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{203377,
        author = {Rohan Rajendra Deshmukh and Prof. S.B. Khandagale},
        title = {AirFence: IoT-Based Wireless Threat Detection and Evil Twin Identification Using ESP32 and Machine Learning},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {11187-11194},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=203377},
        abstract = {The use of networks is getting bigger and bigger in schools, public places and smart homes. This has brought some security problems that the old ways of keeping things safe cannot deal with. One of the threats is something called Evil Twin attacks. This is when someone creates a network that looks just like a real one. They make the Evil Twin network look so real that people think it is the network they want to use. The Evil Twin network is made to trick people into using it of the real network. When people do not know and connect to these networks, they are putting their personal information and passwords at risk. The systems that big companies use to find and stop these attacks are very expensive and hard to set up. This means that small schools and places cannot use them. So, there is a need for something that is cheaper and easier to use.
This research paper is about something called AirFence. It is a system that uses computers and special communication parts to find and stop wireless threats. AirFence is always looking at the networks around it and collecting important information like the names of the networks how strong the signals are, what kind of security they use and how they are sending out their signals. This information is sent to a server, stored in a database and then looked at using special computer programs that can learn and get better over time.
The computer programs in AirFence can put the networks into four groups: Secure, Risky, Critical and Unknown. These programs were taught using a set of information from about 170 networks that were looked at in the real world. Things like networks that are not encrypted signals that're stronger or weaker than they should be and names that look suspicious are used to figure out if a network is an Evil Twin. AirFence also uses some rules to check what the computer programs think to make sure it is getting the right answers.
When we tested AirFence it worked well and did not use too much power. We made some pictures to help us understand how the wireless networks were behaving what kinds of threats were there and how the signals were related to each other. AirFence is an affordable way to keep an eye on wireless security. It can be used in schools, offices and places with a lot of smart devices. In the future we want to make AirFence even better by connecting it to cloud computing using GPS to map out threats sending alerts to phones and using more advanced computer programs to make it even more accurate. AirFence and wireless security are very important. We need to keep working on making them better. Wireless networks and AirFence are the keys, to keeping our information safe.},
        keywords = {IoT Security, Evil Twin Attack Detection, ESP32, Machine Learning, Wireless Intrusion Detection, Rogue Access Point, Network Monitoring, RSSI Analysis, Wireless Threat Classification, Cybersecurity},
        month = {May},
        }