Smartphone Security: Your phone can be tracked even when locked or in flight mode, Chinese researchers claim..

If you believe that locking your smartphone, enabling flight mode, or turning off the internet guarantees complete privacy, you are mistaken. A new study from China might surprise you. Researchers at the People's Public Security University of China claim to have developed a technology capable of identifying which app is running on a smartphone and what the user is doing by analyzing the faint electromagnetic radio signals emitted by the device. The researchers state that this technique does not require unlocking the phone, accessing the operating system, or extracting stored data. The study was published on May 22 in the peer-reviewed journal *Radioengineering*; however, testing has so far been limited to a controlled laboratory environment.

What is this new technology, and how does it work?
Researchers at the People's Public Security University of China developed this technology for digital forensics. According to the research paper, it is a non-contact forensic technique, meaning information can be gathered without physically touching the phone or accessing its operating system and stored data. The researchers explain that smartphones emit very weak, low-frequency electromagnetic signals during operation. These signals fluctuate based on the specific app being used and the user's activities. By analyzing these signals, it is possible to determine which app is active on the phone. The study suggests that this method could help gather additional technical evidence during digital investigations.

Claim of up to 99.07% accuracy
The research team tested this technology on the Apple iPhone 15 Pro, Xiaomi 15 Pro, and Oppo Reno 13. According to the researchers, the model identified the mobile app being used with 99.07% accuracy. The tests covered apps such as Douyin, WeChat video calls, Baidu Maps, SMS, the web browser, the camera, and cloud storage services. Moreover, the model identified—with 95.61% accuracy—whether a user was pausing the video or audio, playing it at normal speed, or watching it at double speed. This testing was conducted on platforms such as YouTube, Bilibili, and Tencent Video.

How does the technology detect this using radio signals?
According to the research paper, every mobile app utilizes the phone's hardware differently. Some apps place a heavier load on the processor and graphics chip, while others heavily activate components like the GPS, Wi-Fi module, storage controller, or wireless communication circuits. This alters power consumption patterns and generates distinct low-frequency electromagnetic signals. For instance, the graphics processor and decoding hardware remain highly active during video streaming, whereas navigation apps utilize GPS and storage intermittently. By recognizing these distinct signals, artificial intelligence infers which app is running on the phone and the nature of the activity taking place.

How were the signals collected, and what is the entire process?
Researchers attached a specialized induction coil to the back of the smartphone and connected it to a digital monitoring receiver. This device recorded electromagnetic signals in the 150 kHz to 650 kHz range. Subsequently, these signals were converted into time-frequency spectrograms and analyzed using a spectrogram transformer deep learning model. Researchers state that this method could assist investigative agencies in gathering behavioral technical indicators without extracting actual data from the phone. They have described it as a useful technique for conducting investigations discreetly under specific circumstances.

What are the next steps for the research?
Although the researchers have asserted the accuracy of this technology, they have also acknowledged that all tests were conducted within a controlled laboratory environment. In this setup, a high-precision digital monitoring receiver was placed in very close proximity to the smartphone. The research paper did not specify how effective this technology would be in crowded electromagnetic environments, over greater distances, or under real-world conditions. The team states that future work will focus on identifying previously unknown devices, creating unique electromagnetic fingerprints for each device, and capturing signals using less specialized equipment. Researchers also suggested exploring the possibility of capturing such signals in the future using smartphone magnetometers or electrodes from wearable electronic devices.

Disclaimer: This content has been sourced and edited from TV9. While we have made modifications for clarity and presentation, the original content belongs to its respective authors and website. We do not claim ownership of the content.