Accuracy and Stability Trade-off in RSSI Window Aggregation for MLP-Based Wi-Fi Indoor Positioning

Authors

Wahyu Arthanugraha
Universitas Muhammadiyah Parepare image/svg+xml
Abdul Halim Ismail
University of Malaysia, Perlis image/svg+xml
Shahrul Nizam Yaakob
University of Malaysia, Perlis image/svg+xml
DOI: https://doi.org/10.52549/ijeei.v14i3.8256
Download PDF
Published: 30-09-2026 • Issue: Vol. 14 No. 3: September 2026 • Section: Regular paper

Abstract

Wi-Fi RSSI fingerprinting is practical for indoor positioning, but short-term signal fluctuations can reduce classification consistency. This study evaluates how fixed RSSI window size affects the accuracy, stability, and sample availability of an MLP-based indoor positioning system. RSSI from three Access Points was collected at 45 ordered corridor positions. Non-overlapping windows with n=1,3,5,7,9,11,15,21 and 29 were represented by the mean and standard deviation of three RSSI channels. The MLP was assessed using stratified 5-fold cross-validation with supplementary order-blocked validation. Relative to the n=1 baseline Accuracy of 68.24%, n=7 achieved 90.60 ± 0.71% Accuracy, 95.08 ± 1.31% Top-2 Accuracy, 0.9047 ± 0.0060 Macro-F1, and 0.4319 ± 0.1364 Ordinal Error. Larger windows achieved higher mean Accuracy but sharply reduced window-level samples and increased fold-to-fold variability. Order-blocked validation produced more conservative estimates and shifted the strongest performance region toward n=9 to n=11. These findings show that fixed-window selection should consider predictive performance, evaluation stability, sample availability, and validation design rather than maximum Accuracy alone.
Keywords: Wi-Fi indoor positioning; RSSI window aggregation; multilayer perceptron; fingerprinting; accuracy and stability trade-off

References

[1] He S, Chan SHG. Wi-Fi fingerprint-based indoor positioning: Recent advances and comparisons. IEEE Communications Surveys and Tutorials 2016;18:466–90. https://doi.org/10.1109/COMST.2015.2464084.

[2] Shang S, Wang L. Overview of WiFi fingerprinting‐based indoor positioning. IET Communications 2022;16:725–33. https://doi.org/10.1049/cmu2.12386.

[3] Singh N, Choe S, Punmiya R, Kaur N. XGBLoc: XGBoost-Based Indoor Localization in Multi-Building Multi-Floor Environments. Sensors 2022;22:6629. https://doi.org/10.3390/s22176629.

[4] Lin X, Gan J, Jiang C, Xue S, Liang Y. Wi-Fi-Based Indoor Localization and Navigation: A Robot-Aided Hybrid Deep Learning Approach. Sensors 2023;23:6320. https://doi.org/10.3390/s23146320.

[5] Martín-Frechina S, Dura E, Miralles I, Torres-Sospedra J. From Fingerprinting to Advanced Machine Learning: A Systematic Review of Wi-Fi and BLE-Based Indoor Positioning Systems. Sensors 2025;25:6946. https://doi.org/10.3390/s25226946.

[6] Neupane I, Shahrestani S, Ruan C. Wi-Fi RSS Fingerprinting-Based Indoor Localization in Large Multi-Floor Buildings. Electronics (Basel) 2025;15:183. https://doi.org/10.3390/electronics15010183.

[7] Zholamanov B, Saymbetov A, Nurgaliyev M, Bolatbek A, Dosymbetova G, Kuttybay N, et al. RSSI Fingerprint-Based Indoor Localization Solutions Using Machine Learning Algorithms: A Comprehensive Review. Smart Cities 2025;8:153. https://doi.org/10.3390/smartcities8050153.

[8] Yang L, Finnerty P, Ohta C. Indoor localization using router-to-router RSSI and transfer learning for dynamic environments. Ad Hoc Networks 2025;178:103938. https://doi.org/10.1016/j.adhoc.2025.103938.

[9] Silva I, Pendão C, Moreira A. Collection of a Continuous Long-Term Dataset for the Evaluation of Wi-Fi-Fingerprinting-Based Indoor Positioning Systems. Sensors 2022;22:8585. https://doi.org/10.3390/s22228585.

[10] Maduranga MWP, Tilwari V, Abeysekera R. Improved-RSSI-based indoor localization by using pseudo-linear solution with machine learning algorithms. Journal of Electrical Systems and Information Technology 2024;11:10. https://doi.org/10.1186/s43067-024-00138-0.

[11] Wan S, Gu C, Shu Y, Shi Z. Last-seen time is critical: Revisiting RSSI-based WiFi indoor localization. Signal Processing 2025;231:109756. https://doi.org/10.1016/j.sigpro.2024.109756.

[12] Yang T, Cabani A, Chafouk H. A Survey of Recent Indoor Localization Scenarios and Methodologies. Sensors 2021;21:8086. https://doi.org/10.3390/s21238086.

[13] Ma Z, Shi K. Few-Shot Learning for WiFi Fingerprinting Indoor Positioning. Sensors 2023;23:8458. https://doi.org/10.3390/s23208458.

[14] Cha J, Lim E. A hierarchical auxiliary deep neural network architecture for large-scale indoor localization based on Wi-Fi fingerprinting. Appl Soft Comput 2022;120:108624. https://doi.org/10.1016/j.asoc.2022.108624.

[15] Feng X, Nguyen KA, Luo Z. A survey of deep learning approaches for WiFi-based indoor positioning. Journal of Information and Telecommunication 2022;6:163–216. https://doi.org/10.1080/24751839.2021.1975425.

[16] Gao B, Yang F, Cui N, Xiong K, Lu Y, Wang Y. A Federated Learning Framework for Fingerprinting-Based Indoor Localization in Multibuilding and Multifloor Environments. IEEE Internet Things J 2023;10:2615–29. https://doi.org/10.1109/JIOT.2022.3214211.

[17] Bi J, Zhao M, Yao G, Cao H, Feng Y, Jiang H, et al. PSOSVRPos: WiFi indoor positioning using SVR optimized by PSO. Expert Syst Appl 2023;222:119778. https://doi.org/10.1016/j.eswa.2023.119778.

[18] Zhu H, Cheng L, Li X, Yuan H. Neural-Network-Based Localization Method for Wi-Fi Fingerprint Indoor Localization. Sensors 2023;23. https://doi.org/10.3390/s23156992.

[19] Turgut Z, Kakisim AG. An explainable hybrid deep learning architecture for WiFi-based indoor localization in Internet of Things environment. Future Generation Computer Systems 2024;151:196–213. https://doi.org/10.1016/j.future.2023.10.003.

[20] Arslantas H, Okdem S. Indoor Localization With an Autoencoder-Based Convolutional Neural Network. IEEE Access 2024;12:46059–66. https://doi.org/10.1109/ACCESS.2024.3382135.

[21] Yoo J. Wi-Fi Fingerprint Indoor Localization by Semi-Supervised Generative Adversarial Network. Sensors 2024;24:5698. https://doi.org/10.3390/s24175698.

[22] Kim D, Park J-H, Suh Y-J. A Wi-Fi Fingerprinting Indoor Localization Framework Using Feature-Level Augmentation via Variational Graph Auto-Encoder. Electronics (Basel) 2025;14:2807. https://doi.org/10.3390/electronics14142807.

[23] Li S, Kim KS, Tang Z, Smith JS. Hierarchical Stagewise Training of Linked Deep Neural Networks for Multibuilding and Multifloor Indoor Localization Based on Wi-Fi RSSI Fingerprinting. IEEE Sens J 2025;25:23341–51. https://doi.org/10.1109/JSEN.2024.3455554.

[24] Esmaeili Gorjan H, Gil Jiménez VP. Improving Indoor WiFi Localization by Using Machine Learning Techniques. Sensors 2024;24:6293. https://doi.org/10.3390/s24196293.

[25] Tang Z, Li S, Kim KS, Smith JS. Multi-Dimensional Wi-Fi Received Signal Strength Indicator Data Augmentation Based on Multi-Output Gaussian Process for Large-Scale Indoor Localization †. Sensors 2024;24. https://doi.org/10.3390/s24031026.

[26] Reyes JMR, Ho IW-H, Mak M-W. Wi-Fi CSI fingerprinting-based indoor positioning using deep learning and vector embedding for temporal stability. Expert Syst Appl 2025;264:125802. https://doi.org/10.1016/j.eswa.2024.125802.

[27] Shen Q, Cui Y. WIFI Fingerprint Correction for Indoor Localization Based on Distance Metric Learning. IEEE Sens J 2024;24:36167–77. https://doi.org/10.1109/JSEN.2024.3462759.

[28] Aviles M, Alvarez-Alvarado JM, Robles-Ocampo J-B, Sevilla-Camacho PY, Rodríguez-Reséndiz J. Optimizing RNNs for EMG Signal Classification: A Novel Strategy Using Grey Wolf Optimization. Bioengineering 2024;11:77. https://doi.org/10.3390/bioengineering11010077.

[29] Sun C, Zhou J, Jang K, Kim Y. Indoor Localization Based on Integration of Wi-Fi with Geomagnetic and Light Sensors on an Android Device Using a DFF Network. Electronics (Basel) 2023;12:5032. https://doi.org/10.3390/electronics12245032.

[30] Tseng CH, Tsaur W-J. FFK: Fourier-Transform Fuzzy-c-means Kalman-Filter Based RSSI Filtering Mechanism for Indoor Positioning. Sensors 2023;23:8274. https://doi.org/10.3390/s23198274.

[31] Wei Z, Zhou Z. A Combined Filtering Method for ZigBee Indoor Distance Measurement. Sensors 2024;24:3164. https://doi.org/10.3390/s24103164.

[32] Antonopoulos K, Skandamis T, Alogdianakis G, Faliagka E, Antonopoulos CP, Voros N. Indoor Localization and ADL Monitoring via RSSI-Driven ML with Feedback Process. Electronics (Basel) 2025;14:3759. https://doi.org/10.3390/electronics14193759.

[33] Pawan, Dhiman R. Machine learning techniques for electroencephalogram based brain-computer interface: A systematic literature review. Measurement: Sensors 2023;28:100823. https://doi.org/10.1016/j.measen.2023.100823.

[34] Ismail AH, Kitagawa H, Tasaki R, Terashima K. WiFi RSS fingerprint database construction for mobile robot indoor positioning system. 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC), IEEE; 2016, p. 001561–6. https://doi.org/10.1109/SMC.2016.7844461.

[35] Gomes EL, Fonseca MSP, Lazzaretti AE, Munaretto A, Guerber CR. Sliding Window, Hierarchical Classification, Regression, and Genetic Algorithm for RFID Indoor Positioning Systems. Expert Syst Appl 2024;238:122298. https://doi.org/10.1016/j.eswa.2023.122298.

[36] Kamal AHM, Alam MdGR, Hassan MR, Apon TS, Hassan MM. Explainable indoor localization of BLE devices through RSSI using recursive continuous wavelet transformation and XGBoost classifier. Future Generation Computer Systems 2023;141:230–42. https://doi.org/10.1016/j.future.2022.11.001.

How to Cite

[1]
W. Arthanugraha, A. H. Ismail, and S. N. Yaakob, “Accuracy and Stability Trade-off in RSSI Window Aggregation for MLP-Based Wi-Fi Indoor Positioning”, IJEEI, vol. 14, no. 3, pp. 861–873, Sep. 2026, doi: 10.52549/ijeei.v14i3.8256.