DopplerCan
Main Page IAM
The aluminum cans are about 2.5" in diameter and 3" long.
The below graph shows example data of a single person running around the outside of the tower, about 1-2 meters from the base circumference. If you zoom in on the pic, it will show voltages and timing. With 8 sensors on the exterior of the tower, pointing down and out, then electronics would see a propagated wave as shown here, from sensor to sensor.
Note that the sensors can easily distinguish walking from running, in the case of a single person moving around the tower.
Here is data on essential attack/decay time of the sensors. Attack time to identify motion is about 1 second, decay time back to baseline (no motion detected) is about 2 seconds.
The below graph shows example data of a single person stepping towards, then away, from a tower strut repeated 6 times. The person would be about 3 meters from the base of the tower, about 6 meters away from the sensor, assuming sensor is about 4 meters off of ground.
Case 1)
When just one (or very few people) are performing some symmetrical synchronized activity inside or around the tower, I think the radar dopplers' will provide a very high degree of specification. They will act to give highly recognizable data to the host processor. This data could just be directly reflected to the light or laser controls to effect a near perfect mimicking of that activity in photonic form. This might be cool because no decision making (about the activity) will be required by processor. The processor just has to map the waveforms directly to light control. In this mode, tower LEDs would be seen to move in direct synchrony to runner (or runners) on the ground.
Case 2) Here I'm imagining a large group of people milling about the tower. It seems this might be the typical case. If the entire group froze, within a distance of 5 meters from the tower (and no big art cars moving within 30 meters!) then the tower dopplers could easily identify this state. As the crowd made increasingly frenetic activity starting from a distance of 5 meters away and then moving towards tower, then overall Doppler signal would rise. If the entire crowd starting jumping up and down, within 1-2 meters of tower, then Doppler signal would maximize. This maximal state could then set off some fantastic effect.
This is the form of the finished Dopplers. Radar board mounts inside. Height, with bracket is 5". Bracket is setup to permit direct mounting to bamboo (Guadua) primary vertical supports for tower. Doppler's best placement will likely be on outside of bamboo supports, pointing to ground, parallel to the support. Sensing cone is approximately 90 degrees, so a 4 meter high placement, on support, will sense up to an 8 meter area (cone base), with Doppler at vertex of cone. All Dopplers will have a RJ45 cable linking power and analog signal from a host junction box placed on one of the supports. Primary 5VDC and GND applied to junction box for distribution. All analog signals accessible from junction box via DSUB 37 pin male or female connector.