Е-Публикации
Технически университет - София

Abstract: Ensuring ground and air safety is top priority in building and maintaining existing aviation and future advanced air mobility systems. Changes to both aviation and connected systems can have an unexpected effect on safety and, therefore, must be evaluated and monitored. Such is the case with the ubiquitous rollout of 5G communications. In this study, flight hazards from 5G interference are reviewed and quantified. For the purpose, an artificial neural network language model was developed and trained to filter narratives of relevant safety reports from the pre- and post- 5G period. The results are interpreted in the context of a qualitative flight safety risk model. The model allows us to evaluate the current measures to assure the level of safety.

References

1. IATA & IFALPA, “Problem statement - 5G interference with radar altimeter frequency band”, (Montreal, 27 Nov 2020).
2. RTCA Paper No. 274-20/PMC-2073, “Assessment of C-Band mobile telecommunications interference impact on low range radar altimeter operations”, (October 7, 2020).
3. Airlines for America, 5G Interference Won't Fly. Learn Why..., https://www.airlines.org/5g/, Accessed on 17.01.2023.
4. ICAO, Information Paper: “Study of interference with radio altimeters from potential 5G implementations in the United States”, FSMP-WG/11 IP/07, 2021-03-01.
5. W. Bellamy, FAA Issues New Radar Altimeter 5G C-Band Risk Assessment Request to Aviation Industry, December 23, 2021, https://www.aviationtoday.com/2021/12/23/faa-issues-new-radar-altimeter-5g-c-bandrisk-assessment-request-aviation-industry/
6. M. Harris, How 5G's Rollout Rattled Hundreds of Pilots Complaints About Altimeter Failures Jumped After Consumer C-Band Deployment, 13 OCT 2022, https://spectrum.ieee.org/faa-5g
7. M. Thurber, NASA Reports Show Possible 5G Interference Issues, Aviation International News, Air Transport News, 2023, https://www.ainonline.com/aviation-news/air-transport/2023-02-01/nasa-reports-show-possible-5g-interference-issues, Accessed on 10.07.2023.
8. FAA, Radio Altimeters and Administration 5G C-Band Deployment, April 2022. https://www.faa.gov/air_traffic/flight_info/aeronav/acf/media/Presentations/22-01-Radio-Altimeters-5g-Deployment-ClausnitzerSilagyi.pdf
9. FAA, 5G and Aviation Safety, https://www.faa.gov/5g, Accessed on 10.07.2023.
10. M. H. Everdij, H. A. Blom, Safety Methods Database, Maintained by Netherlands Aerospace Centre NLR, The Netherlands, http://www.nlr.nl/documents/flyers/SATdb.pdf, Version 1.1, Aug. 2016.
11. K. Georgiev, “Aviation safety training methodology”, Heliyon 7(12), e08511 (2021).
12. M. Ivanov, K. Georgiev, “Impact of 5G on flight safety”, in Proceedings of International Scientific Conference on Aeronautics, Automotive and Railway Engineering and Technologies BulTrans-2022, pp. 6-13 (2022), (in Bulgarian).
13. L. Tanguy, N. Tulechki, A. Urieli, E. Hermann, C. Raynal, “Natural language processing for aviation safety reports: From classification to interactive analysis”, Computers in Industry 78, pp. 80-95 (2016).
14. K. Marev, K. Georgiev, “Automated aviation occurrences categorization”, in International Conference on Military Technologies (ICMT), (Brno, Czech Republic, 2019), pp. 1-5.
15. N. Amin, T. Yother, M. Johnson, J. Rayz, “Exploration of Natural Language Processing (NLP) applications in aviation”, Collegiate Aviation Review International 40(1), (2022).
16. X. Jing et al., “BERT for aviation text classification”, in AIAA AVIATION Forum, (12-16 June, 2023, San Diego, CA and Online).
17. J. Howard, S. Ruder, “Universal language model fine-tuning for text classification”, in ACL. Association for Computational Linguistics, (Melbourne, 2018).
18. OpenAI. GPT-4 Technical Report. 2023.
19. J. Ge, H. Luo, Y. Kim, J. Glass, “Entailment as Robust Self-Learner”, in Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics 1: Long Papers, (Toronto, Canada, Association for Computational Linguistics. 2023), pp. 13803-13817.
20. Guidelines for Hazard Evaluation Procedures, Second Edition with Worked Examples (Center for Chemical Process Safety, 1992).
21. J. Howard, S. Gugger, “NLP Deep Dive: RNNs”, Chapter 10 in Deep Learning for Coders with Fastai and PyTorch: AI Applications Without a PhD (O'Reilly Media, Incorporated, 2020).

Issue

Copyright AIP Publishing