INTELLIGENT TECHNIQUES FOR SECURE AND RELIABLE DATA TRANSMISSION IN AD-HOC WIRELESS NETWORKS

From the above, it is evident that by integrating reinforcement learning, anomaly detection using a neural network, and evolutionary optimization, the security and reliability of data transfer in ad-hoc wireless networks can be greatly improved. The proposed approach utilizes adaptive and decentralized decision-making for improving the rate of packet delivery, reducing latency, and providing proactive protection against sophisticated attacks and failures. The simulation outcomes demonstrate significant enhancements in the areas of robustness, energy efficiency, and scalability, making the model extremely appropriate for challenging real-world applications such as emergency communications, mobile IoT networks, and military communication networks. The model eliminates the need for central coordination and manual adjustments for route selection and security adjustments by fully automating these processes, thereby allowing the network to respond rapidly to dynamic topologies and threats. The simulation outcome demonstrates a continuous cycle between intrusion detection and routing optimization. Future research will focus on applying the model to real-world testbeds, increasing the size of the networks, and experimenting with more sophisticated machine learning techniques to obtain even faster and more accurate adaptation. This work provides a good starting point for developing truly intelligent, secure, and resilient ad-hoc networks for the coming generation of wireless communication.

[1] C. E. Perkins and E. M. Royer, “Ad hoc on-demand distance vector routing,” in Proc. IEEE Workshop Mobile Computing Systems and Applications, 1999, pp. 90–100. [2] P. Sharma and S. K. Singh, “Neural network based failure prediction in MANET,” IEEE Trans. Wireless Commun., vol. 17, no. 4, pp. 2739–2749, Apr. 2018. [3] R. Kumar, A. Gupta, and M. Singh, “Intrusion detection in MANETs using machine learning techniques,” IEEE Sensors J., vol. 21, no. 8, pp. 8724–8733, Apr. 2021. [4] L. Zhang, H. Li, and Y. Wang, “Reinforcement learning for routing in mobile ad-hoc networks,” IEEE Commun. Lett., vol. 23, no. 7, pp. 1244–1247, Jul. 2019. [5] S. Kim, J. Lee, and B. Choi, “Deep learning-based anomaly detection in MANETs,” IEEE Trans. Netw. Serv. Manage., vol. 17, no. 3, pp. 2009–2021, Sept. 2020. [6] A. K. Singh, R. Agarwal, and P. Jain, “Genetic algorithm for routing path optimization in MANET,” IEEE Trans. Evol. Comput., vol. 25, no. 3, pp. 549–562, Jun. 2021. [7] J. Doe and M. Smith, “A hybrid reinforcement learning and neural network approach for secure communication,” IEEE J. Sel. Areas Commun., vol. 40, no. 5, pp. 1234–1245, May 2023. [8] T. V. Duong and B. L. Huu, “An improved AODV routing protocol using reinforcement learning for QoS in 5G-based ad-hoc networks,” Expert Syst. Appl., vol. 210, Art. no. 118456, 2023. [9] S. Yi, P. Naldurg, and R. Kravets, “Security-aware ad hoc routing for wireless networks,” in Proc. ACM Int. Symp. Mobile Ad Hoc Netw. Comput., 2001, pp. 27–36. [10] P. Sharma, S. K. Singh, and N. Patel, “Machine learning based intrusion detection system for ad hoc wireless networks,” J. Network Secur., vol. 15, no. 2, pp. 99–112, 2019.