Passive Radar Sensors and Hardware Clutter Cancellation Techniques for Practical Sensing

As the number of wirelessly connected sensors, communications nodes, and IoT devices increases, users and designers alike are realizing that the electromagnetic spectrum is a finite resource that must be conserved. Existing works have leveraged cognitive radar sensors or large-scale digital radio systems to perform sensing using ubiquitous wireless signals but have so far not demonstrated the level of practicality needed for future use cases due to large, expensive components being critical to existing designs. This dissertation presents new radar sensing architectures that leverage passive sensing paradigms, new beamforming techniques, low-cost fast-startup baseband amplifiers, and analog clutter cancellation to improve the utility and applicability of low-cost radar sensors. The dissertation begins by introducing analog passive motion sensing, and verifies its utility through theoretical modeling and experimental analysis. A beamforming passive radar is then presented, with results showing the ability to localize the angular position of a human target while simultaneously measuring its respiration frequency without contributing to spectrum congestion. An optical phased array device is then discussed, which leverages a similar theory and a high-voltage mixed-signal driver circuit to independently control up to 1,024 optical pixels. A fast-startup baseband amplifier is then presented, which offers a low-cost, high performance solution to low-frequency amplification for future sensing leveraging diodes and MOSFETs. Results using these architectures show a low degree of distortion with a vastly improved startup time. The dissertation concludes by presenting a new FMCW clutter cancellation technique, which leverages analog feedforward cancellation to remove the impacts of stationary clutter. The results demonstrate that the technique can effectively clear the scene of clutter, allowing for improved ranging using simple, low-cost architectures.