Enhancing IOT Security Threat Detection with Big-Data Analytics and Localized Clustered Anomaly Detection
Vinny Sukhija, Brij Mohan Goel · Stallion Journal for Multidisciplinary Associated Research Studies · 2025
IoT devices have unlocked new opportunities for automation & connectivity across industries like never before. But this growth has also added significant security challenges, such as data breaches, unauthorized access, and malware attacks. Static security mechanisms frequently overlook the adaptive characteristics of these attacks, resulting in elevated false positive rates and latency of response. This work presents an improved threat detection framework for the IoT that effectively employs Big Data Analytics and localized anomaly detection techniques to improve the accuracy and efficiency of IoT security. The proposed system also discusses the behavioural model processing of IOT devices locally that significantly reduces server load and response delay and provides a scalable solution for large-scale IoT networks.