Paper Type
Complete
Abstract
The seamless connectivity of smart homes introduces new vulnerabilities, particularly to Distributed Denial-of-Service (DDoS) attacks. Artificial intelligence (AI) offers promising capabilities for detecting and mitigating these attacks; however, existing studies primarily focus on general IoT environments rather than resource-constrained smart home settings. This systematic literature review synthesizes peer-reviewed studies on AI-based DDoS detection and mitigation in smart homes. Following PRISMA guidelines, this review analyzes success conditions, classifies AI techniques, and identifies design trade-offs. Results show a strong emphasis on detection, but limited mitigation and no recovery mechanisms. Classical machine learning and hybrid architectures dominate, with the latter primarily deployed at the gateway and edge levels. Key gaps include incomplete protection lifecycle coverage, limited privacy preservation, and lack of explainability mechanisms. Quality assessment shows high methodological rigor, with real testbeds scoring highest. These findings inform the design of resilient, privacy-preserving, and explainable AI-based defenses for smart home security.
Paper Number
1167
Recommended Citation
Niroomand, Hajar and El-Gayar, Omar, "AI for Mitigating DDoS Threat in Smart Homes - A Systematic Review" (2026). AMCIS 2026 Proceedings. 1.
https://aisel.aisnet.org/amcis2026/sig_sec/sig_sec/1
AI for Mitigating DDoS Threat in Smart Homes - A Systematic Review
The seamless connectivity of smart homes introduces new vulnerabilities, particularly to Distributed Denial-of-Service (DDoS) attacks. Artificial intelligence (AI) offers promising capabilities for detecting and mitigating these attacks; however, existing studies primarily focus on general IoT environments rather than resource-constrained smart home settings. This systematic literature review synthesizes peer-reviewed studies on AI-based DDoS detection and mitigation in smart homes. Following PRISMA guidelines, this review analyzes success conditions, classifies AI techniques, and identifies design trade-offs. Results show a strong emphasis on detection, but limited mitigation and no recovery mechanisms. Classical machine learning and hybrid architectures dominate, with the latter primarily deployed at the gateway and edge levels. Key gaps include incomplete protection lifecycle coverage, limited privacy preservation, and lack of explainability mechanisms. Quality assessment shows high methodological rigor, with real testbeds scoring highest. These findings inform the design of resilient, privacy-preserving, and explainable AI-based defenses for smart home security.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIG SEC