Fuzzy Decision-Making Framework for Explainable Golden Multi-Machine Learning Models for Real-Time Adversarial Attack Detection in Vehicular Ad-hoc Networks

  1. Albahri, A.S.
  2. Hamid, Rula A.
  3. Abdulnabi, Ahmed Raheem
  4. Albahri, O.S.
  5. Alamoodi, A.H.
  6. Deveci, Muhammet
  7. Pedrycz, Witold
  8. Alzubaidi, Laith
  9. Santamaría, Jose
  10. Gu, Yuantong
Aldizkaria:
Information Fusion

ISSN: 1566-2535

Argitalpen urtea: 2024

Orrialdeak: 102208

Mota: Artikulua

DOI: 10.1016/J.INFFUS.2023.102208 GOOGLE SCHOLAR lock_openSarbide irekia editor

Beste argitalpen batzuk: Information Fusion

Garapen Iraunkorreko Helburuak

Laburpena

his paper addresses various issues in the literature concerning adversarial attack detection in Vehicular Ad-hoc Networks (VANETs). These issues include the failure to consider both normal and adversarial attack perspectives simultaneously in Machine Learning (ML) model development, the lack of diversity preprocessing techniques for VANETs communication datasets, the inadequate selection guidelines for real-time adversarial attack detection models, and the limited emphasis on explainability in adversarial attack detection. In this study, we propose an original fuzzy decision-making framework that incorporates multiple fusion standpoints. Our framework aims to evaluate multi-ML models for real-time adversarial attack detection in VANETs, focusing on three stages. The first stage involves identifying and preprocessing Dedicated Short-Range Communication (DSRC) data using standard and fusion preprocessing approaches. Two communication scenarios, normal and jammed, are considered, resulting in two DSRC datasets. In the second stage, we develop multi-ML models based on the DSRC datasets using standard preprocessing and feature fusion preprocessing for dataset-1 and dataset-2, respectively. The third stage evaluates the multi-ML models using a fuzzy decision-making approach based on the Fuzzy Decision by Opinion Score Method (FDOSM) and an adversarial attack decision fusion matrix. The External Fusion Decision (EFD) settings of the FDOSM address individual ranking variance, provide a unique rank and select the best model. Experimental results demonstrate that the K-Nearest Neighbors Algorithm (kNN) model achieves the highest explain score of 0.2048 in dataset-1 using standard preprocessing, while the Random Forest (RF) model applied to dataset-2 using fusion preprocessing emerges as the most robust and golden model against adversarial attacks, with a score of 0.1819. This finding suggests that the fusion preprocessing approach using Principal Component Analysis (PCA) is more suitable for addressing normal and adversarial attack perspectives. Furthermore, our fuzzy framework undergoes evaluation in terms of systematic rank, sensitivity analysis, explainability analysis, and comparison analysis. Overall, this framework provides valuable insights for researchers and practitioners in VANETs, informing the execution, selection, and interpretation of multi-ML models to tackle adversarial attack detection problems effectively. The new fuzzy framework demonstrates that multi-ML models based on feature fusion preprocessing are more effective.

€ Ikusi finantzaketa

Finantzaketari buruzko informazioa

Finantzatzaile

Erreferentzia bibliografikoak

  • Cheng, (2017), IEEE Intell. Syst., 32, pp. 53, 10.1109/MIS.2017.53
  • Araniti, (2013), IEEE Commun. Mag., 51, pp. 148, 10.1109/MCOM.2013.6515060
  • Liang, (2017), IEEE Trans. Veh. Technol., 66, pp. 10647, 10.1109/TVT.2017.2750903
  • Peng, (2019), IEEE Trans. Veh. Technol., 68, pp. 1064, 10.1109/TVT.2018.2833427
  • Zhang, (2015), IEEE Trans. Intell. Transp. Syst., 16, pp. 411, 10.1109/TITS.2014.2335746
  • Cheng, (2015), IEEE Trans. Intell. Transp. Syst., 16, pp. 1784, 10.1109/TITS.2014.2377074
  • Liang, (2019), IEEE Internet Things J, 6, pp. 124, 10.1109/JIOT.2018.2872122
  • Verma, (2013), Wirel. Pers. Commun., 73, pp. 95, 10.1007/s11277-013-1161-5
  • Jain, (2018), 2018-Janua, pp. 1957
  • I. Standard, “INTERNATIONAL STANDARD ISO /IEC / IEEE Telecommunications and information,” vol. 2012, 2012.
  • Kim, (2022), Comput. Mater. Contin., 71, pp. 3589
  • M. R. Ghori, A. S. Sadiq, and A. Ghani, “VANET Routing Protocols : Review, Implementation and Analysis VANET Routing Protocols : Review, Implementation and Analysis,” 2018.
  • Vijayakumar, (2016), IEEE Trans. Intell. Transp. Syst., 17, pp. 1015, 10.1109/TITS.2015.2492981
  • Ounoughi, (2023), Inf. Fusion, 89, pp. 267, 10.1016/j.inffus.2022.08.016
  • Khatri, (2021), Peer-to-Peer Netw. Appl, 14, pp. 1778, 10.1007/s12083-020-00993-4
  • Jordan, (2015), Science (80-.), 349, pp. 255, 10.1126/science.aaa8415
  • Tuna, (2022), Multimed. Tools Appl., 81, pp. 11479, 10.1007/s11042-022-12132-7
  • Deldjoo, (2021), ACM Comput. Surv., 54, pp. 1, 10.1145/3439729
  • Yang, (2021), Appl. Soft Comput., 102, 10.1016/j.asoc.2021.107096
  • Adesina, (2023), IEEE Commun. Surv. Tutorials, 25, pp. 77, 10.1109/COMST.2022.3205184
  • Hasbullah, (2010), World Acad. Sci. Eng. Technol., 65, pp. 411
  • Pekár, (2017), Elife, 6, pp. e22089, 10.7554/eLife.22089
  • Gautam, (2022), IEEE Access, 10, pp. 90369, 10.1109/ACCESS.2022.3200149
  • D. Kosmanos, D. Karagiannis, A. Argyriou, S. Lalis, and L. Maglaras, “RF Jamming Classification using Relative Speed Estimation in Vehicular Wireless Networks,” pp. 1–10, 2019.
  • Sharma, (2019), pp. 1
  • Ji, (2021), pp. 520
  • Tian, (2022), IEEE Internet Things J, 9, pp. 22399, 10.1109/JIOT.2021.3111024
  • Hassan, (2022), Wasit J. Comput. Math. Sci., 1
  • Okolie, (2022), ISPRS J. Photogramm. Remote Sens., 188, pp. 1, 10.1016/j.isprsjprs.2022.03.016
  • Puttagunta, (2023), Multimed. Tools Appl., pp. 1
  • Trivedi, (2020), Technol. Soc., 63, 10.1016/j.techsoc.2020.101413
  • Albahri, (2019), IEEE Access, 7, pp. 50052, 10.1109/ACCESS.2019.2910411
  • Al-Safwani, (2014), J. Appl. Sci., 14, 10.3923/jas.2014.1865.1870
  • Mahmoud, (2022), J. Ambient Intell. Humaniz. Comput.
  • Alqaysi, (2022), Comput. Math. Methods Med., 2022, 10.1155/2022/9410222
  • Ahmed, (2023), Int. J. Inf. Technol. Decis. Mak., 10.1142/S0219622023500463
  • Al-Samarraay, (2022), Appl. Soft Comput., 116, 10.1016/j.asoc.2021.108284
  • Akram, (2022), Math. Probl. Eng., 2022, 10.1155/2022/1501880
  • Alsalem, (2021), J. Infect. Public Health, 14, pp. 1513, 10.1016/j.jiph.2021.08.026
  • Alamoodi, (2022), Complex Intell. Syst., 8, pp. 3479, 10.1007/s40747-022-00689-7
  • Al-Qaysi, (2023), Phys. Eng. Sci. Med., pp. 1
  • Barredo Arrieta, (2020), Inf. Fusion, 58, pp. 82, 10.1016/j.inffus.2019.12.012
  • Ribeiro, (2016), pp. 97
  • Payrovnaziri, (2020), J. Am. Med. Informatics Assoc., 27, pp. 1173, 10.1093/jamia/ocaa053
  • Malebary, (2017), pp. 1
  • Chen, (2023), Inf. Fusion, 92, pp. 205, 10.1016/j.inffus.2022.11.021
  • Liu, (2023), Inf. Fusion, 91, pp. 205, 10.1016/j.inffus.2022.09.030
  • Patro, (2015), arXiv Prepr. arXiv1503.06462
  • Jain, (2018), Expert Syst. Appl., 106, pp. 252, 10.1016/j.eswa.2018.04.008
  • Dey, (2019), Inf. Fusion, 49, pp. 205, 10.1016/j.inffus.2019.01.002
  • Shlens, (2014), arXiv Prepr. arXiv1404.1100
  • Jolliffe, (2016), Philos. Trans. R. Soc. A Math. Phys. Eng. Sci., 374
  • Ansari, (2018), Int. Commun. Heat Mass Transf., 91, pp. 158, 10.1016/j.icheatmasstransfer.2017.12.012
  • Uddin, (2019), BMC Med. Inform. Decis. Mak., 19, pp. 281, 10.1186/s12911-019-1004-8
  • Schrider, (2018), Trends Genet, 34, pp. 301, 10.1016/j.tig.2017.12.005
  • “Getting started with machine-learning.”
  • SMITH, (1993), Teach. Stat., 15, pp. 31, 10.1111/j.1467-9639.1993.tb00258.x
  • Nakata, (2017), pp. 473
  • Atkinson, (2016), Expert Syst. Appl., 47, pp. 35, 10.1016/j.eswa.2015.10.049
  • Bokulich, (2018), Microbiome, 6, pp. 1, 10.1186/s40168-018-0470-z
  • Tantithamthavorn, (2016), 14-22-May, pp. 321
  • Shafahi, (2019), Adv. Neural Inf. Process. Syst., 32
  • Li, (2021), pp. 2022
  • Peng, (2021), Comput. Secur., 107, 10.1016/j.cose.2021.102317
  • Massoli, (2021), Comput. Vis. Image Underst., 202, 10.1016/j.cviu.2020.103103
  • Liu, (2022), IEEE Access, 10, pp. 50974, 10.1109/ACCESS.2022.3171659
  • Zhang, (2023), Inf. Sci. (Ny)., 633, pp. 491, 10.1016/j.ins.2023.03.084
  • Gao, (2021), pp. 109
  • Salih, (2020), Appl. Soft Comput. J., 96, 10.1016/j.asoc.2020.106595
  • Albahri, (2021), Int. J. Intell. Syst., 36, pp. 796, 10.1002/int.22322
  • Albahri, (2023), Comput. Intell., 39, pp. 225, 10.1111/coin.12562
  • Wu, (2021), pp. 11240
  • Qader, (2017), Meas. J. Int. Meas. Confed., 111, pp. 38, 10.1016/j.measurement.2017.07.024
  • Kalid, (2018), J. Med. Syst., 42, pp. 69, 10.1007/s10916-018-0916-7
  • Alamoodi, (2023), Neural Comput. Appl., 35, pp. 6185, 10.1007/s00521-022-07998-5
  • Dožić, (2015), Transp. Res. Procedia, 10, pp. 910, 10.1016/j.trpro.2015.09.044
  • Moradi, (2020), Energy Strateg. Rev., 29
  • Yagmahan, (2022), Environ. Dev. Sustain., pp. 1
  • Ribeiro, (2018), 32, pp. 1527
  • Li, (2023), Artif. Intell., 314, 10.1016/j.artint.2022.103823
  • Chan, (2022), BMC Med. Inform. Decis. Mak., 22, pp. 75, 10.1186/s12911-022-01817-6
  • Visani, (2022), J. Oper. Res. Soc., 73, pp. 91, 10.1080/01605682.2020.1865846
  • Guidotti, (2019), IEEE Intell. Syst., 34, pp. 14, 10.1109/MIS.2019.2957223
  • Gabbay, (2021), Appl. Sci., 11, pp. 10417, 10.3390/app112110417