GPS tracker detection made easy with off-the-shelf hardware

Cyberstalkers are increasingly turning to cheap GPS trackers to secretly monitor people in real time. These devices, which often cost less than $30 and run on 4G LTE networks, are small, easy to hide under a bumper or in a glovebox, and can go undetected for months.

A new paper from researchers at NYU, You Can Drive But You Cannot Hide, presents an affordable, practical method for detecting these hidden cellular GPS trackers using off-the-shelf hardware. The approach is designed to help victims and security pros identify surveillance devices attached to vehicles, without expensive gear.

“We present an innovative algorithm designed for effective uplink frequency analysis, enabling dependable detection within a three-foot range when utilizing standard commercial hardware,” the authors write.

The threat: cheap, powerful, and hard to spot

Stalking and surveillance using GPS trackers is not a fringe threat. In the U.S. alone, 2.7 million people were stalked using technology in 2019, and about 14% of victims were tracked with electronic devices, including GPS hardware.

Trackers such as the Amcrest AM-GL300W, and Brickhouse Spark Nano 7 operate using 4G LTE IoT bands like LTE Cat-M and NB-IoT. These devices send updates every minute when a vehicle moves and rely on low-power cellular connections to communicate with cloud servers.

Because they transmit in uplink bands that are narrow, weak, and buried under normal cellular noise, these signals are hard to detect using conventional RF tools.

Law enforcement has reported cases where cyberstalking led to break-ins and even murder. In 2023, a man pled guilty in Missouri after placing a GPS tracker on his ex-girlfriend’s car, then using the location data to stalk and kill her.

The detection method: uplink signal scanning

The NYU team focused on scanning the uplink frequency range, the path from the tracker to the cell tower, rather than the more common downlink direction. This change allowed for greater signal isolation and better results using inexpensive gear.

The tinySA Ultra portable mini spectrum analyzer covers a frequency range of 100 kHz to 5.3 GHz

At the heart of the system is a $140 tool called the tinySA Ultra, a portable RF spectrum analyzer capable of capturing signals in the 700–1900 MHz range. Combined with public data from OpenCellID and CellMapper, the tool helps pinpoint which frequencies to monitor based on local towers and carrier bands.

Once the system knows where to look, it listens for periodic signal spikes that match known GPS tracker behavior, typically one burst every 60 seconds when a car is moving. The spikes stand out clearly on the tinySA display, especially when using the “maximum hold” and “maximum decay” visualization modes.

“We were able to detect the presence of a 4G LTE IoT cellular GPS vehicle tracking device in the laboratory environment and vehicle environment using the tinySA Ultra spectrum analyzer,” the authors write.

Real-world testing: what works, what doesn’t

The team tested five commercial GPS trackers operating on AT&T and T-Mobile LTE bands. Detection was most reliable within a three-foot range, enough to sweep a vehicle from several positions. Indoor and outdoor experiments showed strong correlation between detected signal peaks and GPS tracker transmissions.

The system reliably detected trackers placed in glove compartments and mounted under bumpers. At three feet, signal strength was clear; beyond that, it dropped sharply. When asked whether this would hold up in real-world use, the researchers ran tests while driving.

“If the user observes similar regular peaks on the tinySA while driving or immediately following a drive, they can confidently suspect the presence of a cellular GPS tracker on their vehicle,” they write.

However, the method doesn’t work for passive GPS trackers that store data locally instead of transmitting. It also can’t detect embedded or modified devices designed to evade detection. And Verizon-based trackers weren’t tested due to lack of compatible hardware.

“We did not evaluate embedded vehicle locating devices or devices that have been modified to limit their transmission rates to avoid detection,” the researchers note.

Implications and future directions

For physical security teams, domestic violence advocates, and corporate protection units, this research offers a DIY toolkit that can be deployed quickly and cheaply. It also lays the groundwork for mobile apps or dedicated detection hardware that could warn users of tracking attempts in real time.

The team is exploring partnerships with support groups such as the Clinic to End Tech Abuse (CETA), which works with domestic violence survivors. They also propose building a smartphone-connected “black box” that automates scanning and alerts.

“Our aim is to explore the development of a system similar to an AirTag, which uses a smartphone to alert users about the possibility of a tracking device following them,” the paper states.

As 5G and 6G networks expand, the detection method will need updates to keep pace with cellular protocols. But for now, the researchers have delivered a rare win in the fight against covert surveillance, one that doesn’t require a law enforcement badge or a five-figure RF scanner.