Scalable lightweight blockchain with anomaly detection using deep learning for IoT devices
The Internet of Things (IoT) process data from sensors to analyze long-term trends but, they suffer from security vulnerabilities. To handle these issues, the blockchain (BC) framework can be integrated with IoT. However, conventional blockchains face scalability issues, that limits intrusion detection. To overcome these, a novel BLOCKchain Anomaly Identification using Deep learning (BLOCK-AID) framework is introduced. The proposed model detects malicious packets using Betweenness Centrality based Improved Elman Spike Neural Network (IESNN). The malicious packets are blocked using Type 2 fuzzy logic (T2FL) and consensus node is selected using Firefly algorithm. The developed approach has been assessed using accuracy (AC), energy consumption (EC), packet loss rate (PaLR), and False Alarm Rate (FAR). The BLOCK-AID approach obtains the accuracy of 99.72 %, whereas the MADCS, Block-Fest and CACVO attains 78 %, 86 %, and 93.8 % accuracy respectively. These findings reveals that the BLOCK-AID model presents a secure, and energy-efficient solution for IoT environments.
Authors
- A. Ahilan
- S. Faizal Mukthar Hussain
- S.A. Mohammed Uveise
- S.M.H. Sithi Shameem Fathima
Institutions
- Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology (IN)
Publication Details
- Journal
- Ain Shams Engineering Journal
- Published
- 2026-11-17
- DOI
- https://doi.org/10.1016/j.asej.2026.104460
- Primary Topic
- Blockchain Technology Applications and Security
- Type
- article
- Field-Weighted Citation Impact
- 0.00