Survey on Iot Security: Attacks, Threat Classification, and Detection Techniques

AUTHOR(S):

Aisha Ibrahim Gide, Sagir Ibrahim

TITLE

Survey on Iot Security: Attacks, Threat Classification, and Detection Techniques

ABSTRACT

With billions of connected devices transforming daily life and industries, the Internet of Things (IoT) is growing exponentially. However, there are serious security risks associated with this expansion, such as privacy violations, problems with authentication, and cyberattacks like DoS, MITM, and phishing. This research investigates techniques for intrusion detection including signature-based, anomaly-based, and hybrid as well as IoT security mechanisms like fog computing, blockchain, edge computing, deep learning (DL) and machine learning (ML). It analyzes machine learning (ML) and deep learning (DL) techniques for securing IoT networks, including SVM, CNN, LSTM, and ensemble models. Problems like computing overhead and changing threats continue to arise despite advancements. In order to improve IoT security, future directions will focus on blockchain integration and lightweight, adaptable techniques.

KEYWORDS

Internet of Things (IoT), machine learning, deep learning, intrusion detection system, cyber security

Cite this paper

Aisha Ibrahim Gide, Sagir Ibrahim. (2025) Survey on Iot Security: Attacks, Threat Classification, and Detection Techniques. International Journal of Internet of Things and Web Services, 10, 87-95