Mobile Ad hoc Networks (MANETs) are decentralized, self-organizing networks that offer flexibility and scalability for a variety of applications. However, their dynamic topology and limited resources make them highly susceptible to security threats. Traditional security measures often struggle to address the complexities of evolving and sophisticated attacks within such networks. This paper reviews the role of machine learning (ML) techniques in enhancing anomaly detection capabilities in MANETs, focusing on their applications in intrusion detection, attack prediction, and routing security. Decentralized, self- organizing, infrastructure-less wireless networks known as mobile ad hoc networks (MANETs) allow nodes to speak with one another directly without the use of fixed base stations. Because of these features, MANETs are extremely adaptable and appropriate for a wide range of applications, including remote sensing, smart transportation systems, military and disaster recovery activities, and more. Despite these benefits, MANETs are vulnerable to a variety of security risks because to their changeable topology, limiting bandwidth, and limited computational resources. Traditional security techniques, such rule-based intrusion detection systems and cryptographic processes, frequently find it difficult to handle the dynamic and adaptive character of contemporary threats. Intelligent and flexible technologies that can offer strong security in real time are therefore desperately needed. By leveraging supervised, unsupervised, and reinforcement learning methods, ML models can identify abnormal behaviours, predict potential attacks, and secure routing protocols in real time. The integration of ML offers a powerful means to mitigate risks associated with attacks such as Black A potent paradigm for enhancing MANET security is machine learning (ML), especially in the domains of anomaly detection and intrusion prevention. It is feasible to identify unusual network activities, anticipate probable attacks before they become more serious, and protect routing protocols against malevolent disruptions by taking advantage of machine learning algorithms' capacity to learn patterns from data. Support Vector Machines (SVM) Hole, Sybil, Wormhole, and Denial of Service (DoS). This review critically examines the performance of various ML-based anomaly detection techniques, including SVM, Auto-encoders, K-Means clustering, and Isolation Forests, and discusses the challenges of deploying these methods in resource-constrained environments. It also highlights emerging solutions like federated learning and model compression, which address scalability and computational issues. The paper concludes with a discussion on future research directions, including the potential of explainable AI (XAI) and hybrid approaches for improving the security resilience of MANETs.

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. Margala M, editor. Advances in intelligent systems: paradigms and applications/edited by Manisha Guduri, PhD, Uma Maheswari V. Biosensors and Bioelectronics. 2019;150:111935.
2. Al Ali IA, Alhaidery MM. Machine Learning Techniques for Anomaly Detection in IoT and WSN: A review. Journal of Al- Qadisiyah for Computer Science and Mathematics. 2025 Jun 30;17(2):229-40.
3. Rawat M, Singal G. Surveying Technology Fusion in IoT Networks for IDS: Exploring Datasets, Tools, Challenges, and Research Prospects. ACM Transactions on Intelligent Systems and Technology. 2025.
4. Wang C, Yuan Z, Zhou P, Xu Z, Li R, Wu DO. The security and privacy of mobile-edge computing: An artificial intelligence perspective. IEEE Internet of Things Journal. 2023 Aug 11;10(24):22008-32.
5. Lu G, Feng D, Huang B. Hidden Markov model-based attack detection for networked control systems subject to random packet dropouts. IEEE Transactions on Industrial Electronics. 2020 Jan 17;68(1):642-53.
6. Tiwari A, Darbari M. Emerging Trends in Computer Science and Its Application. Boca Raton, FL, USA: CRC Press; 2025 Apr 8.
7. Arifin MM, Ahmed MS, Ghosh TK, Udoy IA, Zhuang J, Yeh JH. A survey on the application of generative adversarial networks in cybersecurity: Prospective, direction and open research scopes. arXiv preprint arXiv:2407.08839. 2024 Jul 11.
8. Masud MT, Keshk M, Moustafa N, Linkov I, Emge DK. Explainable artificial intelligence for resilient security applications in the Internet of Things. IEEE Open Journal of the Communications Society. 2024 Jun 13;6:2877-906.
9. Qazi EU, Faheem MH, Zia T. HDLNIDS: hybrid deep-learning-based network intrusion detection system. Applied Sciences. 2023 Apr 14;13(8):4921.
10. Khan MU, Azizi M, García-Armada A, Escudero-Garzás JJ. Unsupervised clustering for 5G network planning assisted by real data. IEEE Access. 2022 Apr 8;10:39269-81.
11. Ibrahim ZB, Ghanim MF. Leveraging artificial intelligence for blackhole attack detection in MANETs: A comparative study. Inf. Dyn. Appl. 2024;3(4):245-57.
12. Meddeb R, Jemili F, Triki B, Korbaa O. A deep learning-based intrusion detection approach for mobile Ad-hoc network. Soft Computing. 2023 Jul;27(14):9425-39.

Mobile Ad hoc Networks (MANETs) are decentralized, self-organizing networks that offer flexibility and scalability for a variety of applications. However, their dynamic topology and limited resources make them highly susceptible to security threats. Traditional security measures often struggle to address the complexities of evolving and sophisticated attacks within such networks. This paper reviews the role of machine learning (ML) techniques in enhancing anomaly detection capabilities in MANETs, focusing on their applications in intrusion detection, attack prediction, and routing security. Decentralized, self- organizing, infrastructure-less wireless networks known as mobile ad hoc networks (MANETs) allow nodes to speak with one another directly without the use of fixed base stations. Because of these features, MANETs are extremely adaptable and appropriate for a wide range of applications, including remote sensing, smart transportation systems, military and disaster recovery activities, and more. Despite these benefits, MANETs are vulnerable to a variety of security risks because to their changeable topology, limiting bandwidth, and limited computational resources. Traditional security techniques, such rule-based intrusion detection systems and cryptographic processes, frequently find it difficult to handle the dynamic and adaptive character of contemporary threats. Intelligent and flexible technologies that can offer strong security in real time are therefore desperately needed. By leveraging supervised, unsupervised, and reinforcement learning methods, ML models can identify abnormal behaviours, predict potential attacks, and secure routing protocols in real time. The integration of ML offers a powerful means to mitigate risks associated with attacks such as Black A potent paradigm for enhancing MANET security is machine learning (ML), especially in the domains of anomaly detection and intrusion prevention. It is feasible to identify unusual network activities, anticipate probable attacks before they become more serious, and protect routing protocols against malevolent disruptions by taking advantage of machine learning algorithms' capacity to learn patterns from data. Support Vector Machines (SVM) Hole, Sybil, Wormhole, and Denial of Service (DoS). This review critically examines the performance of various ML-based anomaly detection techniques, including SVM, Auto-encoders, K-Means clustering, and Isolation Forests, and discusses the challenges of deploying these methods in resource-constrained environments. It also highlights emerging solutions like federated learning and model compression, which address scalability and computational issues. The paper concludes with a discussion on future research directions, including the potential of explainable AI (XAI) and hybrid approaches for improving the security resilience of MANETs.

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. Margala M, editor. Advances in intelligent systems: paradigms and applications/edited by Manisha Guduri, PhD, Uma Maheswari V. Biosensors and Bioelectronics. 2019;150:111935.
2. Al Ali IA, Alhaidery MM. Machine Learning Techniques for Anomaly Detection in IoT and WSN: A review. Journal of Al- Qadisiyah for Computer Science and Mathematics. 2025 Jun 30;17(2):229-40.
3. Rawat M, Singal G. Surveying Technology Fusion in IoT Networks for IDS: Exploring Datasets, Tools, Challenges, and Research Prospects. ACM Transactions on Intelligent Systems and Technology. 2025.
4. Wang C, Yuan Z, Zhou P, Xu Z, Li R, Wu DO. The security and privacy of mobile-edge computing: An artificial intelligence perspective. IEEE Internet of Things Journal. 2023 Aug 11;10(24):22008-32.
5. Lu G, Feng D, Huang B. Hidden Markov model-based attack detection for networked control systems subject to random packet dropouts. IEEE Transactions on Industrial Electronics. 2020 Jan 17;68(1):642-53.
6. Tiwari A, Darbari M. Emerging Trends in Computer Science and Its Application. Boca Raton, FL, USA: CRC Press; 2025 Apr 8.
7. Arifin MM, Ahmed MS, Ghosh TK, Udoy IA, Zhuang J, Yeh JH. A survey on the application of generative adversarial networks in cybersecurity: Prospective, direction and open research scopes. arXiv preprint arXiv:2407.08839. 2024 Jul 11.
8. Masud MT, Keshk M, Moustafa N, Linkov I, Emge DK. Explainable artificial intelligence for resilient security applications in the Internet of Things. IEEE Open Journal of the Communications Society. 2024 Jun 13;6:2877-906.
9. Qazi EU, Faheem MH, Zia T. HDLNIDS: hybrid deep-learning-based network intrusion detection system. Applied Sciences. 2023 Apr 14;13(8):4921.
10. Khan MU, Azizi M, García-Armada A, Escudero-Garzás JJ. Unsupervised clustering for 5G network planning assisted by real data. IEEE Access. 2022 Apr 8;10:39269-81.
11. Ibrahim ZB, Ghanim MF. Leveraging artificial intelligence for blackhole attack detection in MANETs: A comparative study. Inf. Dyn. Appl. 2024;3(4):245-57.
12. Meddeb R, Jemili F, Triki B, Korbaa O. A deep learning-based intrusion detection approach for mobile Ad-hoc network. Soft Computing. 2023 Jul;27(14):9425-39.