Intrusion detection systems for IoT based on bio-inspired and machine learning techniques: a systematic review of the literature

Recent technological advancements have significantly expanded both networks and data, thereby introducing new forms of attacks that pose considerable challenges to intrusion detection and network security. With intruders deploying increasingly diverse attack vectors, the need for robust Intrusion Detection Systems (IDSes) has become paramount. IDS serves as a crucial tool for monitoring network traffic to uphold the integrity, confidentiality, and availability of systems. Despite the integration of Machine Learning (ML) and Deep Learning (DL) algorithms into IDS frameworks, achieving higher accuracy levels while minimizing false alarms remains a challenging task, especially when handling large datasets. In response to this challenge, researchers have turned to bio-inspired algorithms as potential solutions to enhance IDS models. This paper undertakes a comprehensive literature review focusing on augmenting the security of Internet of Things (IoT) networks by integrating bio-inspired methodologies with ML and DL techniques. Among 145 published articles, 25 relevant studies were selected to address the defined research objectives. The findings underscore the efficacy of combining bio-inspired techniques with ML and DL approaches in enhancing IDS performance, highlighting their potential to bolster IoT network security. Furthermore, the review incorporates a comparative analysis of the selected articles, considering various factors, and outlines ongoing challenges and future directions in integrating bio-inspired techniques with ML and DL algorithms.