
  short: R. Giryes, Y.C. Eldar, A.M. Bronstein, G. Sapiro, IEEE Transactions on Signal
    Processing 66 (2018) 1676–1690.
date_created: 2024-10-09T07:45:08Z
date_published: 2018-04-01T00:00:00Z
date_updated: 2024-12-18T11:53:16Z
day: '01'
doi: 10.1109/tsp.2018.2791945
extern: '1'
external_id:
  arxiv:
  - '1605.09232'
intvolume: '        66'
issue: '7'
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://doi.org/10.48550/arXiv.1605.09232
month: '04'
oa: 1
oa_version: Preprint
page: 1676-1690
publication: IEEE Transactions on Signal Processing
publication_identifier:
  eissn:
  - 1941-0476
  issn:
  - 1053-587X
publication_status: published
publisher: IEEE
quality_controlled: '1'
scopus_import: '1'
status: public
title: Tradeoffs between convergence speed and reconstruction accuracy in inverse
  problems
type: journal_article
user_id: 3E5EF7F0-F248-11E8-B48F-1D18A9856A87
volume: 66
year: '2018'
...
---
_id: '18419'
abstract:
- lang: eng
  text: Three important properties of a classification machinery are i) the system
    preserves the core information of the input data; ii) the training examples convey
    information about unseen data; and iii) the system is able to treat differently
    points from different classes. In this paper, we show that these fundamental properties
    are satisfied by the architecture of deep neural networks. We formally prove that
    these networks with random Gaussian weights perform a distance-preserving embedding
    of the data, with a special treatment for in-class and out-of-class data. Similar
    points at the input of the network are likely to have a similar output. The theoretical
    analysis of deep networks here presented exploits tools used in the compressed
    sensing and dictionary learning literature, thereby making a formal connection
    between these important topics. The derived results allow drawing conclusions
    on the metric learning properties of the network and their relation to its structure,
    as well as providing bounds on the required size of the training set such that
    the training examples would represent faithfully the unseen data. The results
    are validated with state-of-the-art trained networks.
article_processing_charge: No
arxiv: 1
author:
- first_name: Raja
  full_name: Giryes, Raja
  last_name: Giryes
- first_name: Guillermo
  full_name: Sapiro, Guillermo
  last_name: Sapiro
- first_name: Alexander
  full_name: Bronstein, Alexander
  id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
  last_name: Bronstein
  orcid: 0000-0001-9699-8730
citation:
  ama: 'Giryes R, Sapiro G, Bronstein AM. Deep neural networks with random Gaussian
    weights: A universal classification strategy? <i>IEEE Transactions on Signal Processing</i>.
    2016;64(13):3444-3457. doi:<a href="https://doi.org/10.1109/tsp.2016.2546221">10.1109/tsp.2016.2546221</a>'
  apa: 'Giryes, R., Sapiro, G., &#38; Bronstein, A. M. (2016). Deep neural networks
    with random Gaussian weights: A universal classification strategy? <i>IEEE Transactions
    on Signal Processing</i>. IEEE. <a href="https://doi.org/10.1109/tsp.2016.2546221">https://doi.org/10.1109/tsp.2016.2546221</a>'
  chicago: 'Giryes, Raja, Guillermo Sapiro, and Alex M. Bronstein. “Deep Neural Networks
    with Random Gaussian Weights: A Universal Classification Strategy?” <i>IEEE Transactions
    on Signal Processing</i>. IEEE, 2016. <a href="https://doi.org/10.1109/tsp.2016.2546221">https://doi.org/10.1109/tsp.2016.2546221</a>.'
  ieee: 'R. Giryes, G. Sapiro, and A. M. Bronstein, “Deep neural networks with random
    Gaussian weights: A universal classification strategy?,” <i>IEEE Transactions
    on Signal Processing</i>, vol. 64, no. 13. IEEE, pp. 3444–3457, 2016.'
  ista: 'Giryes R, Sapiro G, Bronstein AM. 2016. Deep neural networks with random
    Gaussian weights: A universal classification strategy? IEEE Transactions on Signal
    Processing. 64(13), 3444–3457.'
  mla: 'Giryes, Raja, et al. “Deep Neural Networks with Random Gaussian Weights: A
    Universal Classification Strategy?” <i>IEEE Transactions on Signal Processing</i>,
    vol. 64, no. 13, IEEE, 2016, pp. 3444–57, doi:<a href="https://doi.org/10.1109/tsp.2016.2546221">10.1109/tsp.2016.2546221</a>.'
  short: R. Giryes, G. Sapiro, A.M. Bronstein, IEEE Transactions on Signal Processing
    64 (2016) 3444–3457.
date_created: 2024-10-15T11:20:55Z
date_published: 2016-07-01T00:00:00Z
date_updated: 2024-12-18T11:49:18Z
day: '01'
doi: 10.1109/tsp.2016.2546221
extern: '1'
external_id:
  arxiv:
  - '1504.08291'
intvolume: '        64'
issue: '13'
language:
- iso: eng
main_file_link:
- open_access: '1'
  url: https://doi.org/10.48550/arXiv.1504.08291
month: '07'
oa: 1
oa_version: Preprint
page: 3444-3457
publication: IEEE Transactions on Signal Processing
publication_identifier:
  eissn:
  - 1941-0476
  issn:
  - 1053-587X
publication_status: published
publisher: IEEE
quality_controlled: '1'
related_material:
  link:
  - relation: erratum
    url: https://doi.org/10.1109/TSP.2019.2961228
scopus_import: '1'
status: public
title: 'Deep neural networks with random Gaussian weights: A universal classification
  strategy?'
type: journal_article
user_id: 3E5EF7F0-F248-11E8-B48F-1D18A9856A87
volume: 64
year: '2016'
...
---
_id: '18418'
abstract:
- lang: eng
  text: In this correspondence, we consider the problem of multi-input multi output
    (MIMO) quasi maximum likelihood (QML) blind deconvolution. We examine two classes
    of estimators, which are commonly believed to be suitable for super and sub-Gaussian
    sources. We state the consistency conditions and demonstrate a source distribution,
    for which the studied estimators are unsuitable, in the sense that they are inconsistent.
article_processing_charge: No
author:
- first_name: Alexander
  full_name: Bronstein, Alexander
  id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
  last_name: Bronstein
  orcid: 0000-0001-9699-8730
- first_name: M.M.
  full_name: Bronstein, M.M.
  last_name: Bronstein
- first_name: M.
  full_name: Zibulevsky, M.
  last_name: Zibulevsky
citation:
  ama: 'Bronstein AM, Bronstein MM, Zibulevsky M. Quasi maximum likelihood MIMO blind
    deconvolution: Super- and sub-Gaussianity versus consistency. <i>IEEE Transactions
    on Signal Processing</i>. 2005;53(7):2576-2579. doi:<a href="https://doi.org/10.1109/tsp.2005.849221">10.1109/tsp.2005.849221</a>'
  apa: 'Bronstein, A. M., Bronstein, M. M., &#38; Zibulevsky, M. (2005). Quasi maximum
    likelihood MIMO blind deconvolution: Super- and sub-Gaussianity versus consistency.
    <i>IEEE Transactions on Signal Processing</i>. Institute of Electrical and Electronics
    Engineers (IEEE). <a href="https://doi.org/10.1109/tsp.2005.849221">https://doi.org/10.1109/tsp.2005.849221</a>'
  chicago: 'Bronstein, Alex M., M.M. Bronstein, and M. Zibulevsky. “Quasi Maximum
    Likelihood MIMO Blind Deconvolution: Super- and Sub-Gaussianity versus Consistency.”
    <i>IEEE Transactions on Signal Processing</i>. Institute of Electrical and Electronics
    Engineers (IEEE), 2005. <a href="https://doi.org/10.1109/tsp.2005.849221">https://doi.org/10.1109/tsp.2005.849221</a>.'
  ieee: 'A. M. Bronstein, M. M. Bronstein, and M. Zibulevsky, “Quasi maximum likelihood
    MIMO blind deconvolution: Super- and sub-Gaussianity versus consistency,” <i>IEEE
    Transactions on Signal Processing</i>, vol. 53, no. 7. Institute of Electrical
    and Electronics Engineers (IEEE), pp. 2576–2579, 2005.'
  ista: 'Bronstein AM, Bronstein MM, Zibulevsky M. 2005. Quasi maximum likelihood
    MIMO blind deconvolution: Super- and sub-Gaussianity versus consistency. IEEE
    Transactions on Signal Processing. 53(7), 2576–2579.'
  mla: 'Bronstein, Alex M., et al. “Quasi Maximum Likelihood MIMO Blind Deconvolution:
    Super- and Sub-Gaussianity versus Consistency.” <i>IEEE Transactions on Signal
    Processing</i>, vol. 53, no. 7, Institute of Electrical and Electronics Engineers
    (IEEE), 2005, pp. 2576–79, doi:<a href="https://doi.org/10.1109/tsp.2005.849221">10.1109/tsp.2005.849221</a>.'
  short: A.M. Bronstein, M.M. Bronstein, M. Zibulevsky, IEEE Transactions on Signal
    Processing 53 (2005) 2576–2579.
date_created: 2024-10-15T11:20:55Z
date_published: 2005-06-20T00:00:00Z
date_updated: 2024-12-12T12:11:49Z
day: '20'
doi: 10.1109/tsp.2005.849221
extern: '1'
intvolume: '        53'
issue: '7'
language:
- iso: eng
month: '06'
oa_version: None
page: 2576-2579
publication: IEEE Transactions on Signal Processing
publication_identifier:
  eissn:
  - 1941-0476
  issn:
  - 1053-587X
publication_status: published
publisher: Institute of Electrical and Electronics Engineers (IEEE)
quality_controlled: '1'
scopus_import: '1'
status: public
title: 'Quasi maximum likelihood MIMO blind deconvolution: Super- and sub-Gaussianity
  versus consistency'
type: journal_article
user_id: 3E5EF7F0-F248-11E8-B48F-1D18A9856A87
volume: 53
year: '2005'
...