Sorry, you need to enable JavaScript to visit this website.

Multi-channel Signal Processing

DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates


An increasing number of distributed machine learning applications require efficient communication of neural network parameterizations. DeepCABAC, an algorithm in the current working draft of the emerging MPEG-7 part 17 standard for compression of neural networks for multimedia content description and analysis, has demonstrated high compression gains for a variety of neural network models. In this paper we propose a method for employing DeepCABAC in a Federated Learning scenario for the exchange of intermediate differential parameterizations.

Paper Details

Authors:
David Neumann, Felix Sattler, Heiner Kirchhoffer, Simon Wiedemann, Karsten Müller, Heiko Schwarz, Thomas Wiegand, Detlev Marpe, Wojciech Samek
Submitted On:
18 November 2020 - 9:06am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates Presentation Slides.pdf

(9)

Subscribe

[1] David Neumann, Felix Sattler, Heiner Kirchhoffer, Simon Wiedemann, Karsten Müller, Heiko Schwarz, Thomas Wiegand, Detlev Marpe, Wojciech Samek, "DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5556. Accessed: Nov. 26, 2020.
@article{5556-20,
url = {http://sigport.org/5556},
author = {David Neumann; Felix Sattler; Heiner Kirchhoffer; Simon Wiedemann; Karsten Müller; Heiko Schwarz; Thomas Wiegand; Detlev Marpe; Wojciech Samek },
publisher = {IEEE SigPort},
title = {DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates},
year = {2020} }
TY - EJOUR
T1 - DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates
AU - David Neumann; Felix Sattler; Heiner Kirchhoffer; Simon Wiedemann; Karsten Müller; Heiko Schwarz; Thomas Wiegand; Detlev Marpe; Wojciech Samek
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5556
ER -
David Neumann, Felix Sattler, Heiner Kirchhoffer, Simon Wiedemann, Karsten Müller, Heiko Schwarz, Thomas Wiegand, Detlev Marpe, Wojciech Samek. (2020). DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates. IEEE SigPort. http://sigport.org/5556
David Neumann, Felix Sattler, Heiner Kirchhoffer, Simon Wiedemann, Karsten Müller, Heiko Schwarz, Thomas Wiegand, Detlev Marpe, Wojciech Samek, 2020. DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates. Available at: http://sigport.org/5556.
David Neumann, Felix Sattler, Heiner Kirchhoffer, Simon Wiedemann, Karsten Müller, Heiko Schwarz, Thomas Wiegand, Detlev Marpe, Wojciech Samek. (2020). "DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates." Web.
1. David Neumann, Felix Sattler, Heiner Kirchhoffer, Simon Wiedemann, Karsten Müller, Heiko Schwarz, Thomas Wiegand, Detlev Marpe, Wojciech Samek. DeepCABAC - Plug&Play Compression of Neural Network Weights and Weight Updates [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5556

DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS


Multichannel processing is widely used for speech enhancement but several limitations appear when trying to deploy these solutions in the real world. Distributed sensor arrays that consider several devices with a few microphones is a viable solution which allows for exploiting the multiple devices equipped with microphones that we are using in our everyday life. In this context, we propose to extend the distributed adaptive node-specific signal estimation approach to a neural network framework.

Paper Details

Authors:
Irina Illina, Slim Essid
Submitted On:
14 May 2020 - 4:09am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Slides of Nicolas Furnon's ICASSP2020 presentation

(44)

Subscribe

[1] Irina Illina, Slim Essid, "DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5261. Accessed: Nov. 26, 2020.
@article{5261-20,
url = {http://sigport.org/5261},
author = {Irina Illina; Slim Essid },
publisher = {IEEE SigPort},
title = {DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS},
year = {2020} }
TY - EJOUR
T1 - DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS
AU - Irina Illina; Slim Essid
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5261
ER -
Irina Illina, Slim Essid. (2020). DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS. IEEE SigPort. http://sigport.org/5261
Irina Illina, Slim Essid, 2020. DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS. Available at: http://sigport.org/5261.
Irina Illina, Slim Essid. (2020). "DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS." Web.
1. Irina Illina, Slim Essid. DNN-BASED DISTRIBUTED MULTICHANNEL MASK ESTIMATION FOR SPEECH ENHANCEMENT IN MICROPHONE ARRAYS [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5261

LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction


Time series data compression is emerging as an important problem with the growth in IoT devices and sensors. Due to the presence of noise in these datasets, lossy compression can often provide significant compression gains without impacting the performance of downstream applications. In this work, we propose an error-bounded lossy compressor, LFZip, for multivariate floating-point time series data that provides guaranteed reconstruction up to user-specified maximum absolute error.

Paper Details

Authors:
Kedar Tatwawadi, Chengtao Wen, Lingyun Wang, Juan Aparicio, Tsachy Weissman
Submitted On:
18 March 2020 - 5:02pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Session:
Document Year:
Cite

Document Files

slides.pdf

(77)

Subscribe

[1] Kedar Tatwawadi, Chengtao Wen, Lingyun Wang, Juan Aparicio, Tsachy Weissman, "LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/4997. Accessed: Nov. 26, 2020.
@article{4997-20,
url = {http://sigport.org/4997},
author = {Kedar Tatwawadi; Chengtao Wen; Lingyun Wang; Juan Aparicio; Tsachy Weissman },
publisher = {IEEE SigPort},
title = {LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction},
year = {2020} }
TY - EJOUR
T1 - LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction
AU - Kedar Tatwawadi; Chengtao Wen; Lingyun Wang; Juan Aparicio; Tsachy Weissman
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/4997
ER -
Kedar Tatwawadi, Chengtao Wen, Lingyun Wang, Juan Aparicio, Tsachy Weissman. (2020). LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction. IEEE SigPort. http://sigport.org/4997
Kedar Tatwawadi, Chengtao Wen, Lingyun Wang, Juan Aparicio, Tsachy Weissman, 2020. LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction. Available at: http://sigport.org/4997.
Kedar Tatwawadi, Chengtao Wen, Lingyun Wang, Juan Aparicio, Tsachy Weissman. (2020). "LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction." Web.
1. Kedar Tatwawadi, Chengtao Wen, Lingyun Wang, Juan Aparicio, Tsachy Weissman. LFZip: Lossy Compression of Multivariate Floating-Point Time Series Data via Improved Prediction [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/4997

HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS


Time-delay estimation is an essential building block of many signal processing applications. This paper follows up on earlier work for acoustic source localization and time delay estimation using pattern recognition techniques; it presents high performance results obtained with supervised training of neural networks which challenge the state of the art and compares its performance to that of well-known methods such as the Generalized Cross-Correlation or Adaptive Eigenvalue Decomposition.

Paper Details

Authors:
Pooyan Safari, Climent Nadeu
Submitted On:
20 February 2020 - 1:22pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

paperludwighouegnigan_shortversion.pdf

(97)

Subscribe

[1] Pooyan Safari, Climent Nadeu, "HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/4992. Accessed: Nov. 26, 2020.
@article{4992-20,
url = {http://sigport.org/4992},
author = {Pooyan Safari; Climent Nadeu },
publisher = {IEEE SigPort},
title = {HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS},
year = {2020} }
TY - EJOUR
T1 - HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS
AU - Pooyan Safari; Climent Nadeu
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/4992
ER -
Pooyan Safari, Climent Nadeu. (2020). HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS. IEEE SigPort. http://sigport.org/4992
Pooyan Safari, Climent Nadeu, 2020. HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS. Available at: http://sigport.org/4992.
Pooyan Safari, Climent Nadeu. (2020). "HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS." Web.
1. Pooyan Safari, Climent Nadeu. HIGH PERFORMANCE SUPERVISED TIME-DELAY ESTIMATION USING NEURAL NETWORKS [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/4992

ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation


This paper is concerned with estimating unknown multidimensional frequencies from linear compressive measurements. This is accomplished by employing the recently proposed atomic norm minimization framework to recover these frequencies under a sparsity prior without imposing any grid restriction on these frequencies. To this end, we give a rigorous derivation of an iterative scheme called alternating direction of multipliers method, which is able to incorporate multiple compressive snapshots from a multi-dimensional superposition of complex harmonics.

Paper Details

Authors:
Sebastian Semper, Florian Roemer
Submitted On:
20 May 2019 - 8:31am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Slides for the presentation at ICASSP 2019 (SAM-L1.3)

(158)

Subscribe

[1] Sebastian Semper, Florian Roemer, "ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4553. Accessed: Nov. 26, 2020.
@article{4553-19,
url = {http://sigport.org/4553},
author = {Sebastian Semper; Florian Roemer },
publisher = {IEEE SigPort},
title = {ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation},
year = {2019} }
TY - EJOUR
T1 - ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation
AU - Sebastian Semper; Florian Roemer
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4553
ER -
Sebastian Semper, Florian Roemer. (2019). ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation. IEEE SigPort. http://sigport.org/4553
Sebastian Semper, Florian Roemer, 2019. ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation. Available at: http://sigport.org/4553.
Sebastian Semper, Florian Roemer. (2019). "ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation." Web.
1. Sebastian Semper, Florian Roemer. ADMM for ND Line Spectral Estimation using Grid-Free Compressive Sensing from Multiple Measurements with Applications to DOA Estimation [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4553

Learning Overcomplete Dictionaries from Markovian Data

Paper Details

Authors:
Submitted On:
30 June 2018 - 4:35am
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

LEARNING OVERCOMPLETE DICTIONARIES FROM MARKOVIAN DATA

(191)

Subscribe

[1] , "Learning Overcomplete Dictionaries from Markovian Data", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3344. Accessed: Nov. 26, 2020.
@article{3344-18,
url = {http://sigport.org/3344},
author = { },
publisher = {IEEE SigPort},
title = {Learning Overcomplete Dictionaries from Markovian Data},
year = {2018} }
TY - EJOUR
T1 - Learning Overcomplete Dictionaries from Markovian Data
AU -
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3344
ER -
. (2018). Learning Overcomplete Dictionaries from Markovian Data. IEEE SigPort. http://sigport.org/3344
, 2018. Learning Overcomplete Dictionaries from Markovian Data. Available at: http://sigport.org/3344.
. (2018). "Learning Overcomplete Dictionaries from Markovian Data." Web.
1. . Learning Overcomplete Dictionaries from Markovian Data [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3344

Unmixing of Absence Epileptic Seizures in GAERS

Paper Details

Authors:
Submitted On:
30 June 2018 - 4:37am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Unmixing of Absence Epileptic Seizures in GAERS

(180)

Subscribe

[1] , "Unmixing of Absence Epileptic Seizures in GAERS", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/3343. Accessed: Nov. 26, 2020.
@article{3343-18,
url = {http://sigport.org/3343},
author = { },
publisher = {IEEE SigPort},
title = {Unmixing of Absence Epileptic Seizures in GAERS},
year = {2018} }
TY - EJOUR
T1 - Unmixing of Absence Epileptic Seizures in GAERS
AU -
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/3343
ER -
. (2018). Unmixing of Absence Epileptic Seizures in GAERS. IEEE SigPort. http://sigport.org/3343
, 2018. Unmixing of Absence Epileptic Seizures in GAERS. Available at: http://sigport.org/3343.
. (2018). "Unmixing of Absence Epileptic Seizures in GAERS." Web.
1. . Unmixing of Absence Epileptic Seizures in GAERS [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/3343

Revisiting the Kronecker Array Transform


It is known that the calculation of a matrix–vector product can be accelerated if this matrix can be recast (or approximated) by the Kronecker product of two smaller matrices. In array signal processing, the manifold matrix can be described as the Kronecker product of two other matrices if the sensor array displays a separable geometry. This forms the basis of the Kronecker Array Transform (KAT), which was previously introduced to speed up the calculations of acoustic images with microphone arrays.

Paper Details

Authors:
Bruno S, Masiero, Vítor H. Nascimento
Submitted On:
17 April 2018 - 11:55am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP2018.pdf

(334)

Subscribe

[1] Bruno S, Masiero, Vítor H. Nascimento, "Revisiting the Kronecker Array Transform", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2777. Accessed: Nov. 26, 2020.
@article{2777-18,
url = {http://sigport.org/2777},
author = {Bruno S; Masiero; Vítor H. Nascimento },
publisher = {IEEE SigPort},
title = {Revisiting the Kronecker Array Transform},
year = {2018} }
TY - EJOUR
T1 - Revisiting the Kronecker Array Transform
AU - Bruno S; Masiero; Vítor H. Nascimento
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2777
ER -
Bruno S, Masiero, Vítor H. Nascimento. (2018). Revisiting the Kronecker Array Transform. IEEE SigPort. http://sigport.org/2777
Bruno S, Masiero, Vítor H. Nascimento, 2018. Revisiting the Kronecker Array Transform. Available at: http://sigport.org/2777.
Bruno S, Masiero, Vítor H. Nascimento. (2018). "Revisiting the Kronecker Array Transform." Web.
1. Bruno S, Masiero, Vítor H. Nascimento. Revisiting the Kronecker Array Transform [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2777

Channel Dependent Mutual Information in Index Modulations

Paper Details

Authors:
Pol Henarejos, Ana I. Pérez-Neira, Anxo Tato, Carlos Mosquera
Submitted On:
13 April 2018 - 4:05am
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP18_Slides_AP.pdf

(283)

Subscribe

[1] Pol Henarejos, Ana I. Pérez-Neira, Anxo Tato, Carlos Mosquera, "Channel Dependent Mutual Information in Index Modulations", IEEE SigPort, 2018. [Online]. Available: http://sigport.org/2648. Accessed: Nov. 26, 2020.
@article{2648-18,
url = {http://sigport.org/2648},
author = {Pol Henarejos; Ana I. Pérez-Neira; Anxo Tato; Carlos Mosquera },
publisher = {IEEE SigPort},
title = {Channel Dependent Mutual Information in Index Modulations},
year = {2018} }
TY - EJOUR
T1 - Channel Dependent Mutual Information in Index Modulations
AU - Pol Henarejos; Ana I. Pérez-Neira; Anxo Tato; Carlos Mosquera
PY - 2018
PB - IEEE SigPort
UR - http://sigport.org/2648
ER -
Pol Henarejos, Ana I. Pérez-Neira, Anxo Tato, Carlos Mosquera. (2018). Channel Dependent Mutual Information in Index Modulations. IEEE SigPort. http://sigport.org/2648
Pol Henarejos, Ana I. Pérez-Neira, Anxo Tato, Carlos Mosquera, 2018. Channel Dependent Mutual Information in Index Modulations. Available at: http://sigport.org/2648.
Pol Henarejos, Ana I. Pérez-Neira, Anxo Tato, Carlos Mosquera. (2018). "Channel Dependent Mutual Information in Index Modulations." Web.
1. Pol Henarejos, Ana I. Pérez-Neira, Anxo Tato, Carlos Mosquera. Channel Dependent Mutual Information in Index Modulations [Internet]. IEEE SigPort; 2018. Available from : http://sigport.org/2648

Atomic Norm Minimization for Modal Analysis with Random Spatial Compression


Identifying characteristic vibrational modes and frequencies is of great importance for monitoring the health of structures such as buildings and bridges. In this work, we address the problem of estimating the modal parameters of a structure from small amounts of vibrational data collected from wireless sensors distributed on the structure. We consider a randomized spatial compression scheme for minimizing the amount of data that is collected and transmitted by the sensors.

Paper Details

Authors:
Shuang Li, Dehui Yang, Michael Wakin
Submitted On:
12 March 2017 - 4:50pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP2017.pdf

(508)

Subscribe

[1] Shuang Li, Dehui Yang, Michael Wakin, "Atomic Norm Minimization for Modal Analysis with Random Spatial Compression", IEEE SigPort, 2017. [Online]. Available: http://sigport.org/1750. Accessed: Nov. 26, 2020.
@article{1750-17,
url = {http://sigport.org/1750},
author = {Shuang Li; Dehui Yang; Michael Wakin },
publisher = {IEEE SigPort},
title = {Atomic Norm Minimization for Modal Analysis with Random Spatial Compression},
year = {2017} }
TY - EJOUR
T1 - Atomic Norm Minimization for Modal Analysis with Random Spatial Compression
AU - Shuang Li; Dehui Yang; Michael Wakin
PY - 2017
PB - IEEE SigPort
UR - http://sigport.org/1750
ER -
Shuang Li, Dehui Yang, Michael Wakin. (2017). Atomic Norm Minimization for Modal Analysis with Random Spatial Compression. IEEE SigPort. http://sigport.org/1750
Shuang Li, Dehui Yang, Michael Wakin, 2017. Atomic Norm Minimization for Modal Analysis with Random Spatial Compression. Available at: http://sigport.org/1750.
Shuang Li, Dehui Yang, Michael Wakin. (2017). "Atomic Norm Minimization for Modal Analysis with Random Spatial Compression." Web.
1. Shuang Li, Dehui Yang, Michael Wakin. Atomic Norm Minimization for Modal Analysis with Random Spatial Compression [Internet]. IEEE SigPort; 2017. Available from : http://sigport.org/1750

Pages