Lightweight Intrusion Detection for SOHO and IoT Networks: A Comprehensive Survey

Abstract

The exponential proliferation of Internet of Things (IoT) devices has fundamentally altered the landscape of Small Office/Home Office (SOHO) networks, introducing a heterogeneous and expanding attack surface that traditional security paradigms struggle to address. As consumer-grade appliances, industrial sensors, and smart infrastructure components become ubiquitous, they bring with them severe resource constraints—limited processing power, memory, and energy—that render conventional, resource-intensive Intrusion Detection Systems (IDS) obsolete. This survey provides an exhaustive critical analysis of the emerging field of lightweight IDS, specifically tailored for these constrained environments. We propose a multidimensional taxonomy categorizing state-of-the-art systems based on architectural deployment, detection methodology, and advanced optimization techniques (TinyML, quantization, structural pruning). We systematically benchmark commonly used IoT IDS datasets and explicitly contrast academic lightweight systems with commercial SOHO solutions. Furthermore, this report bridges the technical-cognitive gap by exploring the integration of Generative AI, Large Language Models (LLMs), and gamification strategies to enhance security awareness among non-expert users, highlighting recent empirical case studies. Through a rigorous comparative analysis of performance metrics, we identify the optimal trade-offs required for securing the modern edge. The survey concludes by outlining significant knowledge gaps, the necessity for statistical validation in benchmarking, and future research directions for robust, user-centric SOHO security.