Abraham Tetteh, Sebial J. Archival, Angie M. Ceniza-Canillo
The opportunistic Internet of Things (O-IoT) enhances traditional IoT by using adaptive and spontaneous resource and data accessibility. It establishes networks based on events and requirements rather than relying on permanent infrastructure. This is particularly beneficial in situations when establishing stable infrastructure is impractical or costly. Nevertheless, the intermittent connectivity of O-IoT increases security flaws such as the possibility for malicious nodes to interrupt operations, participate in surveillance, and initiate attacks. It is crucial to identify these potential threats in order to preserve the security and reliability of the network. This study presents a comprehensive examination of the security hurdles and techniques for identifying malicious nodes in Opportunistic Internet of Things (O-IoT) networks. These networks utilize adaptable and context-driven resource and data sharing. The study emphasizes the distinct vulnerability of O-IoT, specifically since it depends on dynamic and spontaneous connections rather than a stable infrastructure. The primary objective of this study is the thorough analysis of several security concerns, such as possible delays, eavesdropping, and attacks carried out by rogue nodes. In order to mitigate these risks, the study examines a number of essential techniques for detecting them: anomaly detection to discover anomalies in behavior, machine learning classification to differentiate between benign and hazardous nodes, and trust management frameworks to ensure the reliability of nodes. The study is also to increase user trust and stimulate the usage of O-IoT in critical applications, such as disaster response and remote surveillance. This will be achieved by highlighting the significance of including rigorous safety precautions to minimize risks and maximize the advantages of O-IoT. © 2024 IEEE.
University of San Carlos, Department of Computer, Information Sciences and Mathematics, Cebu City, Philippines
