The proliferation of Internet of Things (IoT) devices has revolutionized multiple sectors, promising significant societal benefits. With an estimated 29 billion IoT devices expected to be interconnected by 2030, these devices span from common household items to advanced sensors and applications across various domains. However, the extensive scale of IoT networks has introduced security challenges, including vulnerabilities, cyber-attacks, and a lack of standardized protocols. In response to evolving threats, machine learning techniques, particularly for malware detection, have made significant strides. This survey focuses on a less-explored aspect of IoT security: the potential of energy-based attack detection. We aim to provide an up-to-date, comprehensive understanding of this approach by analyzing the existing body of research. We explore the diverse landscape of machine learning methodologies employed in IoT security, emphasizing the energy-based approach as a valuable tool for detecting and mitigating attacks. Furthermore, this survey underscores the significance of power consumption analysis in identifying deviations from expected behavior, enabling the detection of ongoing attacks or security vulnerabilities. Our survey offers insights into the state-of-the-art techniques, methodologies, and advancements in energy-based attack detection for IoT devices. By presenting a structured roadmap through the literature, research methodology, and in-depth discussion, we aim to aid researchers, practitioners, and policymakers in enhancing IoT security. This survey's unique contribution lies in bridging the gap in the literature regarding energy- based approaches and underscoring their potential for fortifying IoT security. Future research in this direction promises to significantly enhance the safety and resilience of the IoT landscape.

Energy-based approach for attack detection in IoT devices: A survey

Merlino, Valentino
;
Allegra, Dario
2024-01-01

Abstract

The proliferation of Internet of Things (IoT) devices has revolutionized multiple sectors, promising significant societal benefits. With an estimated 29 billion IoT devices expected to be interconnected by 2030, these devices span from common household items to advanced sensors and applications across various domains. However, the extensive scale of IoT networks has introduced security challenges, including vulnerabilities, cyber-attacks, and a lack of standardized protocols. In response to evolving threats, machine learning techniques, particularly for malware detection, have made significant strides. This survey focuses on a less-explored aspect of IoT security: the potential of energy-based attack detection. We aim to provide an up-to-date, comprehensive understanding of this approach by analyzing the existing body of research. We explore the diverse landscape of machine learning methodologies employed in IoT security, emphasizing the energy-based approach as a valuable tool for detecting and mitigating attacks. Furthermore, this survey underscores the significance of power consumption analysis in identifying deviations from expected behavior, enabling the detection of ongoing attacks or security vulnerabilities. Our survey offers insights into the state-of-the-art techniques, methodologies, and advancements in energy-based attack detection for IoT devices. By presenting a structured roadmap through the literature, research methodology, and in-depth discussion, we aim to aid researchers, practitioners, and policymakers in enhancing IoT security. This survey's unique contribution lies in bridging the gap in the literature regarding energy- based approaches and underscoring their potential for fortifying IoT security. Future research in this direction promises to significantly enhance the safety and resilience of the IoT landscape.
2024
IoT
Attack detection
Power consumption
Malware detection
Anomaly detection
Energy consumption
File in questo prodotto:
File Dimensione Formato  
Energy-based approach for attack detection in IoT devices A survey.pdf

accesso aperto

Tipologia: Versione Editoriale (PDF)
Licenza: Creative commons
Dimensione 568.77 kB
Formato Adobe PDF
568.77 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.11769/640594
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 1
  • ???jsp.display-item.citation.isi??? 1
social impact