Abstract:WiFi connections are vulnerable to simulated attacks from rogue access points (AP) or devices whose SSID and/or MAC/IP address are the same as legitimate devices. This kind of attack is difficult to counter with traditional network security mechanisms. In this paper, we propose a new security mechanism that uses environment-independent features extracted from channel state information (CSI) to detect and identify rogue WiFi devices or APs, and reject their connections. We find that due to the I/Q imbalance and imperfect oscillator of each WiFi network card (NIC), the nonlinear phase error of different subcarriers will vary with the NIC. Through our experimental verification, this cross-subcarrier phase feature is invariant to the location and the environment. We deploy systems on two platforms that can extract constant phase errors from the constantly changing CSI in less than one second, which is at least 8× faster than that of the state-of-the-art solution. Extensive experiments on commercial routers and end devices in different scenarios show that based on the Industral Platform Computer (IPC) platform, where only non-encrypted rogue connections can be detected, the detection accuracy rate reaches 96%, and the false alarm rate is less than 2%. Based on the ASUS router platform (a commercial WiFi router), WiFi channels and smart device types are not restricted, which greatly improved universality, and the accuracy of device connection detection can even reach more than 99%. We improve a device-type identification method based on the communication traffic features of the device when connected to WiFi. Experiments show that even if there are multiple similar devices from the same manufacturer, the accuracy of device type detection exceeds 99%.

Revisiting Indoor Intrusion Detection with WiFi Signals: Do Not Panic over a Pet!

A real-time and robust intrusion detection system with commodity wi-fi

An Environmental Intrusion Detection Technology Based on WiFi

On Detecting Hidden Wireless Cameras: A Traffic Pattern-based Approach.

AR-Alarm: An Adaptive and Robust Intrusion Detection System Leveraging CSI from Commodity Wi-Fi

FreeSense:Indoor Human Identification with WiFi Signals

WiHunter: Enabling Real-time Small Object Detection Via Wireless Sensing.

Artificial intelligence driven Wi‐Fi CSI data mining: Focusing on the intrusion detection applications

WiCross: I Can Know when You Cross Using COTS WiFi Devices

Device-Free Stationary Human Detection with WiFi in Through-the-Wall Scenarios

Detecting Radio Frequency Interference For Csi Measurements On Cots Wifi Devices

Walls Have No Ears: A Non-Intrusive WiFi-Based User Identification System for Mobile Devices

Large-Area Human Behavior Recognition with Commercial Wi-Fi Devices.

Wi-Dog: Monitoring School Violence with Commodity WiFi Devices.

Attention-Enhanced Deep Learning for Device-Free Through-the-Wall Presence Detection Using Indoor WiFi Systems

How Can I Guard My Ap? Non-Intrusive User Identification For Mobile Devices Using Wifi Signals

Robust and Passive Motion Detection with COTS WiFi Devices

WiPIN: Operation-free Person Identification Using WiFi Signals.

Security Analysis of WiFi-based Sensing Systems: Threats from Perturbation Attacks

Real-Time Identification of Rogue WiFi Connections in the Wild