Many smartphone users believe their privacy is protected once they lock their device, turn on airplane mode, or disconnect from the internet. However, a new research paper from China claims that smartphones may still reveal certain activity patterns through weak electromagnetic signals emitted during operation.

Researchers from the People's Public Security University of China say they have developed a technology that can analyze these signals to identify which apps are running and what type of activity is taking place—without unlocking the phone or accessing its internal data.

The research was published in the peer-reviewed journal Radio Engineering on May 22. However, it is important to note that the experiments were conducted only in a controlled laboratory environment and have not yet been proven effective in everyday conditions.

How Does This Technology Work?

The technique is described as a non-contact digital forensic method, meaning researchers attempt to gather information without directly accessing the smartphone's operating system or stored files.

According to the researchers, smartphones produce very weak, low-frequency electromagnetic signals while performing different tasks. These signals change depending on how the phone's hardware is being used.

For example:

• Video apps may increase activity of the processor and graphics chip.
• Navigation apps may activate GPS and storage systems.
• Communication apps may use wireless components differently.

By studying these variations, artificial intelligence models can reportedly identify patterns linked to specific apps or activities.

Researchers Claim High Accuracy

The research team tested the technology on:

• Apple iPhone 15 Pro
• Xiaomi 15 Pro
• Oppo Reno13

Researchers claimed that the system identified the app being used with 99.07% accuracy during testing.

The experiments included apps such as video platforms, messaging apps, maps, browsers, cameras, and cloud storage services.

The team also claimed the system could identify video playback actions—such as pausing, normal playback, or 2x speed—with around 95.61% accuracy on tested platforms.

How Are These Signals Captured?

The researchers used specialized equipment placed near the smartphone.

The process involved:

1. Attaching an induction coil to the back of the phone.
2. Connecting it to a digital monitoring receiver.
3. Recording electromagnetic signals in the 150 kHz to 650 kHz range.
4. Converting the signals into visual patterns called spectrograms.
5. Using a deep-learning model to analyze the patterns.

The researchers believe this approach could potentially support digital investigations where direct access to a device is difficult.

Does This Mean Anyone Can Track Your Phone?

Not at this stage. Experts and the researchers themselves point out several limitations.

The tests were conducted:

• In a controlled laboratory environment
• Using highly sensitive equipment
• With the receiver placed very close to the smartphone

The study does not confirm whether the method would work:

• From a distance
• In crowded electromagnetic environments
• On a wide range of everyday devices

The technology currently appears to be a research concept rather than a practical surveillance tool available for general use.

What Could Happen Next?

Researchers say future studies may focus on:

• Identifying unknown devices through electromagnetic patterns
• Creating unique electromagnetic fingerprints for smartphones
• Exploring whether signals can be detected through less specialized hardware, such as smartphone sensors or wearable devices

While the research raises interesting questions about smartphone privacy, more testing will be needed before understanding its real-world impact. For now, standard security practices—such as keeping software updated, using strong passwords, enabling device encryption, and reviewing app permissions—remain important ways to protect personal information.