| Description |
The wireless radio channel is typically thought of as a means to move information from transmitter to receiver, but the radio channel can also be used to detect changes in the environment of the radio link. This dissertation is focused on the measurements we can make at the physical layer of wireless networks, and how we can use those measurements to obtain information about the locations of transceivers and people. The first contribution of this work is the development and testing of an open source, 802.11b sounder and receiver, which is capable of decoding packets and using them to estimate the channel impulse response (CIR) of a radio link at a fraction of the cost of traditional channel sounders. This receiver improves on previous implementations by performing optimized matched filtering on the field-programmable gate array (FPGA) of the Universal Software Radio Peripheral (USRP), allowing it to operate at full bandwidth. The second contribution of this work is an extensive experimental evaluation of a technology called location distinction, i.e., the ability to identify changes in radio transceiver position, via CIR measurements. Previous location distinction work has focused on single-input single-output (SISO) radio links. We extend this work to the context of multiple-input multiple-output (MIMO) radio links, and study system design trade-offs which affect the performance of MIMO location distinction. The third contribution of this work introduces the "exploiting radio windows" (ERW) attack, in which an attacker outside of a building surreptitiously uses the transmissions of an otherwise secure wireless network inside of the building to infer location information about people inside the building. This is possible because of the relative transparency of external walls to radio transmissions. The final contribution of this dissertation is a feasibility study for building a rapidly deployable radio tomographic (RTI) imaging system for special operations forces (SOF). We show that it is possible to obtain valuable tracking information using as few as 10 radios over a single floor of a typical suburban home, even without precise radio location measurements. |
