FTM ve RSSI Ölçümleri Kullanarak Paket Kaybının Mekansal ve Sınıflandırma Tabanlı Analizi

Öz

Günümüzde Wi-Fi tabanlı iç mekan konumlandırma üzerine birçok çalışma yürütülmektedir. Çoğunlukla RSSI (Received Signal Strength Indicator) ve FTM (Fine Time Measurement) tabanlı yöntemler üzerine kurulu bu çalışmalar, ortamda bulunan fiziksel engeller ve donanım bazlı sebeplerle oluşan gürültülü sonuçlara çözüm aramaktadır. Veride olan gürültülere ek olarak, bağlantı kopması gibi uç durumlar da ortam dinamiğinin etkilerinden biridir. Veri iletimi sırasında bağlantı kopması ve paket kaybı gibi uç durumların nadiren gözlemlenmesi, bu olayların analitik olarak incelenmesini kısıtlamakta ve bu alandaki çalışmaların sınırlı kalmasına yol açmaktadır. Fakat yapılan birçok araştırma, anomali durumlarının rastgele olaylardan ziyade, çevresel birçok faktöre dayalı bir durum olduğunu göstermektedir. Bu çalışmanın amacı, farklı konumlardaki sinyal davranışları ile paket kaybı anomalileri arasındaki ilişkiyi incelemek için istatistiksel ve makine öğrenmesi tabanlı bir değerlendirme sunmaktır. Bu bağlamda, çevresel gürültü faktörlerine sahip bir endüstriyel otomasyon laboratuvarında elde edilmiş, RSSI ve FTM temelli mesafe ölçümünde paket kayıplarının analizini sağlayacak bir veri seti kullanılmıştır. Paket kaybı ile ilişkili anomali durumları ile normal çalışma koşullarını ayırt etmek için sınıflandırma ve Moran’s I tabanlı bir analiz yaklaşımı geliştirilmiştir. Bölgesel paket kaybı davranışlarını analiz edebilmek için her bir konumdan eşit örnekleme sayısı ile zaman serisi veri elde edilmiştir. Zaman serisi veri üzerinde kayan pencere tabanlı öznitelik çıkarımı uygulanmıştır. Kayan pencere yaklaşımı ile RSSI Slope ve FTM Distance Error Slope öznitelikleri çıkarılmıştır. Öznitelikler üzerinde korelasyon analizleri hem Pearson hem de Spearman katsayıları kullanılarak gerçekleştirilmiştir. Elde edilen ilişkilerin anlamlılığı, permütasyon testi kullanılarak değerlendirilmiştir. Anomali durumlarının tespiti problemi, RandomizedSearchCV ve Random Under Sampler yöntemleri kullanılarak denetimli öğrenme çerçevesinde ele alınmıştır. Sınıflandırma performansı Lojistik Regresyon, Random Forest XGBoost ve Isolation Forest modelleri kullanılarak değerlendirilmiştir. Ayrıca, bölgesel ilişkileri analiz etmek için Local Moran’s I yöntemi uygulanmıştır. Permütasyon testi sonuçları, Pearson korelasyonunun istatistiksel olarak anlamlı olmasına rağmen ilişkinin büyüklüğünün sınırlı olduğunu ortaya koymuştur. Elde edilen sonuçlar, paket kaybı olmayan normal koşullar altında, konumlar genelinde RSSI Slope ve FTM Distance Error Slope davranışlarının düşük korelasyon sergilediğini ve bu korelasyon değerlerinin standart sapmasının 0.0708-0.0855 aralığında bulunduğunu göstermektedir. Buna karşılık, paket kaybı anomalileri için bu standart sapma yaklaşık 4–6 kat artarak 0.3898-0.4163 seviyelerine ulaşmaktadır. Anomali tespitinde denetimli öğrenme sonuçlarında, XGBoost modeli, hem 0.7351 Doğruluk hem de 0.7348 Makro-F1 skoru üretirken, Isolation Forest en düşük etkinliği göstermiştir. Bölgesel analiz sonuçlarında, FTM Distance Error tabanlı özniteliklerin Moran’s I skoru 0.20 elde edilmiştir. Elde edilen bulgular, paket kaybı olaylarının rastgele değil, ortamın elektromanyetik ve yapısal özellikleriyle ilişkili sistematik davranışlar sergilediğini göstermektedir. Anomali durumları için elde edilen korelasyon değerlerindeki belirgin artış, paket kaybı durumlarının zamansal ve bölgesel olarak karakterize edilebileceğini ortaya koymaktadır. Denetimli öğrenme sonuçları, özellikle ağaç tabanlı yöntemlerin doğrusal modellere kıyasla anomali durumlarını daha etkin yakaladığını göstermiştir. Ayrıca, Local Moran’s I analizi ile elde edilen bölgesel kümelenme bulguları, paket kaybı davranışlarının bölgesel bağımlılık içerdiğini ve belirli bölgelerde yoğunlaştığını göstermektedir.

Anahtar Kelimeler

• Wi-Fi
• FTM
• RSSI
• Endüstriyel Nesnelerin İnterneti
• Bölge Kümeleme
• İç Mekan Konumlandırma

Spatial and Classification-Based Analysis of Packet Loss Using FTM and RSSI Measurements

Öz

Nowadays, numerous studies focus on Wi-Fi-based indoor positioning, primarily utilizing RSSI (Received Signal Strength Indicator) and FTM (Fine Time Measurement). These studies aim to mitigate noisy results caused by physical obstacles and hardware limitations. Beyond standard noise, edge cases like connection loss represent critical environmental dynamics. The rarity of such disconnections and packet loss events complicates analytical investigation and results in limited literature. However, research suggests that these anomalies are driven by environmental factors rather than being purely random events. This study provides a statistical and machine learning-based evaluation to examine the relationship between signal behaviors and packet loss anomalies across different locations. A dataset from an industrial automation laboratory is utilized to analyze packet losses within RSSI and FTM-based distance measurements. A classification and Moran’s I-based framework is developed to distinguish anomaly states from normal operating conditions. To analyze regional packet loss behaviors, time series data are collected with equal sampling at each location. Features such as RSSI Slope and FTM Distance Error Slope are extracted using a sliding window approach. Relationships are analyzed via Pearson and Spearman correlation coefficients, with significance validated through permutation tests. Anomaly detection is framed as a supervised learning problem, utilizing RandomizedSearchCV and Random Under Sampler. Performance is evaluated using Logistic Regression, Random Forest, XGBoost, and Isolation Forest models. Additionally, Local Moran’s I is applied to analyze regional spatial relationships. Permutation tests revealed that while Pearson correlations are statistically significant, the magnitude of the relationship remained limited. Under normal conditions, the standard deviation of correlation values for RSSI and FTM features ranged between 0.0708–0.0855. In contrast, for packet loss anomalies, this standard deviation increased approximately 4–6 fold, reaching levels between 0.3898 and 0.4163. In supervised learning trials, XGBoost achieved the highest performance with 0.7351 Accuracy and a 0.7348 Macro-F1 score, whereas Isolation Forest showed the lowest effectiveness. Regional analysis yielded a Moran’s I score of 0.20 for FTM distance error-based features. The findings demonstrate that packet loss events are not random but exhibit systematic behaviors linked to electromagnetic and structural environmental properties. The marked increase in correlation values during anomalies shows that packet loss can be characterized both temporally and regionally. Results indicate that tree-based methods capture these anomalies more effectively than linear models. Furthermore, Local Moran’s I analysis confirms that packet loss behaviors possess regional dependency and spatial clustering.

Anahtar Kelimeler

• Wi-Fi
• FTM
• RSSI
• IIoT
• Region clustering
• Indoor Positioning

Kaynakça

1. [1] A. S. Alluhaidan, L. A. Almusfar, A. Y. Shdefat, A. E. Mansour, D. S. Abdelminaam, Y. Alkady, “From GPS to AR: Leveraging Augmented Reality and Grid-Based Systems for Improved Indoor Navigation”, IEEE Access, 13, 55211–55230, 2025. https://doi.org/10.1109/ACCESS.2025.3552319.
2. [2] A. Sesyuk, S. Ioannou, M. Raspopoulos, “Radar-Based Millimeter-Wave Sensing for Accurate 3-D Indoor Positioning: Potentials and Challenges”, IEEE Journal of Indoor and Seamless Positioning and Navigation, 2, 61–75, 2024. https://doi.org/10.1109/JISPIN.2024.3359151.
3. [3] R. Jurdi, H. Chen, Y. M. Zhu, B. L. Ng, N. Dawar, C. R. Zhang, J. K.-H. Han, “WhereArtThou: A WiFi-RTT-Based Indoor Positioning System”, IEEE Access, 12, 41084–41101, 2024, https://doi.org/10.1109/ACCESS.2024.3377237.
4. [4] J. Nikonowicz, A. Mahmood, M. I. Ashraf, E. Björnson, M. Gidlund, “Indoor Positioning in 5G-Advanced: Challenges and Solution Toward Centimeter-Level Accuracy with Carrier Phase Enhancements”, IEEE Wireless Communications, 31(4), 268–275, 2024, https://doi.org/10.1109/MWC.023.2200633.
5. [5] H. Wymeersch, H. Chen, H. Guo, M. F. Keskin, B. M. Khorsandi, M. H. Moghaddam, A. Ramirez, K. Schindhelm, A. Stavridis, T. Svensson, V. Yajnanarayana, “6G Positioning and Sensing Through the Lens of Sustainability, Inclusiveness, and Trustworthiness”, IEEE Wireless Communications, 32(1), 68–75, 2025, https://doi.org/10.1109/MWC.011.2400055.
6. [6] Y. X. Guo, Q. W. Jiang, M. Y. Xu, W. Fang, Q. W. Liu, G. Yan, Q. H. Yang, H. Lu, “Resonant Beam Enabled DoA Estimation in Passive Positioning System”, IEEE Transactions on Wireless Communications, 23(11), 16290–16300, 2024. https://doi.org/10.1109/TWC.2024.3439703.
7. [7] M. Luckner, S. Sowik, P. Brida, “Selection of Signal Sources Influence at Indoor Positioning System”, IEEE Transactions on Wireless Communications, 23(1), 45–57, 2024. https://doi.org/10.1109/TWC.2023.3275537.
8. [8] G. Guo, R. Chen, K. Yan, P. Li, L. Yuan, L. Chen, “Multichannel and Multi-RSS Based BLE Range Estimation for Indoor Tracking of Commercial Smartphones”, IEEE Sensors Journal, 23(24), 30728–30738, 2023. https://doi.org/10.1109/JSEN.2023.3328711.

9. [9] Y. C. Wang, S. Y. Sun, F. J. Miao, Y. Xia, “WiFi Indoor Positioning Based on EGA-PF and Fernet Algorithm”, IEEE Access, 13, 16683–16696, 2025. https://doi.org/10.1109/ACCESS.2025.3532331.
10. [10] H. J. Zhang, G. H. Liu, Y. Xu, T. Jiang, “LoRaAid: Underground Joint Communication and Localization System Based on LoRa Technology”, IEEE Transactions on Wireless Communications, 23(5), 5248–5260, 2024. https://doi.org/10.1109/TWC.2023.3325330.
11. [11] H. Soganci, S. Gezici, H. V. Poor, “Accurate positioning in ultra-wideband systems”, IEEE Wireless Communications, 18(2), 19–27, 2011. https://doi.org/10.1109/MWC.2011.5751292.
12. [12] P. E. Numan, H. Park, C. Laoudias, S. Horsmanheimo, S. Kim, “Smartphone-Based Indoor Localization via Network Learning With Fusion of FTM/RSSI Measurements”, IEEE Networking Letters, 5(1), 21–25, 2023. https://doi.org/10.1109/LNET.2022.3226462.
13. [13] J. A. López-Pastor, P. Arques-Lara, J. J. Franco-Peñaranda, A. J. García-Sánchez, J. L. Gómez-Tornero, “Wi-Fi RTT-Based Active Monopulse RADAR for Single Access Point Localization”, IEEE Access, 9, 34755–34766, 2021. https://doi.org/10.1109/ACCESS.2021.3062085.
14. [14] H. Moon, H. Park, J. Seo, “HELPS for Emergency Location Service: Hyper-Enhanced Local Positioning System”, IEEE Wireless Communications, 31(4), 276–282, 2024. https://doi.org/10.1109/MWC.011.2300354.
15. [15] H. Vo, V. H. L. Nguyen, V. L. Tran, F. Ferrero, F.-Y. Lee, M.-H. Tsai, “Advance Path Loss Model for Distance Estimation Using LoRaWAN Network’s Received Signal Strength Indicator (RSSI)”, IEEE Access, 12, 83205–83216, 2024. https://doi.org/10.1109/ACCESS.2024.3412849.
16. [16] D. Scazzoli, S. Moro, V. Teeda, P. K. Upadhyay, M. Magarini, “Experimental Comparison of UAV-Based RSSI and AoA Localization”, IEEE Sensors Letters, 8(1), 6000104, 2024. https://doi.org/10.1109/LSENS.2023.3336375.
17. [17] B. Yang, L. Guo, R. Guo, M. Zhao, T. Zhao, “A Novel Trilateration Algorithm for RSSI-Based Indoor Localization”, IEEE Sensors Journal, 20(14), 8164–8172, 2020. https://doi.org/10.1109/JSEN.2020.2980966.
18. [18] F. B. Sorbelli, C. M. Pinotti, S. Silvestri, S. K. Das, “Measurement Errors in Range-Based Localization Algorithms for UAVs: Analysis and Experimentation”, IEEE Transactions on Mobile Computing, 21(4), 1291–1304, 2022. https://doi.org/10.1109/TMC.2020.3020584.
19. [19] R. Çakır, S. Doganalp, “Determination of beacon distance values according to ambient conditions by polynomial interpolation and artificial neural networks”, Pamukkale University Journal of Engineering Sciences, 31(3), 351–356, 2025. https://doi.org/10.5505/pajes.2024.48079.
20. [20] E. J. C. E. Silva, K. Y. G. V. Guedes, P. A. A. D. D. N. Alves, P. R. D. Ribeiro, A. O. Barradas Filho, “Systematic Review of Radio Wave Techniques for Indoor Positioning Systems”, IEEE Latin America Transactions, 23(3), 205–215, 2025. https://doi.org/10.1109/TLA.2025.10879196.
21. [21] N. S. Ahmad, “Recent Advances in WSN-Based Indoor Localization: A Systematic Review of Emerging Technologies, Methods, Challenges, and Trends”, IEEE Access, 12, 180674–180714, 2024. https://doi.org/10.1109/ACCESS.2024.3509516.
22. [22] M. Girolami, F. Mavilia, F. Furfari, P. Barsocchi, “An Experimental Evaluation Based on Direction Finding Specification for Indoor Localization and Proximity Detection”, IEEE Journal of Indoor and Seamless Positioning and Navigation, 2, 36–50, 2024. https://doi.org/10.1109/JISPIN.2023.3345268.
23. [23] Z. Y. Liu, T. Hakala, J. Hyyppä, A. Kukko, R. Z. Chen, “Performance Comparison of UWB IEEE 802.15.4z and IEEE 802.15.4 in Ranging, Energy Efficiency, and Positioning”, IEEE Sensors Journal, 24(8), 12481–12489, 2024. https://doi.org/10.1109/JSEN.2024.3368113.
24. [24] G. Y. Guo, R. Z. Chen, F. Ye, X. S. Peng, Z. Y. Liu, Y. J. Pan, “Indoor Smartphone Localization: A Hybrid WiFi RTT-RSS Ranging Approach”, IEEE Access, 7, 176767–176781, 2019. https://doi.org/10.1109/ACCESS.2019.2957753.
25. [25] Y. Shida, S. Aikawa, S. Yamamoto, “Correction of Round-Trip Time and Selection of Access Points for Estimating Wireless LAN Locations by Multilateration”, IEICE Communications Express, 13(8), 290–294, 2023. https://doi.org/10.23919/comex.2024TCL0002.
26. [26] Z. L. Liu, L. Chen, X. Zhou, Z. H. Jiao, G. Y. Guo, R. Z. Chen, “Machine Learning for Time-of-Arrival Estimation With 5G Signals in Indoor Positioning”, IEEE Internet of Things Journal, 10(11), 9782–9795, 2023. https://doi.org/10.1109/JIOT.2023.3234123.
27. [27] S. Aditya, A. F. Molisch, H. M. Behairy, “A Survey on the Impact of Multipath on Wideband Time-of-Arrival Based Localization”, Proceedings of the IEEE, 106(7), 1183–1203, 2018. https://doi.org/10.1109/JPROC.2018.2819638.
28. [28] M. Rea, A. Fakhreddine, D. Giustiniano, V. Lenders, “Filtering Noisy 802.11 Time-of-Flight Ranging Measurements From Commoditized WiFi Radios”, IEEE/ACM Transactions on Networking, 25(4), 2514–2527, 2017. https://doi.org/10.1109/TNET.2017.2700430.
29. [29] Y. Görmez, H. Arslan, Y. E. Işık, S. Tomaç, “A novel hybrid model for bluetooth low energy-based indoor localization using machine learning in the internet of things”, Pamukkale University Journal of Engineering Sciences, 30(1), 36–43, 2024. https://doi.org/10.5505/pajes.2023.57088.
30. [30] D. Xia, J. H. Shi, K. Wan, J. F. Wan, M. Martínez-García, X. Guan, “Digital Twin and Artificial Intelligence for Intelligent Planning and Energy-Efficient Deployment of 6G Networks in Smart Factories”, IEEE Wireless Communications, 30(3), 171–179, 2023. https://doi.org/10.1109/MWC.017.2200495.
31. [31] T. T. Sari, B. Bayram, B.-S. Kim, G. Seçinti, “Semantic Flow Control for Task-Oriented Position Tracking”, IEEE Access, 13, 53547–53557, 2025. https://doi.org/10.1109/ACCESS.2025.3552927.
32. [32] X. L. Yang, Z. Liu, W. Nie, W. He, Q. L. Pu, “AP Optimization for Wi-Fi Indoor Positioning-Based on RSS Feature Fuzzy Mapping and Clustering”, IEEE Access, 8, 153599–153609, 2020. https://doi.org/10.1109/ACCESS.2020.3018147.
33. [33] B. Altintas, T. Serif, “Improving RSS-Based Indoor Positioning Algorithm via K-Means Clustering”, 17th European Wireless 2011 - Sustainable Wireless Technologies, Vienna, Austria, 27-29 April 2011.
34. [34] J. X. Bi, L. Huang, H. J. Cao, G. B. Yao, W. G. Sang, J. Zhen, Y. Y. Liu, “Improved Indoor Fingerprinting Localization Method Using Clustering Algorithm and Dynamic Compensation”, ISPRS International Journal of Geo-Information, 10(9), 613, 2021. https://doi.org/10.3390/ijgi10090613.
35. [35] X. Feng, K. A. Nguyen, Z. Y. Luo, “WiFi Access Points Line-of-Sight Detection for Indoor Positioning Using the Signal Round Trip Time”, Remote Sensing, 14(23), 6052, 2022. https://doi.org/10.3390/rs14236052.
36. [36] E. Skýpalová, M. Boroš, T. Loveček, A. Veľas, “Innovative Indoor Positioning: BLE Beacons for Healthcare Tracking”, Electronics, 14(10), 2018, 2025. https://doi.org/10.3390/electronics14102018.
37. [37] Z. Y. Lyu, T. T. L. Chan, G. C. M. Leung, D. P. K. Lun, M. G. Pecht, “Counteracting Packet Loss in Fingerprint-Based Indoor Positioning via Spatially Regularized Entropy and Ground-Truth Prior Variational Inference”, IEEE Sensors Journal, 24(9), 15651–15664, 2024. https://doi.org/10.1109/JSEN.2024.3381077.
38. [38] Md. O. Khyam, S. S. Ge, X. D. Li, M. R. Pickering, “Highly Accurate Time-of-Flight Measurement Technique Based on Phase-Correlation for Ultrasonic Ranging”, IEEE Sensors Journal, 17(2), 434–443, 2017. https://doi.org/10.1109/JSEN.2016.2631244.
39. [39] M. Welzl, S. Islam, M. von Stephanides, “Real-Time TCP Packet Loss Prediction Using Machine Learning”, IEEE Access, 12, 159622–159634, 2024. https://doi.org/10.1109/ACCESS.2024.3488511.
40. [40] H. Wu, Y. Liu, S. S. Ni, G. Cheng, X. Y. Hu, “LossDetection: Real-Time Packet Loss Monitoring System for Sampled Traffic Data”, IEEE Transactions on Network and Service Management, 20(1), 30–45, 2023. https://doi.org/10.1109/TNSM.2022.3203389.
41. [41] D. Neuhold, J. F. Schmidt, J. Klaue, D. Schupke, C. Bettstetter, “Experimental Study of Packet Loss in a UWB Sensor Network for Aircraft”, Proceedings of the 20th ACM International Conference on Modelling, Analysis and Simulation of Wireless and Mobile Systems, Miami, FL, USA, 21-25 November 2017, 137–142. https://doi.org/10.1145/3127540.3127549.
42. [42] I. Srisomboon, S. Lee, “Positioning and Navigation Approaches Using Packet Loss-Based Multilateration for UAVs in GPS-Denied Environments”, IEEE Access, 12, 13355–13369, 2024. https://doi.org/10.1109/ACCESS.2024.3357148.
43. [43] K. J. Ahn, G. Cormode, S. Guha, A. McGregor, A. Wirth, “Correlation Clustering in Data Streams”, Algorithmica, 83(7), 1980–2017, 2021. https://doi.org/10.1007/s00453-021-00816-9.
44. [44] F. Bonchi, D. García-Soriano, K. Kutzkov, “Local correlation clustering”, arXiv, 2013, https://doi.org/10.48550/arXiv.1312.5105.
45. [45] C. S. Zhang, L. Luo, W. L. Xu, V. Ledwith, “Use of local Moran’s I and GIS to identify pollution hotspots of Pb in urban soils of Galway, Ireland”, Science of The Total Environment, 398(1–3), 212–221, 2008. https://doi.org/10.1016/j.scitotenv.2008.03.011.
46. [46] T. Y. Tsui, A. Derumigny, D. Peck, A. van Timmeren, A. Wandl, “Spatial clustering of waste reuse in a circular economy: A spatial autocorrelation analysis on locations of waste reuse in the Netherlands using global and local Moran’s I”, Frontiers in Built Environment, 8, 954642, 2022. https://doi.org/10.3389/fbuil.2022.954642.
47. [47] T. Koutsos, G. Menexes, “The role of Spatial Autocorrelation on spatially correlated data for Hierarchical Cluster Analysis”, Data Analysis Bulletin, 20(1), 109–119, 2024, https://ejournals.epublishing.ekt.gr/index.php/dab/article/view/33519.
48. [48] H. E. Gorjan, V. P. G. Jiménez, “Improving Indoor WiFi Localization by Using Machine Learning Techniques”, Sensors, 24(19), 6293, 2024. https://doi.org/10.3390/s24196293.
49. [49] S. Kalime, K. Sagar, “Anomaly Detection in Real-Time IoT Systems Using Machine Learning”, 2025 9th International Conference on Inventive Systems and Control (ICISC), Coimbatore, India, 12-13 August 2025, 783–791. https://doi.org/10.1109/ICISC65841.2025.11188217.
50. [50] I. Ozturk, E. Payal, AIoTx Research Lab, “AIoTxWallSense1.0”, Zenodo, 2026. https://doi.org/10.5281/zenodo.20592832.
51. [51] S. M. Miraftabzadeh, M. Longo, S. Leva, N. Matera, “Data anomaly detection in photovoltaic power time-series via unsupervised deep learning with insufficient information”, Sustainable Energy, Grids & Networks, 43, 101769, 2025. https://doi.org/10.1016/j.segan.2025.101769.