Derin Öğrenme Destekli Çok Kanallı Karar Destek Sistemi İle Askeri Deniz Platformlarında Haberleşme Optimizasyonu

Abstract

Bu çalışma, askeri deniz platformlarında kullanılan çok kanallı haberleşme sistemlerinin etkinliğini artırmak amacıyla, derin öğrenme tabanlı bir karar destek sistemi geliştirmektedir. Ka Band, UHF/VHF, GSM ve gemi içi kablosuz iletişim modülleri (KİM) gibi farklı karakteristiklere sahip iletişim kanalları, sinyal gücü, paket kaybı, gecikme, elektromanyetik parazit ve görüş hattı gibi çok boyutlu parametreler üzerinden analiz edilerek değerlendirilmektedir. Geliştirilen yapay zeka modeli, mevcut kanal koşullarına göre en uygun iletişim kanalını otomatik olarak seçmekte ve böylece iletişim sürekliliği ile güvenliği artırmaktadır. Modelin deneysel değerlendirmelerinde %91.4 doğruluk ve 0.906 F1-skoru elde edilmiştir. Özellikle UHF ve Ka Band gibi stabil kanallarda %93’ün üzerinde başarı sağlanırken, GSM ve KİM gibi değişken ortam koşullarına sahip kanallarda performans görece düşük kalmıştır. Bu sonuçlar, sistemin heterojen veri setleriyle daha da güçlendirilebileceğini göstermektedir. Geliştirilen sistem, geleneksel kanal seçim süreçlerine kıyasla operasyonel çeviklik, gerçek zamanlı karar alma ve öngörü yeteneği açısından önemli üstünlükler sunmaktadır. Bu yönüyle çalışma, yalnızca iletişim ağlarının mevcut durumunu değerlendirmekle kalmayıp, geçmiş verilerden öğrenerek gelecekteki kanal davranışlarını tahmin edebilmekte ve olası iletişim kesintilerine karşı önleyici aksiyonlar alabilmektedir. Literatürde yer alan benzer çalışmalara göre, çoklu kanal yapısının tek bir karar mimarisi altında ele alınması ve sistemin gerçek zamanlı çalışabilme kapasitesi, çalışmayı özgün ve uygulamaya dönük kılmaktadır. Elde edilen bulgular, önerilen yaklaşımın modern askeri haberleşme sistemlerinde yüksek güvenlik, esneklik ve adaptasyon kabiliyeti sunduğunu ortaya koymaktadır.

Keywords

• Derin öğrenme
• UHF/VHF
• Ka Band
• GSM

Communication Optimization in Military Naval Platforms Using a Deep Learning-Based Multi-Channel Decision Support System

Abstract

In this study, a deep learning-based decision support system is proposed to dynamically manage multi-channel communication infrastructures used in military naval platforms. The developed model analyzes various communication channels with different characteristics such as Ka Band, UHF/VHF, wireless modules, and GSM based on environmental and operational parameters, ensuring the optimal selection of communication channels in real time. The system processes multi-dimensional inputs such as signal strength, latency, interference level, packet loss, and line-of-sight conditions to enhance both communication continuity and security. Experimental results demonstrate that the model achieves a classification accuracy of 91.4% and an F1-score of 0.906, with higher success rates observed on more stable channels like UHF and Ka Band. The proposed deep neural network not only adapts to current network conditions but also learns from historical data to anticipate future channel behavior, providing a proactive solution against potential communication disruptions. Compared to traditional systems, the proposed architecture offers enhanced operational agility and reduced dependency on manual intervention. Its ability to handle heterogeneous data sources and dynamically manage channel transitions positions it as a robust and scalable solution for modern military communication needs. Future work will focus on extending the system to other military platforms, integrating threat detection modules, and incorporating quantum-secure protocols for enhanced end to end communication security.

Keywords

• Deep learning
• UHF/VHF
• Ka Band
• GSM

References

1. [1] B. Zhao et al., "Multi-agent deep reinforcement learning for RIS-assisted secure UAV communication," Journal of King Saud University - Computer and Information Sciences, vol. 37, no. 133, pp. 1-17, 2025, doi: 10.1007/s44443-025-00145-9.
2. [2] W. Lei, Y. Ye and M. Xiao, "Deep Reinforcement Learning-Based Spectrum Allocation in Integrated Access and Backhaul Networks," IEEE Trans. Cogn. Commun. Netw., vol. 6, no. 3, pp. 970–979, Sep.2020, doi: 10.1109/TCCN.2020.2992628.
3. [3] H. Farhi, A. Messai and T. Berghout, "Enhancing Spectrum Utilization in Cognitive Radio Networks Using Reinforcement Learning with Snake Optimizer: A Meta-Heuristic Approach,"Electronics, vol. 14, no. 13, p. 2525, 2025, doi: 10.3390/electronics14132525.
4. [4] M. Kulin, T. Kazaz, E. De Poorter and I. Moerman, "A Survey on Machine Learning-Based Performance Improvement of Wireless Networks: PHY, MAC and Network Layer," Electronics, vol. 10, no. 3, p. 318, 2021, doi: 10.3390/electronics10030318.
5. [5] H. Xie, Z. Qin, G. Y. Li and B. -H. Juang, "Deep Learning Enabled Semantic Communication Systems," in IEEE Transactions on Signal Processing, vol. 69, pp. 2663-2675, 2021, doi: 10.1109/TSP.2021.307121
6. [6] T. O’Shea and J. Hoydis, "An Introduction to Deep Learning for the Physical Layer," in IEEE Transactions on Cognitive Communications and Networking, vol. 3, no. 4, pp. 563-575, Dec. 2017, doi: 10.1109/TCCN.2017.2758370.
7. [7] R. Safavinejad, H.-H. Chang and L. Liu, "Deep Reinforcement Learning for Dynamic Spectrum Access: Convergence Analysis and System Design," in IEEE Transactions on Wireless Communications, vol. 23, no. 12, pp. 18888-18902, Dec. 2024, doi: 10.1109/TWC.2024.3414428.
8. [8] T. -D. Le and G. Kaddoum, "LSTM-Based Channel Access Scheme for Vehicles in Cognitive Vehicular Networks With Multi-Agent Settings," in IEEE Transactions on Vehicular Technology, vol. 70, no. 9, pp. 9132-9143, Sept. 2021, doi: 10.1109/TVT.2021.3100591.

9. [9] A. Arsalan, T. Umer, R. A. Rehman, B.-S. Kim, M. Murtaza, and C.-T. Cheng, "Semantic-aware communication and computation co-optimization in digital twin-assisted Internet of Drones networks using federated deep reinforcement learning," ICT Express, vol. 12, no. 2, pp. 517-522, Apr. 2026, doi: 10.1016/j.icte.2026.02.001.
10. [10] W. Cao, F. Chu, L. Jia, H. Zhou, and Y. Zhang, "A Multi-Agent Deep Reinforcement Learning Anti-Jamming Spectrum-Access Method in LEO Satellites," Electronics, vol. 14, no. 16, p. 3307, 2025, doi: 10.3390/electronics14163307.
11. [11] J. Sormayli, M. Darvishi, K. Zarrinnegar, et al., "Real-Time jamming detection using windowing and hybrid machine learning models for pre-saturation alerts," Scientific Reports, vol. 15, no. 1, p. 24748, 2025, doi: 10.1038/s41598-025-10567-0.
12. [12] A. S. M. Mohammed, A. I. A. Taman, A. M. Hassan, et al., "Deep Learning Channel Estimation for OFDM 5G Systems with Different Channel Models," Wireless Personal Communications, vol. 128, no. 4, pp. 2891–2912, Feb. 2023, doi: 10.1007/s11277-022-10077-6.
13. [13] N. Guo and X. Wang, "Deep Learning-Based Optimization for Maritime Relay Networks," Applied Sciences, vol. 15, no. 3, p. 1160, 2025, doi: 10.3390/app15031160.
14. [14] S. Wang, S. A. R. Zaidi, M. Hafeez, et al., "Deep Transfer Learning: A Smarter Approach to Wireless Communication Networks," IEEE Commun. Stand. Mag., 2025 (Accepted for publication)
15. [15] D. Liu, K. Su, N. Yang, and J. Zhang, "Joint Resource Allocation for Maritime RIS–RSMA Communications Using Fractal-Aware Robust Deep Reinforcement Learning," Fractal Fract., vol. 10, no. 4, p. 223, 2026, doi: 10.3390/fractalfract10040223.
16. [16] J. Jena and P. K. Sahu, "Rain fade and Ka-band Spot Beam Satellite communication in India," in Proc. Recent Adv. Space Technol. Services Climate Change (RSTS & CC), Chennai, India, 2010, pp. 304-306, doi: 10.1109/RSTSCC.2010.5712816.
17. [17] P. M. Kalaivaanan et al., "Evaluation of Ka-Band Rain Attenuation for Satellite Communication in Tropical Regions Through a Measurement of Multiple Antenna Sizes," in IEEE Access, vol. 8, pp. 18007-18018, 2020, doi: 10.1109/ACCESS.2020.2966873.
18. [18] D. P. Wright and E. A. Ball, "IoT focused VHF and UHF propagation study and comparisons," IET Microw. Antennas Propag., vol. 15, no. 8, pp. 871-884, June 2021, doi: 10.1049/mia2.12101.
19. [19] A. Lopes, L. Fernandes, and P. Chaves, "Detection of Communications Channels in VHF Band for Enhanced Maritime Surveillance," Sensors, vol. 25, no. 23, p. 7258, 2025, doi: 10.3390/s25237258.
20. [20] H. Ismiyanto, Kemalsyah, and Bastari, "Intelligence Interconnect Communication System (IICS) pada Network Centric Warfare Operasi Udara," Jurnal Strategi Pertahanan Udara, vol. 9, no. 1, pp. 25–37, 2023.
21. [21] M. Z. Chowdhury, M. Shahjalal, S. Ahmed, and Y. M. Jang, "6G Wireless Communication Systems: Applications, Requirements, Technologies, Challenges, and Research Directions," in IEEE Open J. Commun. Soc., vol. 1, pp. 957-975, 2020, doi: 10.1109/OJCOMS.2020.3010270.
22. [22] M. Krichen and A. Mihoub, "Long Short-Term Memory Networks: A Comprehensive Survey," AI, vol. 6, no. 9, p. 215, 2025, doi: 10.3390/ai6090215.
23. [23] K. Greff, R. K. Srivastava, J. Koutník, B. R. Steunebrink, and J. Schmidhuber, "LSTM: A Search Space Odyssey," IEEE Trans. Neural Netw. Learn. Syst., vol. 28, no. 10, pp. 2222-2232, Oct. 2017, doi: 10.1109/TNNLS.2016.2582924.
24. [24] L. Shang et al., "CNN-LSTM Hybrid Model to Promote Signal Processing of Ultrasonic Guided Lamb Waves for Damage Detection in Metallic Pipelines," Sensors, vol. 23, no. 16, p. 7059, 2023, doi: 10.3390/s23167059.
25. [25] G. Luo, X. Wang, W. Zhao, S. Tao, and Z. Tang, "ReLU Neural Networks and Their Training," Mathematics, vol. 14, no. 1, p. 39, 2026, doi: 10.3390/math14010039.