A Low-Power DWS-CNN-Based Downsampling Approach for Odor Identification Using Herbal and Organic Datasets
Received: 14 February 2026 | Revised: 2 April 2026 and 23 April 2026 | Accepted: 30 April 2026 | Online: 12 May 2026
Corresponding author: Akshata K. Aldi
Abstract
Odor identification systems play an important role in food safety applications, herbal authentication, and portable sensing devices. These systems help identify the essential properties of natural organic substances with fast processing speed, low power consumption, and high accuracy. Traditional odor analysis methods provide precise detection and structured gas sensor measurements, yet they involve high computational cost, large architectures, slow data inference, and limited suitability for FPGA- and embedded system-based real-time applications. Furthermore, traditional Convolutional Neural Network (CNN) architectures generally require multiple convolutional layers, large memory resources, high hardware utilization, and increased classification latency, making them unsuitable for low-power applications such as odor identification and character recognition. Therefore, the proposed Depthwise Separable Convolutional Neural Network (DWS-CNN) was specifically developed for low-power applications using a 26×26 input window size. The proposed architecture acquires 676 samples every 65.36 μs at a frequency of 100 MHz. In addition, it classifies odor identification data from various herbal and organic datasets using multiclass dense classification with a 3×3 window size. The proposed model simultaneously predicts and classifies 12 different sensing values from 50 distinct categories. The entire architecture was implemented in Verilog HDL and functionally verified using Modelsim. The design was synthesized on an Artix-7 FPGA, achieving a low power consumption of 0.362 W and a latency of 4.146 ns, demonstrating its suitability for real-time odor identification applications.
Keywords:
odor identification, multiclass dense classification, Convolutional Neural Network (CNN), herbal and organic datasetsDownloads
References
P. Singh, U. Habiba, Z. Shafi, A. Noor, V. K. Pandey, and R. Singh, "Understanding the Concepts of Smart E-Nose Technology in Combination With Machine Learning for New Era of Food Safety: An Advanced Review," Food Safety and Health, vol. 3, no. 4, pp. 518–534, Oct. 2025.
F. R. Cavalcante et al., "Determinants of Odor-Related Perception: Analysis of Community Response," Atmosphere, vol. 16, no. 10, Oct. 2025, Art. no. 1176.
S. Fuentes et al., "Assessment of Smoke Contamination in Grapevine Berries and Taint in Wines Due to Bushfires Using a Low-Cost E-Nose and an Artificial Intelligence Approach," Sensors, vol. 20, no. 18, Sept. 2020, Art. no. 5108.
R. Macìas-Quijas, R. Velázquez, C. Del-Valle-Soto, R. Lizut, P. Visconti, and A. Lay-Ekuakille, "Environmental odor detection and classification with electronic nose system," Bulletin of Electrical Engineering and Informatics, vol. 14, no. 2, pp. 1117–1125, Apr. 2025.
H. Patel et al., "Design Principles From Natural Olfaction for Electronic Noses," Advanced Science, vol. 12, no. 12, Mar. 2025, Art. no. 2412669.
E. L. Hines, E. Llobet, and J. W. Gardner, "Electronic noses: a review of signal processing techniques," IEE Proceedings - Circuits, Devices and Systems, vol. 146, no. 6, pp. 297–310, Dec. 1999.
H. Wei and Y. Gu, "A Machine Learning Method for the Detection of Brown Core in the Chinese Pear Variety Huangguan Using a MOS-Based E-Nose," Sensors, vol. 20, no. 16, Aug. 2020, Art. no. 4499.
T. Sanislav, G. D. Mois, S. Zeadally, S. Folea, T. C. Radoni, and E. A. Al-Suhaimi, "A Comprehensive Review on Sensor-Based Electronic Nose for Food Quality and Safety," Sensors, vol. 25, no. 14, July 2025, Art. no. 4437.
J. Segura-Garcia et al., "AI-driven 5G IoT e-nose for whiskey classification," Applied Intelligence, vol. 55, no. 10, Apr. 2025, Art. no. 686.
C. Cheng and K. K. Parhi, "Fast 2D Convolution Algorithms for Convolutional Neural Networks," IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 67, no. 5, pp. 1678–1691, May 2020.
J. Yepez and S.-B. Ko, "Stride 2 1-D, 2-D, and 3-D Winograd for Convolutional Neural Networks," IEEE Transactions on Very Large Scale Integration (VLSI) Systems, vol. 28, no. 4, pp. 853–863, Apr. 2020.
R. Handayani, R. Sarno, D. Rahman Wijaya, K. R. Sungkono, and D. Y. Kristiyanto, "An Approach Sensory Analysis Using Targeted Sensors in Electronic Nose for Assessing Green Tea Quality," in 2024 International Seminar on Intelligent Technology and Its Applications, Mataram, Indonesia, 2024, pp. 746–751.
O. Sosa-Ramos et al., "Study of the effect of relative humidity on the classification of volatile organic compounds for a quartz crystal microbalance sensors array," Sensors and Actuators A: Physical, vol. 387, June 2025, Art. no. 116465.
D. Feng, W. Dai, C. Li, A. Pernigo, and P. P. Liang, "SmellNet: A Large-scale Dataset for Real-world Smell Recognition," presented at the The Fourteenth International Conference on Learning Representations, Rio de Janeiro, Brazil, 2025.
R. L. Doty, "Odors as cognitive constructs: history of odor classification and attempts to map odor percepts to physical and chemical parameters," Chemical Senses, vol. 50, Jan. 2025, Art. no. bjaf022.
M. H.-M. Khan et al., "Multi- class classification of breast cancer abnormalities using Deep Convolutional Neural Network (CNN)," PLOS ONE, vol. 16, no. 8, Aug. 2021, Art. no. e0256500.
Z. Mo, D. Luo, T. Wen, Y. Cheng, and X. Li, "FPGA Implementation for Odor Identification with Depthwise Separable Convolutional Neural Network," Sensors, vol. 21, no. 3, Jan. 2021, Art. no. 832.
D. L. T. Wong, Y. Li, D. John, W. K. Ho, and C. H. Heng, "Resource and Energy Efficient Implementation of ECG Classifier Using Binarized CNN for Edge AI Devices," in 2021 IEEE International Symposium on Circuits and Systems (ISCAS), Daegu, Korea, 2021, pp. 1–5.
L. Bai, Y. Zhao, and X. Huang, "A CNN Accelerator on FPGA Using Depthwise Separable Convolution," IEEE Transactions on Circuits and Systems II: Express Briefs, vol. 65, no. 10, pp. 1415–1419, Oct. 2018.
I. Yuadi, N. U. Nazikhah, C.-C. Hu, K. Nisa’, and V. S. Suryakala, "Deep Learning-Based Galls and Healthy Leaf Recognition Using Depthwise Separable CNN Architecture," Engineering, Technology & Applied Science Research, vol. 15, no. 3, pp. 23777–23782, June 2025.
D. R. Wijaya, R. Handayani, M. D. Badri, S. Shabri, and V. P. Rahadi, "Data set for Gambung green tea aroma using on electronic nose," BMC Research Notes, vol. 17, no. 1, Sept. 2024, Art. no. 244.
D. Ameta, S. Kumar, R. Mishra, L. Behera, A. Chakraborty, and T. Sandhan, "Odor classification: Exploring feature performance and imbalanced data learning techniques," Plos One, vol. 20, no. 5, May 2025, Art. no. e0322514.
X. Chen, J. Ji, S. Mei, Y. Zhang, M. Han, and Q. Du, "FPGA Based Implementation of Convolutional Neural Network for Hyperspectral Classification," in IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, 2018, pp. 2451–2454.
X. Zhao et al., "HDSuper: High-Quality and High Computational Utilization Edge Super-Resolution Accelerator With Hardware-Algorithm Co-Design Techniques," IEEE Transactions on Circuits and Systems I: Regular Papers, vol. 71, no. 4, pp. 1679–1692, Apr. 2024.
W. Li, H. Leung, C. Kwan, and B. R. Linnell, "E-Nose Vapor Identification Based on Dempster–Shafer Fusion of Multiple Classifiers," IEEE Transactions on Instrumentation and Measurement, vol. 57, no. 10, pp. 2273–2282, Oct. 2008.
D. Wu, D. Luo, K.-Y. Wong, and K. Hung, "POP-CNN: Predicting Odor Pleasantness With Convolutional Neural Network," IEEE Sensors Journal, vol. 19, no. 23, pp. 11337–11345, Dec. 2019.
H. Nishijima et al., "Optimizing Odorants for Olfactory Training Based on Olfactory Receptor–Ligand Pair Analysis," The Laryngoscope, vol. 136, no. 1, pp. 384–394, Jan. 2026.
I. Shtepliuk, K. Montelius, J. Eriksson, and D. Puglisi, "Adaptive Machine Learning for Electronic Nose-Based Forensic VOC Classification," Advanced Science, vol. 12, no. 36, Sept. 2025, Art. no. e04657.
A. M. Dhanush Gowda, A. D. Dessai, and U. Y. Nayak, "Electronic-Nose Technology for Lung Cancer Detection: A Non-Invasive Diagnostic Revolution," Lung, vol. 203, no. 1, July 2025, Art. no. 76.
Y. Zhou, J. Wang, and S. Zhou, "A Dense Network Model of Traffic Classification Based on Autoencoder and Attention Mechanism," in 2024 IEEE 12th International Conference on Computer Science and Network Technology, Dalian, China, 2024, pp. 55–59.
H. Zhang, H. Yu, Z. Xu, K. Zheng, and L. Gao, "A Novel Classification Framework for Hyperspectral Image Classification Based on Multi-Scale Dense Network," in 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS, Brussels, Belgium, 2021, pp. 2238–2241.
Y. Zhou, S. Zhou, and J. Cui, "An Encrypted Traffic Classification Model Based on Improved Dense Network and Bidirectional Gated Recurrent Unit," in 2024 IEEE 12th International Conference on Computer Science and Network Technology, Dalian, China, 2024, pp. 60–66.
D. Feng, "MIT-MI/SmellNet." MIT Multisensory Intelligence Group, Apr. 22, 2026, [Online]. Available: https://github.com/MIT-MI/SmellNet.
Downloads
How to Cite
License
Copyright (c) 2026 Akshata K. Aldi, R. L. Raibagkar
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
