This conference presentation discusses research addressing the problem faced by correctional facilities of the smuggling of cell phones and other wireless devices into prison walls.
The authors seek to resolve the problem faced by correctional facilities of regarding the smuggling of cell phones and other wireless devices. They chose to use cell phones of varying models and multiple low-cost software-defined radios (SDRs) in order to map intercepted signals to indoor locations within a few-meter radius. The different types of cell phones provided them with a robust dataset for location fingerprinting due to the different transmitter hardware in each device. The SDRs allowed researchers to receive raw IQ (inphase and quadrature) data from WiFi signals while being more cost-efficient for smaller facilities. That raw data was collected from a prison-like environment in a grid pattern and associated with the location where the signals were captured. The authors used an advanced machine learning network to use the raw signals as input and locations as labels, in order to map the signals to their respective locations. They compared accuracy of their system against related work previously done on this topic. The authors augmented their own original input with values such as channel state information and received signal strength indicator, and measured their effect on the system’s overall performance. The results provide prisons with a tool capable of locating unauthorized devices in zones for confiscation.