According to a recently released report, Wi-Fi routers tend to locate their users’ exact location inside a house. Not only this, but the pose setting they will be in can also be guessed with the help of this device.
The research was carried out by Carnegie Mellon University, situated in the U.S. In this research, the team used several Wi-Fi routers to gather information. While the study came out positive that indeed routers tend to pinpoint the exact location, the results are raising questions over the potential breach of privacy.
Just like sonar uses waves to locate objects, routers use Wi-Fi signals to find people. An artificial intelligence-supported software program will then analyze the differences between the signals going out and coming back in. Signals that return generate wireframe pictures that tell the location of people either in a room, in a house, or a building.
However, the accuracy of the results is still questionable. While most of the results were accurate, there were some cases in which the results either couldn’t be generated or the poses suggested by the system were termed “unnatural.”
Another area that needs to be focused on is the number of targets that can be located. Currently, the devices can locate only three targets; however, it can be expected that the number can be increased by working on them.
To carry out the research, a TP-Link router, which costs only thirty-two dollars, was used. In comparison to this, devices that are specifically designed to locate targets cost way more, which gives routers a good advantage.
Not only this, but Wi-Fi devices also have an advantage over CCTV cameras. Security cameras’ views can easily be blocked by anything, such as a wall or a gate. On the other hand, signals can go beyond these objects and locate anyone who is hiding or blocked from the camera’s view.
Since the university carried out successful research, it appears that the team will be continuing their project to overcome weak areas or fix the limitations that were faced during the study.
Read next: Here’s Why GettyImages and Shutterstock Are Responding So Differently to AI
The research was carried out by Carnegie Mellon University, situated in the U.S. In this research, the team used several Wi-Fi routers to gather information. While the study came out positive that indeed routers tend to pinpoint the exact location, the results are raising questions over the potential breach of privacy.
Just like sonar uses waves to locate objects, routers use Wi-Fi signals to find people. An artificial intelligence-supported software program will then analyze the differences between the signals going out and coming back in. Signals that return generate wireframe pictures that tell the location of people either in a room, in a house, or a building.
However, the accuracy of the results is still questionable. While most of the results were accurate, there were some cases in which the results either couldn’t be generated or the poses suggested by the system were termed “unnatural.”
Another area that needs to be focused on is the number of targets that can be located. Currently, the devices can locate only three targets; however, it can be expected that the number can be increased by working on them.
To carry out the research, a TP-Link router, which costs only thirty-two dollars, was used. In comparison to this, devices that are specifically designed to locate targets cost way more, which gives routers a good advantage.
Not only this, but Wi-Fi devices also have an advantage over CCTV cameras. Security cameras’ views can easily be blocked by anything, such as a wall or a gate. On the other hand, signals can go beyond these objects and locate anyone who is hiding or blocked from the camera’s view.
Since the university carried out successful research, it appears that the team will be continuing their project to overcome weak areas or fix the limitations that were faced during the study.
Read next: Here’s Why GettyImages and Shutterstock Are Responding So Differently to AI
