Add to Google Calendar
Date:  Thu, April 14, 2016
Time:  1:30 PM – 3:00 PM
Location:  Holmes Hall 389
Speaker:  Ashikur Rahman, Graduate Research Assistant, University of Hawaii at Manoa

Abstract

Stationary continuous wave Doppler radar has been used for displacement measurement and vital signs detection. However, on mobile platform, measurements become challenging due to motion artifacts induced by the platform. This thesis thoroughly investigates mobile noncontact vital sign monitoring device for short range application. Signal processing techniques are investigated to classify vital signs and health diagnostic and noncontact unique identification using Doppler radar alongside. Various techniques were experimented in lab environment by creating and using phantoms. Initial study began with mounting radar module on a programmable linear stage, and precise stage movements are monitored by an optical tracking system. The motion artifacts due to radar system movements are removed using IIR filter and adaptive noise cancellation techniques. The system is capable of extracting respiration rate even in the presence of radar module motion. The experiments and theoretical techniques provide a baseline that can be potentially used to measure vital signs from any arbitrarily moving radar system.  To implement a feasible field applicable solution low intermediate frequency (IF) techniques for non-invasive detection of vital signs from a mobile short-range Doppler radar platform were proposed and validated through mechanical and human experiments. RF tag and a low IF radar architecture with an adaptive noise cancellation technique is employed to extract desired vital signs motion information even in the presence of large platform motion. For a diverse range of applications of mobile Doppler radar, a see-through-the-wall (STTW) life sign detection from a mobile platform has been studied. Upon investigating motion artifact cancellation techniques research efforts were given to classify vital sign for useful diagnostic information extraction, i.e., health diagnostic and uniqueness. A continuous-wave (CW) Doppler radar-based unique-identification system has been studied. Experiments have been performed using a neural network based classifier to uniquely identify individuals based on the variation in their breathing energy, frequency and patterns captured by the radar. It has shown the possibility of non-contact unique identification where camera based system is not preferred. It is demonstrated that the system is capable of identifying individuals with more than 90% accuracy. This study also has impact on radar-based breathing pattern classification for health diagnostics. Further research is needed for applying the techniques to a larger dataset.