Enhancing Security in Wireless Mesh Networks: A Deep Learning Approach to Black Hole Attack Detection

Mansi Bhonsle; Gunji Sreenivasulu; Kilaru Chaitanya; Dhumpati Raghu; Gunti Surendra; Konduru Kranthi Kumar; Mandalapu Srinivasa Rao; Kandukuri Prabhakar; Vamsi Krishna Vuppu

doi:10.26877/asset.v7i1.1036

Authors

Mansi Bhonsle MIT Art, Design and Technology University
Gunji Sreenivasulu Madanapalle Institute of Technology & Science
Kilaru Chaitanya SRK Institute of Technology
Dhumpati Raghu Bahrain Polytechnic
Gunti Surendra K L Deemed to be University
Konduru Kranthi Kumar Vasireddy Venkatadri Institute of Technology
Mandalapu Srinivasa Rao R.V.R. & J.C.College of Engineering
Kandukuri Prabhakar Chaitanya Bharathi Institute of Technology
Vamsi Krishna Vuppu Amrita School of Computing

DOI:

https://doi.org/10.26877/asset.v7i1.1036

Keywords:

Deep learning, defense mechanisms, black hole attacks, wireless mesh networks, security, attack mitigation, Cybersecurit, anomaly detection, Network intrusion

Abstract

Wireless Mesh Networks (WMNs) are susceptible to various security threats, including black hole attacks, where malicious nodes attract and drop packets, disrupting network communication. Traditional security mechanisms are often inadequate in detecting and mitigating these attacks due to their dynamic and evolving nature. In this paper, we propose a novel deep learning-based defense mechanism against black hole attacks in WMNs. It utilizes Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) networks to analyze network traffic patterns and detect abnormal behavior indicative of black hole attacks. The proposed approach offers several advantages, including the ability to adapt to new attack patterns and achieve high detection accuracy. The evaluations of this method using an NSL KDD demonstrate its effectiveness in mitigating black hole attacks. Results indicate a significant improvement in attack detection rates compared to traditional rule-based systems, reducing both false positives and the overall impact of such attacks on network performance. The proposed solution not only strengthens WMN security but also has the potential to adapt to evolving attack strategies through continuous learning. This research paves the way for future advancements in adversarial learning and autonomous, self-healing security systems for mesh networks. It offers scalable solutions to secure critical infrastructure like smart cities and IoT ecosystems, ensuring reliable communication. Integrating Deep Learning Algorithms security in WMNs enhances resilience against evolving cyber threats in next-generation wireless networks.

Author Biographies

Mansi Bhonsle, MIT Art, Design and Technology University

Department of CSE, MIT School of Computing, MIT Art, Design and Technology University, Pune, Loni Kalobhor, Pune. India.

Gunji Sreenivasulu, Madanapalle Institute of Technology & Science

Department of CSE, Madanapalle Institute of Technology & Science, Madanapalle. India.

Kilaru Chaitanya, SRK Institute of Technology

Department of CSE, SRK Institute of Technology, Vijayawada. India.

Dhumpati Raghu, Bahrain Polytechnic

Department of CSE, Bahrain Polytechnic, PO Box 3339, Isa Town, Bahrain.

Gunti Surendra, K L Deemed to be University

Department of CSE, K L Deemed to be University, Green Fields, Vaddswaram, AP. India.

Konduru Kranthi Kumar, Vasireddy Venkatadri Institute of Technology

Department of Information Technology, VVIT, Nambur. India

Mandalapu Srinivasa Rao, R.V.R. & J.C.College of Engineering

CSBS Department, RVR&JC, Chowdavaram, Guntur. India.

Kandukuri Prabhakar, Chaitanya Bharathi Institute of Technology

Department, Chaitanya Bharathi Institute of Technology, Osman Sagar Rd, Kokapet, Gandipet, Hyderabad. India.

Vamsi Krishna Vuppu, Amrita School of Computing

Department of CSE, Amrita School of Computing, Amrita Vishwa Vidyapeetham, Chennai, Tamil Nadu, India.

References

[1] Joon, D., & Chopra, K. (2021). Hybrid deep learning prediction model for blackhole attack protection in wireless communication. NVEO-NATURAL VOLATILES & ESSENTIAL OILS Journal| NVEO, 10228-10243.

[2] Ibrahim, Z. B., & Ghanim, M. F. (2024). A Review of AI-Based Approaches against Wormhole and Blackhole Attacks in AODV Protocol.

[3] Sharma, D. K., Dhurandher, S. K., Kumaram, S., Gupta, K. D., & Sharma, P. K. (2022). Mitigation of black hole attacks in 6LoWPAN RPL-based Wireless sensor network for cyber physical systems. Computer Communications, 189, 182-192.

[4] Khan, S., Khan, M. A., & Alnazzawi, N. (2024). Artificial neural network-based mechanism to detect security threats in wireless sensor networks. Sensors, 24(5), 1641.

[5] Saleh, H. M., Marouane, H., & Fakhfakh, A. (2024). A Comprehensive Analysis of Security Challenges and Countermeasures in Wireless Sensor Networks Enhanced by Machine Learning and Deep Learning Technologies. International Journal of Safety & Security Engineering, 14(2).

[6] Meddeb, R., Jemili, F., Triki, B., & Korbaa, O. (2023). A deep learning-based intrusion detection approach for mobile Ad-hoc network. Soft Computing, 27(14), 9425-9439.

[7] SMAILI, A., & KACHOUR, I. E. (2021). Analysis and detection of routing attacks in the internet of Things using Deep learning (Doctoral dissertation, Université Ibn Khaldoun-Tiaret-).

[8] Cakir, S., Toklu, S., & Yalcin, N. (2020). RPL attack detection and prevention in the Internet of Things networks using a GRU based deep learning. IEEE Access, 8, 183678-183689.

[9] Sagayam, K. M., Bhushan, B., Andrushia, A. D., & Albuquerque, V. H. C. D. (Eds.). (2020). Deep Learning Strategies for Security Enhancement in Wireless Sensor Networks. IGI Global.

[10] Regan, R., & Manickam, J. M. L. (2017). Detecting and denying malicious behavior using adaptive learning based routing protocols in wireless mesh network. Appl. Math, 11(4), 1155-1162.

[11] Ardyani, S. S. F., & Sari, C. A. (2024). A Web-Based for Demak Batik Classification Using VGG16 Convolutional Neural Network. Advance Sustainable Science Engineering and Technology, 6(4), 0240406-0240406.

[12] Reji, M., Joseph, C., Thaiyalnayaki, K., & Lathamanju, R. (2023). Genetic-based Fuzzy IDS for Feature Set Reduction and Worm Hole Attack Detection. Computer Systems Science & Engineering, 45(2).

[13] Pawar, M. V. (2023). Detection and prevention of black-hole and wormhole attacks in wireless sensor network using optimized LSTM. International Journal of Pervasive Computing and Communications, 19(1), 124-153.

[14] Hussain, K., Xia, Y., Onaizah, A. N., Manzoor, T., & Jalil, K. (2022). Hybrid of WOA-ABC and proposed CNN for intrusion detection system in wireless sensor networks. Optik, 271, 170145.

[15] Pawar, M. V. (2023). Detection and prevention of black-hole and wormhole attacks in wireless sensor network using optimized LSTM. International Journal of Pervasive Computing and Communications, 19(1), 124-153.

[16] Karunaratne, S., & Gacanin, H. (2019). An overview of machine learning approaches in wireless mesh networks. IEEE Communications Magazine, 57(4), 102-108.

[17] Qazi, E. U. H., Faheem, M. H., & Zia, T. (2023). HDLNIDS: hybrid deep-learning-based network intrusion detection system. Applied Sciences, 13(8), 4921.

[18] Smys, S., & Haoxiang, W. (2021). A secure optimization algorithm for quality-of-service improvement in hybrid wireless networks. IRO Journal on Sustainable Wireless Systems, 3(1), 1-10.

[19] Nivaashini, M., Thangaraj, P., Sountharrajan, S., Suganya, E., & Soundariya, R. S. (2021). Effective Feature Selection for Hybrid Wireless IoT Network Intrusion Detection Systems Using Machine Learning Techniques. Ad Hoc Sens. Wirel. Networks, 49(3-4), 175-206.

[20] Mahajan, S., HariKrishnan, R., & Kotecha, K. (2022). Prediction of network traffic in wireless mesh networks using hybrid deep learning model. IEEE Access, 10, 7003-7015.