AI Powered IoT Security Monitoring System

Abstract

The rapid growth of Internet of Things (IoT) technology has led to its widespread adoption in critical applications such as smart cold storage systems, healthcare, and industrial automation. However, IoT devices are highly vulnerable to cyberattacks due to limited computational power, insecure communication channels, and lack of intelligent security mechanisms. Traditional rule-based security systems fail to detect advanced attacks such as replay attacks and data injection attacks. This paper proposes an AI-powered IoT security monitoring system designed to detect anomalies and replay attacks in a smart cold storage environment. Sensor data collected using NodeMCU (ESP8266) is transmitted to a Flask-based server where a machine learning model analyzes the data for abnormal behavior. If an anomaly is detected, the system automatically blocks the attacker’s IP address and generates alerts; otherwise, the data is forwarded to the cloud for monitoring using ThingSpeak. Experimental results demonstrate that the proposed system improves real-time security, enhances detection accuracy, and ensures reliable cold storage monitoring.

Introduction

The paper addresses security challenges in IoT-based cold storage systems, which are widely used to preserve food, medicines, and vaccines through real-time monitoring of environmental conditions. Despite their benefits, such systems are highly vulnerable to cyberattacks due to weak authentication, insecure communication, and limited device resources. Attacks such as replay attacks, false data injection, and DDoS can compromise sensor data and lead to incorrect control decisions. Traditional rule-based and cryptographic security methods are insufficient to detect evolving and intelligent threats.

To overcome these limitations, the study proposes an AI-powered IoT security monitoring system that combines environmental monitoring with real-time cyberattack detection. Using Machine Learning–based anomaly detection, the system learns normal sensor behavior and identifies malicious activities at both the sensor and network levels. Unlike many existing solutions, the proposed approach includes automatic mitigation mechanisms such as blocking malicious IP addresses and generating alerts.

The literature review highlights growing interest in AI-driven IoT security, emphasizing the effectiveness of ML, Deep Learning, and reinforcement learning in detecting anomalies and zero-day attacks. However, existing systems often lack sensor-level analysis and automated responses, which this work aims to address.

The proposed algorithm initializes sensors and a pre-trained AI model, continuously collects and preprocesses sensor data, and classifies it as normal or anomalous. Normal data is forwarded to the cloud for visualization, while anomalous data triggers security actions such as IP blocking, alert generation, and incident logging. The system also controls cold storage actuators based on verified data, ensuring both operational reliability and security. Overall, the approach provides an intelligent, adaptive, and real-time security solution for IoT-based cold storage environments.

Conclusion

This research presents an AI-powered IoT security monitoring system capable of detecting anomalies and replay attacks in smart cold storage environments. By integrating machine learning with IoT infrastructure, the system enhances security, reliability, and automation. The proposed solution effectively identifies malicious activities and responds automatically through IP blocking and alert generation.

References

[1] N. Sharma and P. Dhiman, “A Survey on IoT Security: Challenges and Their Solutions Using Machine Learning and Blockchain Technology,” Cluster Computing, Springer Nature, vol. 28, pp. 1–19, Apr. 2025. doi: 10.1007/s10586-025-05208-0. [2] V. Menon U., V. B. Kumaravelu, V. Kumar C., R. A., S. Chinnadurai, R. Venkatesan, H. Hai, and P. Selvaprabhu, “AI-Powered IoT: A Survey on Integrating Artificial Intelligence with IoT for Enhanced Security, Efficiency, and Smart Applications,” IEEE Access, vol. 13, pp. 22844–22879, Mar. 2025. doi: 10.1109/ACCESS.2025.3551750. [3] C. Gilbert and M. A. Gilbert, “AI-Driven Threat Detection in the Internet of Things (IoT): Exploring Opportunities and Vulnerabilities,” International Journal of Research Publication and Reviews, vol. 5, no. 11, pp. 219–236, Nov. 2024. doi: 10.2139/ssrn.5259702. [4] M. Prabhakar, S. Selvarani, S. Manikandan, M. Preethika, and N. Suganya, “A Comprehensive Survey on IoT Security Challenges, Threats, and Emerging Countermeasures,” International Journal of Recent Technology and Engineering (IJRTE), vol. 12, issue 4, Apr. 2024. [5] Y. Raj and S. D. Kamble, “A Survey on the Role of Artificial Intelligence in Enhancing IoT Security,” International Journal of Advanced Research in Computer and Communication Engineering (IJARCCE), vol. 12, no. 11, Nov. 2023. [6] T. Mazhar, D. B. Talpur, T. A. Al Shloul, Y. Y. Ghadi, I. Haq, I. Ullah, K. Ouahada, and H. Hamam,“Analysis of IoT Security Challenges and Its Solutions Using Artificial Intelligence,” Brain Sciences, MDPI, vol. 13, no. 4, p. 683, Apr. 2023. doi: 10.3390/brainsci13040683. [7] A. Imran and W. Shah, “AI-Powered Cyber Security for IoT: Enhancing Network Resilience and Privacy,” ResearchGate Preprint, Dec. 2023. doi: 10.13140/RG.2.2.35571.03363. [8] F. Zaheer, “AI-Powered Cybersecurity for IoT Systems: Threat Detection and Privacy Preservation,” International Journal of Advanced Research in Computer Science and Engineering, Dec. 2023.

Copyright

Copyright © 2026 Sarthak Pagar, Shashwat Deshmukh, Vivek Dake, Tushar Dalave, Prof. Suvarna Potdukhe. This 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.