Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals
In recent years, with the growing popularity of complex signal approximation via deep neural networks, people have begun to pay close attention to the spectral bias of neural networks—a problem that occurs when a neural network is used to fit broadband signals. An important direction taken to overco...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-09-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/22/19/7347 |
_version_ | 1797476928754548736 |
---|---|
author | Zhi Zeng Pengpeng Shi Fulei Ma Peihan Qi |
author_facet | Zhi Zeng Pengpeng Shi Fulei Ma Peihan Qi |
author_sort | Zhi Zeng |
collection | DOAJ |
description | In recent years, with the growing popularity of complex signal approximation via deep neural networks, people have begun to pay close attention to the spectral bias of neural networks—a problem that occurs when a neural network is used to fit broadband signals. An important direction taken to overcome this problem is the use of frequency selection-based fitting techniques, of which the representative work is called the PhaseDNN method, whose core idea is the use of bandpass filters to extract frequency bands with high energy concentration and fit them by different neural networks. Despite the method’s high accuracy, we found in a large number of experiments that the method is less efficient for fitting broadband signals with smooth spectrums. In order to substantially improve its efficiency, a novel candidate—the parallel frequency function-deep neural network (PFF-DNN)—is proposed by utilizing frequency domain analysis of broadband signals and the spectral bias nature of neural networks. A substantial improvement in efficiency was observed in the extensive numerical experiments. Thus, the PFF-DNN method is expected to become an alternative solution for broadband signal fitting. |
first_indexed | 2024-03-09T21:10:44Z |
format | Article |
id | doaj.art-0f9696c322864daaa3b5bca300c1510c |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-09T21:10:44Z |
publishDate | 2022-09-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-0f9696c322864daaa3b5bca300c1510c2023-11-23T21:47:42ZengMDPI AGSensors1424-82202022-09-012219734710.3390/s22197347Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband SignalsZhi Zeng0Pengpeng Shi1Fulei Ma2Peihan Qi3School of Mechano-Electronics Engineering, Xidian University, Xi’an 710071, ChinaSchool of Civil Engineering & Institute of Mechanics and Technology, Xi’an University of Architecture and Technology, Xi’an 710055, ChinaSchool of Mechano-Electronics Engineering, Xidian University, Xi’an 710071, ChinaSchool of Telecommunications Engineering, Xidian University, Xi’an 710071, ChinaIn recent years, with the growing popularity of complex signal approximation via deep neural networks, people have begun to pay close attention to the spectral bias of neural networks—a problem that occurs when a neural network is used to fit broadband signals. An important direction taken to overcome this problem is the use of frequency selection-based fitting techniques, of which the representative work is called the PhaseDNN method, whose core idea is the use of bandpass filters to extract frequency bands with high energy concentration and fit them by different neural networks. Despite the method’s high accuracy, we found in a large number of experiments that the method is less efficient for fitting broadband signals with smooth spectrums. In order to substantially improve its efficiency, a novel candidate—the parallel frequency function-deep neural network (PFF-DNN)—is proposed by utilizing frequency domain analysis of broadband signals and the spectral bias nature of neural networks. A substantial improvement in efficiency was observed in the extensive numerical experiments. Thus, the PFF-DNN method is expected to become an alternative solution for broadband signal fitting.https://www.mdpi.com/1424-8220/22/19/7347PFF-DNNspectral biasbroadband signalsfast Fourier analysis |
spellingShingle | Zhi Zeng Pengpeng Shi Fulei Ma Peihan Qi Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals Sensors PFF-DNN spectral bias broadband signals fast Fourier analysis |
title | Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals |
title_full | Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals |
title_fullStr | Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals |
title_full_unstemmed | Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals |
title_short | Parallel Frequency Function-Deep Neural Network for Efficient Approximation of Complex Broadband Signals |
title_sort | parallel frequency function deep neural network for efficient approximation of complex broadband signals |
topic | PFF-DNN spectral bias broadband signals fast Fourier analysis |
url | https://www.mdpi.com/1424-8220/22/19/7347 |
work_keys_str_mv | AT zhizeng parallelfrequencyfunctiondeepneuralnetworkforefficientapproximationofcomplexbroadbandsignals AT pengpengshi parallelfrequencyfunctiondeepneuralnetworkforefficientapproximationofcomplexbroadbandsignals AT fuleima parallelfrequencyfunctiondeepneuralnetworkforefficientapproximationofcomplexbroadbandsignals AT peihanqi parallelfrequencyfunctiondeepneuralnetworkforefficientapproximationofcomplexbroadbandsignals |