Secure Access for Massive Devices in the Industrial Internet of Things (IIoT).

In the context of the widespread adoption of Industrial Internet of Things (IIoT) technology, the number of IoT devices is rapidly increasing, leading to a significant security concern in the form of device impersonation attacks. LoRa devices, as common wireless IoT terminal devices, play a critical role in ensuring the security of the IoT ecosystem through their identity recognition mechanisms. This paper presents a method for secure access of massive IoT devices based on LoRa wireless fingerprints. The proposed method extracts fingerprint features from three types of LoRa devices, which serve as unique wireless identifiers, enabling rapid identification of a large number of LoRa-based IoT devices. The process begins with preprocessing and feature extraction of the collected signals, resulting in three distinctive feature sets that form the wireless fingerprints of the devices. These fingerprints are then used to train a classification model using the K-nearest neighbors (KNN) algorithm. The trained model is subsequently employed to predict the device identity by analyzing the fingerprints of the devices to be identified. Experimental results demonstrate that the proposed system achieves a recognition accuracy of 95.3% under a signal-to-noise ratio of 55dB, while requiring a minimal training time of only 0.3169 seconds.