Monitoring Radio Frekuensi Berbasis Deep Learning Untuk Klasifikasi Sinyal, Clustering dan Rekomendasi Amplifier

Cindy Alnis Rochmadani; Ade Silvia Handayani; Emilia Hesti

doi:10.70340/jirsi.v5i3.542

Authors

Cindy Alnis Rochmadani Politeknik Negeri Sriwijaya
Ade Silvia Handayani Politeknik Negeri Sriwijaya
Emilia Hesti Politeknik Negeri Sriwijaya

DOI:

https://doi.org/10.70340/jirsi.v5i3.542

Keywords:

Deep learning, GoogLeNet, DBSCAN, SDR Sharp, radio spectrum monitoring

Abstract

Monitoring of radio frequency spectrum is an important activity in managing frequency usage to prevent interference between users. This study proposes a deep learning-based radio frequency monitoring system using SDR# spectrum images with RTL-SDR receiver implemented in MATLAB. The system integrates GoogLeNet with transfer learning to extract visual features from spectrum images, Autoencoder to reduce feature dimensions to a 32-dimensional latent space, and K-Means algorithm to classify spectrum conditions into Cluster A (good signal), Cluster B (moderate signal), and Cluster C (weak signal). AM/FM signal modulation identification is performed using the OCR method on the SDR# image header with FM frequency range 87-108 MHz and AM 109-137 MHz. The system was tested at 20 measurement points in 4 cardinal directions with distance variations of 2 km to 10 km using 909 SDR# spectrum images. Test results show an OCR Success Rate of 16.4% as a measure of frequency text-reading success, indicating that SDR# is not optimal as an image source in this approach. Among the images successfully recognized by OCR, signal modulation classification based on frequency range achieved 100% accuracy, precision, and recall as a measure of classification correctness. The system also produced spectrum condition clustering using K-Means (k=3) with a distribution of Cluster A (56.8%), Cluster B (30.4%), and Cluster C (20.7%), and RF Amplifier recommendation based on FM signal dominance. Future research is recommended to use GQRX to improve system performance.

Downloads

Download data is not yet available.

References

Z. Sabrina, T. Djamel, T. Camel, M. Ammar, S. Said, and N. Okba, “SDR Implementation of Spectrum Sensing using Deep Learning,” Int. J. Sensors, Wirel. Commun. Control, vol. 13, no. 4, pp. 226–236, 2023, doi: 10.2174/2210327913666230719152400.

P. Flak, “Hardware-Accelerated Real-Time Spectrum Analyzer With a Broadband Fast Sweep Feature Based on the Cost-Effective SDR Platform,” IEEE Access, vol. 10, no. September, pp. 110934–110946, 2022, doi: 10.1109/ACCESS.2022.3215800.

S. E. Abdelbaset, H. M. Kasem, A. A. Khalaf, A. H. Hussein, and A. A. Kabeel, “Deep Learning-Based Spectrum Sensing for Cognitive Radio Applications,” Sensors, vol. 24, no. 24, 2024, doi: 10.3390/s24247907.

A. Olesiński and Z. Piotrowski, “Clustering Method for Signals in the Wideband RF Spectrum Using Semi-Supervised Deep Contrastive Learning,” Appl. Sci., vol. 14, no. 7, 2024, doi: 10.3390/app14072990.

T. Liang, G. Yilmaz, and C. S. Soon, “Deriving Accurate Nocturnal Heart Rate, rMSSD and Frequency HRV from the Oura Ring,” Sensors, vol. 24, no. 23, 2024, doi: 10.3390/s24237475.

J. H. Kim, S. Y. Kwon, and H. N. Kim, “Enhanced Radar Signal Classification Using AMP and Visibility Graph for Multi-Signal Environments,” Sensors, vol. 24, no. 23, 2024, doi: 10.3390/s24237612.

D. P. Kingma and M. Welling, “Auto-encoding variational bayes,” 2nd Int. Conf. Learn. Represent. ICLR 2014 - Conf. Track Proc., no. Ml, pp. 1–14, 2014, doi: 10.61603/ceas.v2i1.33.

M. Daszykowski and B. Walczak, “2.26 - Density-Based Clustering Methods,” Compr. Chemom. Chem. Biochem. Data Anal. Second Ed. Four Vol. Set, vol. 2, pp. 565–580, 2020, doi: 10.1016/B978-0-444-64165-6.03005-6.

H. Kim, Y. J. Kim, and W. T. Kim, “Multitask Learning-Based Deep Signal Identification for Advanced Spectrum Sensing,” Sensors, vol. 23, no. 24, pp. 1–22, 2023, doi: 10.3390/s23249806.

Z. Jing, P. Li, B. Wu, S. Yuan, and Y. Chen, “An Adaptive Focal Loss Function Based on Transfer Learning for Few-Shot Radar Signal Intra-Pulse Modulation Classification,” Remote Sens., vol. 14, no. 8, 2022, doi: 10.3390/rs14081950.

D. G. Cortes, E. Onieva, I. P. Lopez, L. Trinchera, and J. Wu, “Autoencoder-Enhanced Clustering: A Dimensionality Reduction Approach to Financial Time Series,” IEEE Access, vol. 12, no. January, pp. 16999–17009, 2024, doi: 10.1109/ACCESS.2024.3359413.

L. Yin, H. Hu, K. Li, G. Zheng, Y. Qu, and H. Chen, “Improvement of DBSCAN Algorithm Based on K-Dist Graph for Adaptive Determining Parameters,” Electron., vol. 12, no. 15, 2023, doi: 10.3390/electronics12153213.

T. Zaki Abdulhameed, S. A. Yousif, V. W. Samawi, and H. Imad Al-Shaikhli, “SS-DBSCAN: Semi-Supervised Density-Based Spatial Clustering of Applications With Noise for Meaningful Clustering in Diverse Density Data,” IEEE Access, vol. 12, no. September, pp. 131507–131520, 2024, doi: 10.1109/ACCESS.2024.3457587.

A. Nagasubramanian et al., “OCRNet a robust deep learning framework for alphanumeric character recognition to assist the visually impaired,” Sci. Rep., vol. 15, no. 1, pp. 1–16, 2025, doi: 10.1038/s41598-025-25278-9.

M. Zaid et al., “Optimizing Low Noise Amplifiers: A Two-Stage Approach for Improved Noise Figure and Stability,” IEEE Access, vol. 12, no. March, pp. 53475–53484, 2024, doi: 10.1109/ACCESS.2024.3387108.