The spy cameras planted in various private places such as motels, hotels, homestays (i.e., Airbnb), and restrooms, etc., have raised an immense privacy concern to people. Wi-Fi spy cameras are extensively used by the adversary because of easy installability followed by size reduction. To prevent privacy invasion, the prevalent works detect wireless cameras based on video traffic analysis and require additional synchronous data from external sensors or stimulus hardware to confirm the user’s motion. Such supplements make users uncomfortable and need extra effort and time for setting. To tackle this problem, we propose an effective spy camera detection system called DeepDeSpy for detecting the recording of a spy camera with no effort from the user. The core idea is using the Channel State Information (CSI) and the network traffic from the camera to detect whether the wireless camera record the movements of the user. The CSI signal is prone to motion and detecting motion from an enormous amount of CSI data in real-time is challenging, we handle this by leveraging the Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) deep learning methods. Such synergistic CNN and BiLSTM deep learning model enables instant and accurate detection by extracting meaningful features automatically from the sequential raw CSI data. To verify the feasibility of DeepDeSpy, we implemented it on both a PC and a smartphone and evaluate it in the considered real-life scenarios (e.g., various room sizes and user physical activities). The average accuracy achieves in different real-life settings is around 96% and reaches 98.9% with intensive physical activity in the large-size room. Moreover, the ability to achieve instant detection on a smartphone within only 1 second response time makes it workable for real-time applications.
- Bidirectional Long Short-Term Memory (BiLSTM)
- Channel State Information (CSI)
- Convolutional Neural Network (CNN)
- deep learning
- network traffic
- Wi-Fi monitoring mode