Artificial Immune System Approach to Intrusion Detection in IOT Sensor Networks

Intrusion detection in Internet of Things (IoT) sensor networks is imperative due to their susceptibility to security breaches and data integrity violations. The Artificial Immune System (AIS), inspired by biological immune mechanisms, presents a promising paradigm for anomaly detection in such dynamic environments. Nonetheless, current intrusion detection systems frequently exhibit suboptimal accuracy, elevated false alarm rates, and insufficient adaptability to evolving threats within heterogeneous IoT infrastructures. To address these limitations, a Hybrid Machine Learning Ensemble (HMLE) framework is proposed, integrating AIS with ensemble learning algorithms such as Support Vector Machine (SVM). This methodology enhances classification robustness and improves generalization capabilities to identify both known and novel attack vectors. The proposed HMLE approach employs multi-tiered feature selection and adaptive learning mechanisms to continuously monitor and mitigate security threats in real-time. The method achieves improved detection accuracy of 98.9%, minimal false positive rate of 3.2%, recall of 96.2%, and precision of 96.8%. The system demonstrates robust adaptability and reliability suitable for deployment in resource-constrained IoT environments.

Ҳали таржима қилинмаган