Abstract
The Received Signal Strength Indicator (RSSI) timeseries have been used as a primary variable in many cybersecurity applications, such as wireless-node profiling for the purpose of authentication, localization, and physical security perimeter monitoring. Previous research on the use of RSSI-based wireless node profiling assumes that RSSI timeseries are stationary and independent identically distributed (i.i.d.). Unfortunately, in real-world environments, this assumption is far from the truth and would negatively impact the performance of any system or application built on idealized models of RSSI timeseries data. In other words, a set of real-world RSSI values (depending on the variability of noise produced by objects in the environment) are typically made of sub-segments each with its own statistical characteristics (e.g., mean and variance). Therefore, before any modelling attempt, one must consider breaking down a given RSSI dataset into its constituting sub-segments. Unfortunately, the effect of environmental variables on RSSI values tend to be random, which makes the problem of RSSI timeseries segmentation even more challenging. Thus, it is necessary to study the effectiveness of existing notable timeseries segmentation algorithms against a dataset of RSSI values. The main contributions of our work are that (1) we have demonstrated the non-stationary nature of RSSI timeseries by collecting samples from a real-world IoT network, and (2) through real-world experimentation we have compared the effectiveness of notable timeseries segmentation methods for the discovery of sub-segments in a RSSI timeseries dataset. Our work highlights the importance of accurate detection of change points in RSSI timeseries, which can further facilitate optimal selection and performance of the respective system’s cost and objective functions. Finally, we demonstrate that the \(\ell _1\) cost function can capture a meaningful relationship between neighboring data points in a RSSI timeseries and can result in a stable segmentation across different search methods.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Wenjia, W., Xiaolin, G., Dong, K., Shi, X., Yang, M.: PRAPD: a novel received signal strength-based approach for practical rogue access point detection. Int. J. Distrib. Sens. Netw. 14(8) (2018). https://doi.org/10.1177/1550147718795838
Moosavirad, S.M., Kabiri, P., Mahini, H.: RSSAT: a wireless intrusion detection system based on received signal strength acceptance test. J. Adv. Comput. Res. 4(1), 65–80 (2013)
Demirbas, M., Song, Y.: An RSSI-based scheme for sybil attack detection in wireless sensor networks. In: 2006 International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM 2006), pp. 5-pp. IEEE (2006)
Madani, P., Vlajic, N., Maljevic, I.: Randomized moving target approach for mac-layer spoofing detection and prevention in IoT systems. Digital Threats Res. Pract. (2022)
Madani, P., Vlajic, N.: RSSI-based MAC-layer spoofing detection: deep learning approach. J. Cybersecur. Privacy 1(3), 453–469 (2021)
Sandeepa, C., Moremada, C., Dissanayaka, N., Gamage, T., Liyanage, M.: Social interaction tracking and patient prediction system for potential COVID-19 patients. In: 2020 IEEE 3rd 5G World Forum (5GWF), pp. 13–18. IEEE (2020)
Sugano, M., Kawazoe, T., Ohta, Y., Murata, M.: Indoor localization system using RSSI measurement of wireless sensor network based on ZigBee standard. Wirel. Opt. Commun. 538, 1–6 (2006)
Truong, C., Oudre, L., Vayatis, N.: Selective review of offline change point detection methods. Signal Process. 167, 107299 (2020)
Frick, K., Munk, A., Sieling, H.: Multiscale change point inference. J. R. Stat. Soc. Ser. B (Stat. Methodol.) 76(3), 495–580 (2014)
Nam, C.F.H., Aston, J.A.D., Johansen, A.M.: Quantifying the uncertainty in change points. J. Time Ser. Anal. 33(5), 807–823 (2012)
Yeh, C.-C.M., et al.: Matrix profile I: all pairs similarity joins for time series: a unifying view that includes motifs, discords and shapelets. In: 2016 IEEE 16th International Conference on data mining (ICDM), pp. 1317–1322. IEEE (2016)
Lajugie, R., Bach, F., Arlot, S.: Large-margin metric learning for constrained partitioning problems. In: International Conference on Machine Learning, pp. 297–305. PMLR (2014)
Harchaoui, Z., Lévy-Leduc, C.: Multiple change-point estimation with a total variation penalty. J. Am. Stat. Assoc. 105(492), 1480–1493 (2010)
Safaric, S., Malaric, K.: Zigbee wireless standard. In: Proceedings ELMAR 2006, pp. 259–262. IEEE (2006)
Van Benschoten, A., Ouyang, A., Bischoff, F., Marrs, T.: MPA: a novel cross-language API for time series analysis. J. Open Source Softw. 5(49), 2179 (2020)
Madani, P., Vlajic, N.: Robustness of deep autoencoder in intrusion detection under adversarial contamination. In: Proceedings of the 5th Annual Symposium and Bootcamp on Hot Topics in the Science of Security, pp. 1–8 (2018)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix A
Appendix A
Segmentation visualization of algorithms reported in Table 1 (Figs. 5, 6, 7, 8, 9 and 10).
Binary segmentation with different cost functions.
Windowing segmentation with different cost functions.
Dynamic programming segmentation with different cost functions.
Bottom-up segmentation with different cost functions.
Matrix profile segmentation with different cost functions.
Kernel-based segmentation with different cost functions.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Madani, P., Vlajic, N. (2022). Effective Segmentation of RSSI Timeseries Produced by Stationary IoT Nodes: Comparative Study. In: Li, W., Furnell, S., Meng, W. (eds) Attacks and Defenses for the Internet-of-Things. ADIoT 2022. Lecture Notes in Computer Science, vol 13745. Springer, Cham. https://doi.org/10.1007/978-3-031-21311-3_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-21311-3_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-21310-6
Online ISBN: 978-3-031-21311-3
eBook Packages: Computer ScienceComputer Science (R0)