GELİŞMİŞ GİYİLEBİLİR SENSÖR BİLGİSİ İŞLEME: İNSAN AKTİVİTESİNİN TANINMASI İÇİN OPTİMİZE EDİLMİŞ KALMAN FİLTRESİ

Abstract

Bu çalışma, sinyal filtreleme ve tahmini için geleneksel ve gelişmiş Kalman filtresi yöntemlerini incelemektedir. İlk olarak sentetik sinüzoidal veriler kullanılmış, bunu insan aktivite sinyalleri izlemiştir. Başlangıçta, sentetik sinyaller üzerinde standart Kalman Filtresi (KF), Genişletilmiş Kalman Filtresi (EKF), Kokusuz Kalman Filtresi (UKF) ve otomatik olarak ayarlanmış parametrelere sahip Bayes-Optimize Edilmiş Kalman Filtresi'nin (Bayes-KF) filtreleme performansları araştırılmıştır. İnsan aktivite sinyalleri, UCI'nin İnsan Aktivitesi Tanıma (HAR) veri setinden elde edilen ivme verileri kullanılarak tahmin edilmiştir. Tahmin doğruluğu açısından, Bayes-KF yaklaşımı, filtre parametrelerini (Q ve R) iyileştirerek geleneksel KF, EKF ve UKF yöntemlerinden daha iyi performans göstermiştir. Her yaklaşımın performansını değerlendirmek için Ortalama Karesel Hata (MSE), Ortalama Karesel Hatanın Kökü (RMSE) ve Ortalama Mutlak Hata (MAE) göstergeleri kullanılmıştır. Bayes-KF yöntemi, MSE metriğine göre 5,0996 x 10-5 değeriyle en düşük tahmin hatasını üretti. Karşılaştırmalı çalışma, Bayes-KF'nin hem sentetik hem de gerçek dünyadaki insan aktivite sinyalleri için güvenilir bir yöntem olduğunu göstermektedir. Bu nedenle, hareket analizi ve sensör verisi işleme gerektiren uygulamalarda özellikle faydalıdır.

Keywords

• Sinyal İşleme
• Kalman Fİltre
• Bayez Optimizasyon
• İnsan Aktivitesi Tanımlama

Supporting Institution

Uygulama yoktur.

Ethical Statement

Uygulama yoktur.

Thanks

Uygulama yoktur.

ADVANCED WEARABLE SENSOR INFORMATION PROCESSING: OPTIMIZED KALMAN FILTER FOR RECOGNITION OF HUMAN ACTIVITY

Abstract

This study examines traditional and advanced Kalman filter methods for signal filtering and estimation. Synthetic sinusoidal data were first used, followed by human activity signals. Initially, the filtering performances of the standard Kalman Filter (KF), Extended Kalman Filter (EKF), Unscented Kalman Filter (UKF), and Bayesian-Optimized Kalman Filter (Bayes-KF) with automatically tuned parameters were investigated on synthetic signals. Human activity signals were estimated using acceleration data from the UCI's Human Activity Recognition (HAR) dataset. In terms of prediction accuracy, the Bayesian-KF approach outperformed the traditional KF, EKF, and UKF methods by improving the filter parameters (Q and R). Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE) indicators were used to evaluate the performance of each approach. The Bayesian-KF method produced the lowest prediction error, with a value of 5.0996 x 10-5 based on the MSE metric. The comparative study demonstrates that Bayesian-KF is a reliable method for both synthetic and real-world human activity signals. Therefore, it is particularly useful in applications requiring motion analysis and sensor data processing.

Keywords

• Signal processing
• Kalman filter
• Bayes optimization
• Human activity recognition

Supporting Institution

Not applicable.

Ethical Statement

Not applicable.

Thanks

Not applicable.

References

1. Anguita, D., Ghio, A., Oneto, L., Parra, X., Reyes-Ortiz, J.L., 2013. A public domain dataset for human activity recognition using smartphones. Proceedings of the 21st European Symposium on Artificial Neural Networks (ESANN), Bruges, Belgium, 437–442.
2. Bao, L., Intille, S.S., 2004. Activity recognition from user-annotated acceleration data. Pervasive Computing, Vol. 3001, Berlin, Heidelberg: Springer, 1–17.
3. Belbekri, N., Mezghanni, S., Bouamra, F., 2025. Adaptive feedback-driven segmentation for continuous multi-label human activity recognition. Applied Sciences, 15 (6), 2905.
4. Bulling, A., Blanke, U., Schiele, B., 2014. A tutorial on human activity recognition using body-worn inertial sensors. ACM Computing Surveys (CSUR), 46 (3), 1–33.
5. Calandra, R., Seyfarth, A., Peters, J., Deisenroth, M.P., 2016. Bayesian optimization for learning gaits under uncertainty. NIPS Workshop on Bayesian Optimization.
6. Castañeda, C.E., 2025. AI-based optimization of a neural discrete-time sliding mode controller via Bayesian, particle swarm, and genetic algorithms. Robotics, 14 (9), 128.
7. Chen, L., Hoey, J., Nugent, C., Cook, D., Yu, Z., 2012. Sensor-based activity recognition. IEEE Transactions on Systems, Man, and Cybernetics, 42 (6), 790–808.
8. Elgohary, A.A., Gwinnell, B., Augustine, J., 2025. Investigating the performance of EKF, UKF, and PF for quadrotor position estimation in hurricane wind disturbances. arXiv preprint arXiv:2509.13243.

9. Gao, M., Li, J., Wang, W., Wang, R., Luo, J., Li, J., 2025. Bayesian bidirectional long short-term memory-based kinematics-dynamics fusion for fault-tolerant vehicle state estimation under yaw rate sensor failures. Engineering Applications of Artificial Intelligence, Elsevier.
10. Grewal, M.S., Andrews, A.P., 2015. Kalman Filtering: Theory and Practice Using MATLAB, 4th ed. Hoboken, NJ: Wiley.
11. Hutter, F., Hoos, H.H., Leyton-Brown, K., 2011. Sequential model-based optimization for general algorithm configuration. Proceedings of the LION Conference, Berlin, Heidelberg: Springer, 507–523.
12. John, A., Cardiff, B., John, D., 2024. A review on multisensor data fusion for wearable health monitoring. arXiv preprint arXiv:2412.05895.
13. Ma, J., Wang, F., Liu, B., Tang, M., 2025a. UAV rapid response fire extinguishing method for multi-point fires in complex scenarios. Proceedings of the 6th International Conference on Intelligent Systems and Applications (ISA), IEEE.
14. Ma, Y., Zhang, G., Ye, S., An, D., 2025b. Variable Bayesian-based maximum correntropy criterion cubature Kalman filter with application to target tracking. Entropy, 27 (10), 997.
15. Ordóñez, F., Roggen, D., 2016. Deep convolutional and LSTM recurrent neural networks for multimodal wearable activity recognition. Sensors, 16 (1), 115.
16. Shahriari, B., Swersky, K., Wang, Z., Adams, R.P., De Freitas, N., 2016. Taking the human out of the loop: A review of Bayesian optimization. Proceedings of the IEEE, 104 (1), 148–175.
17. Simon, D., 2006. Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches. Hoboken, NJ: Wiley.
18. Snoek, J., Larochelle, H., Adams, R.P., 2012. Practical Bayesian optimization of machine learning algorithms. Advances in Neural Information Processing Systems (NeurIPS), 25, 2951–2959.
19. Torun, H.M., Swaminathan, M., Kavungal Davis, A., Bellaredj, M.L.F., 2018. A global Bayesian optimization algorithm and its application to integrated system design. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 26 (4), 792–802.
20. Wan, E.A., Van der Merwe, R., 2000. The unscented Kalman filter for nonlinear estimation. Proceedings of IEEE ASSPCC Symposium, 153–158.
21. Weigend, F.C., Liu, X., Ben Amor, H., 2023. Probabilistic differentiable filters enable ubiquitous robot control with smartwatches. arXiv preprint arXiv:2309.06606.
22. Welch, G., Bishop, G., 2006. An introduction to the Kalman filter. University of North Carolina at Chapel Hill, Technical Report 95-041.
23. Wijethunga, S., Senavirathna, S.K., Dissanayake, K., 2023. IMU-based modularized wearable device for human motion classification. arXiv preprint arXiv:2303.16468.
24. Zhang, P., Li, J., Ding, D., Wang, H., Wang, K., Wang, X., 2025. A two-step cellular network traffic forecasting method integrating decomposition and deep neural networks based on Bayesian joint parameter optimization. IEEE Transactions on Network and Service Management, 22 (5), 4274–4287.
25. Zhu, Z., Rezayat Sorkhabadi, S.M., Gu, Y., Zhang, W., 2022. Design and evaluation of an invariant extended Kalman filter for trunk motion estimation with sensor misalignment. arXiv preprint arXiv:2205.07800.