Advancements in artificial intelligence (AI) are rapidly transforming technology, but they also bring forth new privacy concerns. Who-Fi, an innovative technology, poses a significant threat by enabling the identification and tracking of individuals without the need for cameras. This system utilizes AI to analyze and interpret Wi-Fi signals, creating a unique profile for each person.
According to research published in the online journal arXiv, the technology leverages standard 2.4 GHz Wi-Fi signals to identify and monitor a person’s activities. Who-Fi employs a combination of Wi-Fi signals and transformer-based neural networks to analyze Channel State Information (CSI). This involves assessing how Wi-Fi signals behave within a space and how they change when they interact with a person’s body. The system functions similarly to radar or sonar, using these changes to create a biometric profile.
When a person is near a Wi-Fi signal, their presence creates a distinct pattern within the signal. This pattern is unique enough to be used for identification, similar to fingerprints, facial features, or retinal scans. Once trained, the system can track activities and re-identify individuals entering the network area. It can also capture data on body movements and even recognize sign language. The most notable aspect of this system is its ability to function without cameras or microphones, making it cost-effective with the use of a single-antenna transmitter and a three-antenna receiver, as indicated in the study.
