Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency Reconstruction
Time–frequency distributions (TFDs) are crucial for analyzing non-stationary signals. Compressive sensing (CS) in the ambiguity domain offers an approach for TFD reconstruction with high performance, but selecting the optimal regularization parameter for various signals remains challenging. Traditio...
Saved in:
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Technologies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7080/12/12/251 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846102545358913536 |
|---|---|
| author | Vedran Jurdana |
| author_facet | Vedran Jurdana |
| author_sort | Vedran Jurdana |
| collection | DOAJ |
| description | Time–frequency distributions (TFDs) are crucial for analyzing non-stationary signals. Compressive sensing (CS) in the ambiguity domain offers an approach for TFD reconstruction with high performance, but selecting the optimal regularization parameter for various signals remains challenging. Traditional methods for parameter selection, including manual and experimental approaches, as well as existing optimization procedures, can be imprecise and time-consuming. This study introduces a novel approach using deep neural networks (DNNs) to predict regularization parameters based on Wigner–Ville distributions (WVDs). The proposed DNN is trained on a comprehensive dataset of synthetic signals featuring multiple linear and quadratic frequency-modulated components, with variations in component amplitudes and random positions, ensuring wide applicability and robustness. By utilizing DNNs, end-users need only provide the signal’s WVD, eliminating the need for manual parameter selection and lengthy optimization procedures. Comparisons between the reconstructed TFDs using the proposed DNN-based approach and existing optimization methods highlight significant improvements in both reconstruction performance and execution time. The effectiveness of this methodology is validated on noisy synthetic and real-world signals, emphasizing the potential of DNNs to automate regularization parameter determination for CS-based TFD reconstruction in diverse signal environments. |
| format | Article |
| id | doaj-art-16f42e7f40c44f07a16718618c61d8d3 |
| institution | Kabale University |
| issn | 2227-7080 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Technologies |
| spelling | doaj-art-16f42e7f40c44f07a16718618c61d8d32024-12-27T14:56:01ZengMDPI AGTechnologies2227-70802024-12-01121225110.3390/technologies12120251Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency ReconstructionVedran Jurdana0Department of Automation and Electronics, Faculty of Engineering, University of Rijeka, 51000 Rijeka, CroatiaTime–frequency distributions (TFDs) are crucial for analyzing non-stationary signals. Compressive sensing (CS) in the ambiguity domain offers an approach for TFD reconstruction with high performance, but selecting the optimal regularization parameter for various signals remains challenging. Traditional methods for parameter selection, including manual and experimental approaches, as well as existing optimization procedures, can be imprecise and time-consuming. This study introduces a novel approach using deep neural networks (DNNs) to predict regularization parameters based on Wigner–Ville distributions (WVDs). The proposed DNN is trained on a comprehensive dataset of synthetic signals featuring multiple linear and quadratic frequency-modulated components, with variations in component amplitudes and random positions, ensuring wide applicability and robustness. By utilizing DNNs, end-users need only provide the signal’s WVD, eliminating the need for manual parameter selection and lengthy optimization procedures. Comparisons between the reconstructed TFDs using the proposed DNN-based approach and existing optimization methods highlight significant improvements in both reconstruction performance and execution time. The effectiveness of this methodology is validated on noisy synthetic and real-world signals, emphasizing the potential of DNNs to automate regularization parameter determination for CS-based TFD reconstruction in diverse signal environments.https://www.mdpi.com/2227-7080/12/12/251time–frequency distributionsparse reconstructionregularization parameterdeep learningneural networkcompressive sensing |
| spellingShingle | Vedran Jurdana Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency Reconstruction Technologies time–frequency distribution sparse reconstruction regularization parameter deep learning neural network compressive sensing |
| title | Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency Reconstruction |
| title_full | Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency Reconstruction |
| title_fullStr | Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency Reconstruction |
| title_full_unstemmed | Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency Reconstruction |
| title_short | Deep Neural Networks for Estimating Regularization Parameter in Sparse Time–Frequency Reconstruction |
| title_sort | deep neural networks for estimating regularization parameter in sparse time frequency reconstruction |
| topic | time–frequency distribution sparse reconstruction regularization parameter deep learning neural network compressive sensing |
| url | https://www.mdpi.com/2227-7080/12/12/251 |
| work_keys_str_mv | AT vedranjurdana deepneuralnetworksforestimatingregularizationparameterinsparsetimefrequencyreconstruction |