Smart vacuum cleaners, which use smart sensors to work independently, have gained great traction over the past few years.
The LidarPhone attack specifically targets sensor vacuum cleaners LiDAR.
LiDAR systems allow light detection and range determination, and is a remote sensing method that uses light in the form of a pulsed laser to measure distances to and from near objects.
This technology helps smart vacuum cleaners bypass obstacles on the floor while cleaning them.
The attack appears to be complex, as the attackers need to penetrate the device itself, and the attackers must be connected to the victim’s local network to launch the attack.
The team of researchers from the University of Maryland and the National University of Singapore said: We are developing a system to reuse the LiDAR sensor to sense audio signals in the environment, collect data remotely from the cloud, process the primary signal to extract information, and we call this eavesdropping system LidarPhone.
The basic idea of the attack was to remotely access the LiDAR readings of the smart vacuum cleaner and analyze the combined audio signals.
The researchers said: This would allow the attacker to listen to private conversations, which could reveal credit card data, or provide potentially criminal information that could be used for extortion.
The researchers were able to use the LidarPhone attack with the Xiaomi Roborock smart vacuum cleaner as a proof of concept.
The LidarPhone attack achieves an average accuracy of 91 percent in number ratings, and an average of 90 percent accuracy in music ratings, the researchers said.
The researchers were able to discover different sounds around the house, such as: the sound of cloth moving, the sound of using a trash can, and the sound of different musical compositions of popular news channels on television.
Several conditions can make the attack less effective. The distance from the vacuum cleaner and different volume of noise affect the overall effectiveness, and background noise levels and lighting conditions also affect the attack.
The attack is an important reminder that the proliferation of smart sensors in our homes opens the door open to many attacks.