Estimating Object Location in RF Communication by Using RSSI Values Through k-NN and Deep Learning Techniques

Öz

GPS-based positioning faces significant challenges in accuracy and reliability, especially due to environmental factors such as signal interruptions, multi-path propagation, and poor satellite visibility. This study explores using RF signal strength (RSSI) to estimate object positions, comparing different algorithms in indoor and open-air environments. For indoor localization, the Mean Absolute Error (MAE) algorithm achieved a limited 66% success rate, primarily due to RSSI fluctuations caused by signal reflections from obstacles. In open-air settings, Neural Net Fitting (NNF) outperformed Machine Learning (ML). NNF demonstrated high accuracy of approximately 94.05%, indicating effective learning and minimal overfitting. The ML model achieved 74.4% accuracy, showing less stability and overall accuracy compared to NNF. Results suggest NNF is more effective for RF-based localization, particularly in open-air environments where signal propagation is less complex.

Anahtar Kelimeler

• RSSI
• RF Positioning
• Object localization
• indoor-outdoor localiziation
• wireless sensor networks

RF Haberleşmesinde RSSI Değerleri Kullanılarak k-NN ve Derin Öğrenme Yöntemleri ile Nesne Konumunun Tahmin Edilmesi

Öz

GPS tabanlı konumlandırma, özellikle sinyal kesintileri, çoklu yol (multi-path) yayılımı ve zayıf uydu görünürlüğü gibi çevresel faktörler nedeniyle doğruluk ve güvenilirlik açısından önemli zorluklarla karşı karşıyadır. Bu çalışma, nesne konumlarını tahmin etmek için RF sinyal gücünü (RSSI) kullanmayı araştırmakta ve farklı algoritmaları kapalı alan ve açık hava ortamlarında karşılaştırmaktadır. Kapalı alan konumlandırmasında, Ortalama Mutlak Hata (MAE) algoritması, sinyallerin engellerden yansıması nedeniyle oluşan RSSI dalgalanmalarından ötürü %66 ile sınırlı bir başarı oranına ulaşmıştır. Açık hava ortamlarında ise Sinir Ağı Uydurması (NNF), Makine Öğrenimi'ne (ML) kıyasla daha iyi performans göstermiştir. NNF yaklaşık %94,05 gibi yüksek bir doğruluk oranı ile etkili öğrenme gerçekleştirmiş ve aşırı öğrenme (overfitting) göstermemiştir. ML modeli ise %74,4 doğruluk oranına ulaşarak NNF'ye kıyasla daha düşük kararlılık ve genel doğruluk sergilemiştir. Sonuçlar, sinyal yayılımının daha az karmaşık olduğu açık hava ortamlarında RF tabanlı konumlandırma için NNF'nin daha etkili olduğunu göstermektedir.

Anahtar Kelimeler

• RSSI
• RF konumlama
• Nesne konumlandırma
• kapalı-açık alan konumlandırma
• kablosuz sensor ağları

Kaynakça

1. [1] M. Feng, S. C. Shen, D. C. Caruth, and J. J. Huang, “Device technologies for RF front-end circuits in next-generation wireless communications,” Proceedings of the IEEE, vol. 92, no. 2, pp. 354–374, 2004, doi: 10.1109/JPROC.2003.821903.
2. [2] B. Clerckx, R. Zhang, R. Schober, D. W. K. Ng, D. I. Kim, and H. V. Poor, “Fundamentals of wireless information and power transfer: From RF energy harvester models to signal and system designs,” IEEE Journal on Selected Areas in Communications, vol. 37, no. 1, pp. 4–33, 2019, doi: 10.1109/JSAC.2018.2872615.
3. [3] Y. Zeng, B. Clerckx, and R. Zhang, “Communications and Signals Design for Wireless Power Transmission,” IEEE Transactions on Communications, vol. 65, no. 5, pp. 2264–2290, 2017, doi: 10.1109/TCOMM.2017.2676103.
4. [4] A. A. Abidi, “Low-power radio-frequency IC’s for portable communications,” Integrated Circuits for Wireless Communications, vol. 83, no. 4, pp. 3–28, 1998, doi: 10.1109/9780470544952.ch1.
5. [5] D. N. Purnamasari and M. Ulum, “LoRa 920 MHz as a Green Technology in Maritime Education : Optimizing Modulation Settings for Real-Time Environmental Data,” vol. 05004, 2024.
6. [6] P. D. P. Adi and A. Kitagawa, “A performance of radio frequency and signal strength of LoRa with BME280 sensor,” TELKOMNIKA (Telecommunication Computing Electronics and Control), vol. 18, no. 2, pp. 649–660, Apr. 2020, doi: 10.12928/TELKOMNIKA.V18I2.14843.
7. [7] T. Du, Y. H. Su, and A. Sample, “2D+Depth RF Localization via a Low-Cost Receiver,” IEEE Wireless Communications and Networking Conference, WCNC, pp. 1–6, 2024, doi: 10.1109/WCNC57260.2024.10571215.
8. [8] A. Kleniatis, A. Dimitriou, and A. Bletsas, “Device-Free Localization of Multiple Humans with Passive RFID and Joint RSSI-Phase Techniques,” 2024 IEEE International Conference on RFID, RFID 2024, pp. 35–40, 2024, doi: 10.1109/RFID62091.2024.10582592.

9. [9] V. Teeda, S. Moro, D. Scazzoli, L. Reggiani, and M. Magarini, “Differentiation and Localization of Ground RF Transmitters Through RSSI Measures from a UAV,” IEEE Trans Veh Technol, vol. PP, pp. 1–10, 2024, doi: 10.1109/TVT.2024.3458416.
10. [10] L. Peled-Eitan and E. Greenberg, “Mobile User Localization Based on Wireless Sensor Network Signals in Smart Cities,” 2024 IEEE International Conference on Microwaves, Communications, Antennas, Biomedical Engineering and Electronic Systems, COMCAS 2024, pp. 1–4, 2024, doi: 10.1109/COMCAS58210.2024.10666174.
11. [11] W. Chen, T. Boroushaki, I. Perper, and F. Adib, “Reinforcement Learning for RFID Localization,” 2024 IEEE International Conference on RFID, RFID 2024, pp. 47–52, 2024, doi: 10.1109/RFID62091.2024.10582639.
12. [12] R. Fu, D. Xiao, and Y. Fan, “A novel cell phone localization solution for trapped victims based on compressed RSSI fluctuation range and PSO-BP neural network,” Meas. J. Int. Meas. Confed., vol. 225, no. September 2023, p. 114014, 2024, doi: 10.1016/j.measurement.2023.114014.
13. [13] A. Ferrero-López, A. J. Gallego, and M. A. Lozano, “Bluetooth low energy indoor positioning: A fingerprinting neural network approach,” Internet Things (The Netherlands), vol. 31, no. March, p. 101565, 2025, doi: 10.1016/j.iot.2025.101565.
14. [14] A. G. Kakisim and Z. Turgut, “Multi-channel convolutional neural network with attention mechanism using dual-band WiFi signals for indoor positioning systems in smart buildings,” Internet Things (The Netherlands), vol. 29, no. May 2024, p. 101435, 2025, doi: 10.1016/j.iot.2024.101435.
15. [15] A. S. Lutakamale, H. C. Myburgh, and A. de Freitas, “RSSI-based fingerprint localization in LoRaWAN networks using CNNs with squeeze and excitation blocks,” Ad Hoc Networks, vol. 159, no. March, p. 103486, 2024, doi: 10.1016/j.adhoc.2024.103486.
16. [16] R. R. Guerra, A. Vizziello, P. Savazzi, E. Goldoni, and P. Gamba, “Forecasting LoRaWAN RSSI using weather parameters: A comparative study of ARIMA, artificial intelligence and hybrid approaches,” Comput. Networks, vol. 243, no. September 2023, p. 110258, 2024, doi: 10.1016/j.comnet.2024.110258.