System-on-chip based doppler radar occupancy sensor with add-on passive node

Occupancy sensors can save up to 80% of energy used for lighting and heating ventilation and air conditioning (HVAC) systems, resulting in huge financial savings. However, currently available occupancy sensors --- most commonly passive infrared (PIR) and ultra-sonic (US) sensors --- have significant drawbacks, including high rates of false-positives and failures to detect stationary persons. Newly availably high-efficiency Doppler radar transceivers can be used to discern human cardiopulmonary motion thus providing a practical solution for such problems in occupancy sensing. In addition, the advent of the integrated low-power microprocessor/RF-transceivers provides a new platform which combines sensing, processing and communication to form the core of a wireless smart sensor network (WSSN) for applications such as "smart buildings".;A significant challenge for wide adoption of occupancy sensors is to demonstrate reliable system performance at low power with low cost. This research demonstrates the feasibility of low cost, low power Doppler radar occupancy sensor by building a customized passive sensor node into commercially available SoC's (TI's CC2530 and CC430). Experiments using periodic moving mechanical target illustrate that these SoC based Doppler radar sensors are able to accurately detect the motion of the target under continuous wave (CW), modulated CW and packet operation modes. The study on sensitivity and power consumption under these modes indicates the most cost efficiency and power efficiency can be achieved by operating the sensor under packet mode with an optimum output power level. A comparison between passive sensor node and quadrature receiver shows sensor node does not sacrifice the sensitivity by using simpler and lowercosting configuration. Null-point sensitivity study shows that though the configuration with add-on passive sensor node suffers from the decreased sensitivity, the specific pattern at this point can be used for a decision on occupancy detection. Simulation and experiment demonstrate characteristic pattern of the respiration signal detected by the Doppler radar occupancy sensor by charting how the variation of target position impacts the strength of the real motion frequency and its second harmonic. An algorithm detects the true presence occupancy is developed based on this pattern analysis, and is evaluated to be effective with human testing. This research also includes a broad-band mixer design for the add-on passive node using 0.18-mum IBM7HP CMOS process. The broad-band performance of this mixer is presented.;Future work will include single channel Doppler radar measuring the displacement of target in periodic motion by making use of the relationship between the detected characteristic pattern and the target location , and harvesting ambient RF energy by the same passive sensor node configured in this dissertation.