Lightweight and Intelligent DDoS Detection for IoT Networks: A Comprehensive Survey, Comparative Benchmark, and Research Roadmap

Abstract

The integration of Internet of Things (IoT) technologies into critical systems has driven major advancements across industries. However, this integration has also introduced new avenues for security breaches, particularly Distributed Denial-of-Service (DDoS) attacks. These attacks are especially damaging in IoT environments due to the constrained processing power and limited defenses of many devices, which are often deployed without sufficient safeguards. This review sets out to explore and analyze the range of techniques proposed for detecting and mitigating DDoS attacks within IoT frameworks. Specifically, it organizes existing research into four primary categories: traditional detection methods, machine learning based techniques, deep learning models, and hybrid approaches that combine different strategies for enhanced performance. Using a methodology aligned with PRISMA guidelines, the study systematically filtered and reviewed 26 papers. Each paper was assessed according to its detection mechanism, dataset type, performance outcomes, and suitability for real world deployment. The findings highlight the increasing effectiveness of machine learning and deep learning methods. This is particularly evident when they are employed in hybrid configurations that improve detection rates and adaptability. Despite these advancements, several issues persist, such as the lack of standardized datasets and the challenge of deploying models on resource constrained devices. This review highlights these gaps and serves as a guide for researchers seeking to develop more practical and resilient IoT security solutions.