Radar Sensors Measure Heartbeat and Reconstruct Speech

Small, low power radars use Doppler and powerful analysis to monitor heartbeats, detect home entry and reconstruct audio signal from vocal cords.
By Hamilton Carter, Senior Editor, Semi IP Systems
 
It seems that the days of the tricorder will soon be at hand and a research team led by Dr. Schreurs of the University of Leuven will be ushering them in.  At IMS2015 Dr. Schreurs spoke of small, low-power radars installed in the home, capable of performing a variety of medical sensing tasks.  All of these tasks are enabled by radar Doppler shift measurements and powerful data analysis software.
One of the most easily deployed applications discussed was a radar enabled fall detector.  The detector will enable medical professionals to intervene immediately when a patient falls.  We’re talking about the ultimate “I’ve fallen and I can’t get up” device.  Radar enabled fall detectors can be easily mounted on the ceiling or walls of a room.  The question of course arises, which location is more effective.   On occasion, simple questions lead to simple experiments.  Dr. Schreurs‘ team outfitted a room with a radar in each of the orientations.  They also placed an ample supply of pillows on the ground to cushion the test subjects’ falls.  Who were the test subjects you ask?  As it turns out, that’s what grad students are for!
In addition to the best orientation of a fall detector, Dr. Schreurs also presented data for a number of other medical applications requiring much higher resolution.  Among these were systems for monitoring heart and respiration rates by measuring the Doppler shift induced by minute movements in a patient’s body.   Dr. Schreurs’ team is able to analyze the data returned from their radar systems with such a degree of accuracy that they can distinguish between signals generated by a runner’s heartbeat and their respiration, providing two different diagnostic data channels.
The heart monitor functions by detecting movements of the patient’s chest wall that indicate heart movement.  How much does the chest wall move when the heart beats you ask?  It moves roughly .08 mm.  It’s not much, but it’s enough to detect.  Dr. Schreurs’ apparatus is able to reconstruct the data  using Fourier analysis to separate the more even heartbeat signals from the much larger movement signals of a shifting body.  Separating out the Doppler shift from the patient’s respiration is a bit harder to do, but it’s an achievable goal with yet another clever application of Fourier analysis.

Lest you fret that all these radar advances will bear you no fruit until your dotage, fear not!  The fall detector is a not-so-distant cousin of smart house technology.  The same type of system will be used to detect when you enter and exit rooms in your home.  Using this information, your smart-house can adjust each room to your unique specifications as you enter  To save energy, and by extension, of course, the planet, as the radar detects your exit, the system will shut everything back off.
While the NSA hardly needs more snooping toys they’re sure to get a few new ones out of this research.  Dr. Schreurs pointed out that the heartbeat/respiration technology could be used to reconstruct audio signal from an individual’s vocal cords.  The spy applications don’t stop there.  Right now, the smart-home technology is being promoted—innocuously enough—as a way to track elderly people in their homes.  The pitch is that by monitoring the motion of the elderly, doctors can detect the early onset of frailty.  However, if the system is hacked, it doesn’t take too much imagination to guess that it could be used to keep track of your whereabouts within your own home in real time.  Wait for it to appear in the next Jack Ryan movie.
In closing, if you were wondering how to best detect the fall of a grad student, (in the non-metaphorical sense of course): place the radar on the wall.
 
Related Stories:

• (Potato) Chip Bags IP