An enhanced detection system against routing attacks in mobile ad-hoc network

Mobile ad-hoc network is a dynamic wireless network that transfers information through neighbor nodes with a temporary configuration. Due to its dynamic nature, it is exposed to attacks and intrusions. Routing disruption attack is the main problem of this network where intermediate nodes act maliciously. An encryption-based security mechanism is a first-line defense system that is efficient. It is still not compatible with the mobile ad-hoc network environment. Malicious nodes can drop encrypted data packets in this network. The lightweight technique analyzes a few parameters that consume few resources and provide comparatively low detection rates. However, an intrusion detection system is a reliable second-line security mechanism. In this paper, we have proposed a detection method that classifies malicious and benign information. The proposed intrusion detection method is based on learning techniques that initially require a dataset to determine mobile nodes' behavior. Subsequently, we perform this work in an order such as mobile ad-hoc network simulation with some malicious nodes, features selection, and data collection using packet captured files. This work is executed through extensive simulations in the NS-3. The proposed method learns the system for information classification, and experimental results that show the proposed method performs better than existing schemes. Moreover, the obtained performance confirms that the suggested feature set is suitable for the intrusion detection system in mobile ad-hoc networks.

Automated summary

Mobile ad-hoc network is vulnerable to attacks, and encryption-based security mechanism is not compatible. Lightweight technique and intrusion detection system serve as reliable security mechanisms. This paper proposes an intrusion detection method based on learning techniques, which outperforms existing schemes in extensive simulations.