Hand Hygiene

Although we’re still collecting data, it is beginning to look like the RSSI signal is messy for reasons that are going to be tricky to clean up. We have a number of ideas that we can pursue:

1. Track the badge-dispenser signal for longer periods and a faster rate, hope that we get enough enough experimental power to reliably distinguish the mean signal strength at .75 and 1.0 m.
2. Change the antenna design within the dispenser unit to provide a more robust signal within the range of interest.
3. Use a pointing antenna mounted above the dispenser to create a zone of detectability right around the dispenser.
4. Add another transducer, like a sonar sensor to provide alternative distance estimate.

Here are the strength and weaknesses of each approach.

Statistical Inference
Ultimately this comes down to the power of a t-test. Based on Russel Lenth’s one-sample t-test power calculator, if we want to detect a difference of 6 rssi units with a reliability of 95% when the standard deviation of the rssi values read is 15, we need 69 samples, which is probably more than we can get practically. If the std dev is only 3, we need just 6 samples. If we think we can get 12 samples reliably (three seconds of rapid sampling), for the 95% confidence we need to achieve a std dev of about 6.  Alternatively, it might be convenient to think of one standard deviation of the error being equal to the resolution that we can achieve.

Antenna Design
One of our theories about the high sensitivity of the signals is that that the signal has a strong polarity. That could cause signal power to be very sensitive to the orientation of the mote receiver relative to the mote transmitter. We’re looking at the possibility of setting up single dipole transmitters with known polarity and seeing if we can get a more robust signal transmission that way. A 1/4-wave diopole antenna is about 4 cm long, so we could perhaps make this work.  The problem may turn out to be that such antennas are too good, making it difficult to attenuate the signal over the range of interest.

Pointing Antenna Design
We’ve constructed a pointing antenna with a coil of wire and that seemed to provide good signal in an area of interest just somewhat smaller than what we want.  The principle question is whether or not we can get the antenna to be an inconspicuous part of the device we’re building and have a reliable field for detection.

Sonar Design
Years ago, I participated in a project in which we used the time of flight of an ultrasonic signal timed with an optical signal.  We could use a similar approach here, combining a radio transmission and an acoustic signal to provide a distance estimate.  We could mount a transmitter in the hand sanitizer and a receiver in the badge.