Strategic Approaches to Intrusion Detection in Cloud-IoT Ecosystem

13Designing a Hybrid Intrusion Detection System for Wireless Acoustic Sensor Networks: Enhancing Security During Audio Transmission

Utpal Ghosh¹ and Uttam Kr. Mondal^2*

¹Dept. of Computer Science, Sarojini Naidu College for Women, Kolkata, WB, India

²Dept. of Computer Science, Vidyasagar University, Midnapore, WB, India

Abstract

Wireless Acoustic Sensor Networks (WASNs) are extensively used in environmental monitoring, security surveillance, and smart homes, but their wireless nature exposes them to various security risks, especially during audio transmission. To enhance the security of WASNs, this article proposes a hybrid intrusion detection system (HIDS) that integrates anomaly, machine learning and signature-based detection techniques. The signature-based module utilizes a database of known attack signatures to quickly identify and neutralize threats, while the anomaly-based module employs machine learning methods like support vector machines (SVM) and neural networks to model normal audio transmission behavior and detect deviations indicative of novel or zero-day attacks. The efficacy of the proposed hybrid IDS was assessed through detailed models and real-world implementations against three attack scenarios: node compromise, eavesdropping, and packet tampering. Performance metrics such as detection accuracy, false positive rate, detection time and resource utilization (memory, central processing unit (CPU), network) were evaluated. The results showed a high detection ...