Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AI
The artificial intelligence (AI) application in instruments such as impedance spectroscopy highlights the difficulty to choose an electronic technology that correctly solves the basic performance problems, adaptation to the context, flexibility, precision, autonomy, and speed of design. Present work...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-06-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/11/13/2064 |
| _version_ | 1797480397202784256 |
|---|---|
| author | Jorge Fe Rafael Gadea-Gironés Jose M. Monzo Ángel Tebar-Ruiz Ricardo Colom-Palero |
| author_facet | Jorge Fe Rafael Gadea-Gironés Jose M. Monzo Ángel Tebar-Ruiz Ricardo Colom-Palero |
| author_sort | Jorge Fe |
| collection | DOAJ |
| description | The artificial intelligence (AI) application in instruments such as impedance spectroscopy highlights the difficulty to choose an electronic technology that correctly solves the basic performance problems, adaptation to the context, flexibility, precision, autonomy, and speed of design. Present work demonstrates that FPGAs, in conjunction with an optimized high-level synthesis (HLS), allow us to have an efficient connection between the signals sensed by the instrument and the artificial neural network-based AI computing block that will analyze them. State-of-the-art comparisons and experimental results also demonstrate that our designed and developed architectures offer the best compromise between performance, efficiency, and system costs in terms of artificial neural networks implementation. In the present work, computational efficiency above 21 Mps/DSP and power efficiency below 1.24 mW/Mps are achieved. It is important to remark that these results are more relevant because the system can be implemented on a low-cost FPGA. |
| first_indexed | 2024-03-09T21:59:24Z |
| format | Article |
| id | doaj.art-3119c1d930f84d428b9d2762f8596856 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-09T21:59:24Z |
| publishDate | 2022-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-3119c1d930f84d428b9d2762f85968562023-11-23T19:52:15ZengMDPI AGElectronics2079-92922022-06-011113206410.3390/electronics11132064Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AIJorge Fe0Rafael Gadea-Gironés1Jose M. Monzo2Ángel Tebar-Ruiz3Ricardo Colom-Palero4Institute for Molecular Imaging Technologies (I3M), Universitat Politècnica de València, 46022 Valencia, SpainInstitute for Molecular Imaging Technologies (I3M), Universitat Politècnica de València, 46022 Valencia, SpainInstitute for Molecular Imaging Technologies (I3M), Universitat Politècnica de València, 46022 Valencia, SpainInstitute for Molecular Imaging Technologies (I3M), Universitat Politècnica de València, 46022 Valencia, SpainInstitute for Molecular Imaging Technologies (I3M), Universitat Politècnica de València, 46022 Valencia, SpainThe artificial intelligence (AI) application in instruments such as impedance spectroscopy highlights the difficulty to choose an electronic technology that correctly solves the basic performance problems, adaptation to the context, flexibility, precision, autonomy, and speed of design. Present work demonstrates that FPGAs, in conjunction with an optimized high-level synthesis (HLS), allow us to have an efficient connection between the signals sensed by the instrument and the artificial neural network-based AI computing block that will analyze them. State-of-the-art comparisons and experimental results also demonstrate that our designed and developed architectures offer the best compromise between performance, efficiency, and system costs in terms of artificial neural networks implementation. In the present work, computational efficiency above 21 Mps/DSP and power efficiency below 1.24 mW/Mps are achieved. It is important to remark that these results are more relevant because the system can be implemented on a low-cost FPGA.https://www.mdpi.com/2079-9292/11/13/2064FPGAimpedance spectroscopyartificial neural networkshigh-level synthesisAI edge computing |
| spellingShingle | Jorge Fe Rafael Gadea-Gironés Jose M. Monzo Ángel Tebar-Ruiz Ricardo Colom-Palero Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AI Electronics FPGA impedance spectroscopy artificial neural networks high-level synthesis AI edge computing |
| title | Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AI |
| title_full | Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AI |
| title_fullStr | Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AI |
| title_full_unstemmed | Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AI |
| title_short | Improving FPGA Based Impedance Spectroscopy Measurement Equipment by Means of HLS Described Neural Networks to Apply Edge AI |
| title_sort | improving fpga based impedance spectroscopy measurement equipment by means of hls described neural networks to apply edge ai |
| topic | FPGA impedance spectroscopy artificial neural networks high-level synthesis AI edge computing |
| url | https://www.mdpi.com/2079-9292/11/13/2064 |
| work_keys_str_mv | AT jorgefe improvingfpgabasedimpedancespectroscopymeasurementequipmentbymeansofhlsdescribedneuralnetworkstoapplyedgeai AT rafaelgadeagirones improvingfpgabasedimpedancespectroscopymeasurementequipmentbymeansofhlsdescribedneuralnetworkstoapplyedgeai AT josemmonzo improvingfpgabasedimpedancespectroscopymeasurementequipmentbymeansofhlsdescribedneuralnetworkstoapplyedgeai AT angeltebarruiz improvingfpgabasedimpedancespectroscopymeasurementequipmentbymeansofhlsdescribedneuralnetworkstoapplyedgeai AT ricardocolompalero improvingfpgabasedimpedancespectroscopymeasurementequipmentbymeansofhlsdescribedneuralnetworkstoapplyedgeai |