Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{159911,
        author = {M.Vishnu Priya and K.Umamageshwari and M.Pavithra and R.Bhavani},
        title = {FOG-BASED ATTACK DETECTION FRAMEWORK FOR INTERNET OF THINGS USING DEEP   Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {},
        volume = {9},
        number = {12},
        pages = {814-829},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=159911},
        abstract = {The number of cyber-attacks and data breaches has immensely increased across different enterprises, companies, and diligence as a result of the exploitation of the sins in securing Internet of Effects ( IoT)  bias. The added number of biases connected to IoT and its different protocols has led to the growing volume of zero-day attacks. Deep literacy( DL) has demonstrated its superiority in big data fields and cyber-security. lately, DL has been used in cyber-attack discovery because of its capability of rooting and learning deep features of known attacks and detecting unknown attacks without the need for homemade point engineering. still, DL can not be enforced on IoT bias with limited coffers because it requires expansive calculation, strong power and storehouse capabilities. This paper presents a comprehensive attack discovery frame of a distributed, robust, and high discovery rate to  descry several IoT cyber-attacks using DL. The proposed frame implements an attack sensor on fog bumps because of its distributed nature, high computational capacity and propinquity to edge  bias. Six DL models are compared to identify the DL model with the stylish performance. All DL models are  estimated using five different datasets, each of which involves  colourful attacks. trials show that the long short-term memory model outperforms the five other DL models. The proposed  frame is effective in terms of response time and discovery  delicacy and can descry several types of cyber-attacks with99.97 discovery rate and99.96 discovery  delicacy in  double bracket and99.65 discovery  delicacy in multi-class bracket.},
        keywords = {Element, formatting, style, styling, insert(  crucial words). },
        month = {},
        }