Efficient Anomaly Detection Technique for Future IoT Applications

Internet of Things (IoT) provides smart wireless connectivity and is the basis of many future applications. IoT nodes are equipped with sensors that obtain application-related data and transmit to the servers using IEEE 802.15.4-based wire-less communications, thus forming a low-rate wireless personal area network. Security is a major challenge in IoT networks as malicious users can capture the network and waste the available bandwidth reserved for legitimate users, thus significantly reducing the Quality of Service (QoS) in terms of transmitted data and transmission delay. In this work, an Anomaly Detection Mechanism for IEEE 802.15.4 standard (ADM15.4) to improve the QoS of the IoT Nodes is proposed. ADM15.4 also proposes a mechanism to block the malicious nodes without affecting the overall performance of the medium. The performance of ADM15.4 is compared with the standard when there is no such anomaly detection is present. The results are obtained for different values of SO and for different sets of GTS requesting nodes and are compared with the standard in the presence and absence of malicious nodes. The simulation results show that the ADM15.4 improves data transmission up to 19.5% from IEEE 802.15.4 standard without attacks and up to 52% when there is malicious attacks. Furthermore, ADM15.4 transmits data 33% reduced time and accommodate 56% more GTS requesting legitimate nodes as compared to the standard in the presence of the malicious attacks.

@article{Alvi2025,
  title     = {Efficient Anomaly Detection Technique for Future IoT Applications},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {16},
  number    = {1},
  year      = {2025},
  publisher = {The Science and Information Organization},
  author    = {Ahmad Naseem Alvi and Muhammad Awais Javed and Bakhtiar Ali and Mohammed Alkhathami},
  doi       = {10.14569/IJACSA.2025.01601110},
  url       = {https://doi.org/10.14569/IJACSA.2025.01601110}
}