1

I hope this question is not out of bounds. I'm a EE with little RF and you guys seem to know a LOT about RF...

I have a small vehicle traveling in a straight line at low speed. I need to know when it crosses a line that is perpendicular to it's travel. I can't use the normal proximity or through beam sensors. I propose to embed an antenna in the "finish line" (which needs to span 25-30 feet), mount a receiver on the vehicle, and monitor something like RSSI to determine when it passes the finish line. The receiver would need to be about 3 feet above ground level where the antenna is.

Is this possible? What frequencies would work best and are available for something like this?

RayZ
  • 11
  • 1
  • 1
    A GPS receiver in the car is probably easier, and more accurate... – Phil Frost - W8II Mar 26 '19 at 14:56
  • 1
    Define "normal proximity" sensor that you cannot use. Why can't you use one? The "rules" say so, or ? – mike65535 Mar 26 '19 at 16:04
  • How about magnets in the ground and Hall effect sensors under the vehicle? – Mike Waters Mar 26 '19 at 16:11
  • Phil - I need centimeter accuracy, and very small size, and battery power. I think this is asking too much from GPS. – RayZ Mar 26 '19 at 20:26
  • Mike65536 - the location of the sensor has to be about 3 feet above the ground where the "finish line" is. I don't know of a prox that can do that. – RayZ Mar 26 '19 at 20:28
  • Mike W - I don't know if a Hall sensor will work from a distance of 3 feet. But maybe with N52 magnets? I'll try to run this down. – RayZ Mar 26 '19 at 20:30
  • @RayZ I don't either. I suggest that you ask on https://electronics.stackexchange.com, as [this is beyond the scope of this site](https://ham.stackexchange.com/help/on-topic). – Mike Waters Mar 26 '19 at 22:29
  • I just shared this in [EE chat](https://chat.stackexchange.com/rooms/15/electrical-engineering). – Mike Waters Mar 27 '19 at 18:41
  • "Better just to paint a stripe of paint on the ground and use a camera or reflective sensor on the vehicle to detect it" - per @ThePhoton in EE chat. – Mike Waters Mar 27 '19 at 18:57
  • 1
    @RayZ A Hall sensor will not give you centimeter accuracy at 2 feet away. BTW, if you type an "@" sign before usernames in your comments (like I just did here), the person will receive a notification that you replied. – Mike Waters Mar 27 '19 at 19:02
  • 1
    @Mike Waters Thank you for helping. Unfortunately, none of the ideas presented so far will work for this application and I still don't have any better handle on using RSSI. I've discovered that Analog Devices makes a couple dozen different log RF power detectors; I'll try to get to an app engineer there. – RayZ Mar 29 '19 at 14:56
  • You can read about my first naive attempt at something similar [on my blog](https://edeca.net/project/kart-timing-mk1/). I did produce an upgraded system that solved many problems but did not document it. Summary: it’s a tricky problem and more EE not amateur radio. – David Mar 31 '19 at 22:19

3 Answers3

1

Use induction. Bury conductors underground just below the surface or on the surface is conductive coating. Run current thru the conductors which creates a fixed magnetic field (DC current). Have a inductor coil of suitable size mounted on the front of the car to sense the magnetic field and create emf and current. Use that current as an indicator to signal you have crossed the line.

K7PEH
  • 2,948
  • 10
  • 14
  • K7PEH - the concept may be valid, but my gut tells me that sensing a small current from 3+ feet away is harder than sensing the RSSI of an antenna, for which there are sophisticated off-the-shelf chips. And if I made a circuit sensitive enough, any nearby stray fields would be a problem. But it's an interesting concept - thanks. – RayZ Mar 27 '19 at 14:05
  • 1
    @RayZ traffic lights detect cars using inductive sensors with nothing in particular on the car designed to help. So it's definitely feasible if you can engineer both the sensor and the car. – Phil Frost - W8II Mar 27 '19 at 17:31
0

It sounds like there are requirements that are not called out in the question but, since it invites some interesting "spitballing," here's a thought: Doppler effect.

I would propose to mount a small transmitter on the vehicle(s) and connecting the antenna to a receiver. As the transmitter approaches the antenna, the receiver will observe a higher carrier frequency than the one on which the transmitter is known to be operating. As it passes the antenna, the receiver's observation of the carrier frequency will coincide with the known frequency. As the transmitter travels away from the antenna, the received frequency will drop.

The higher the carrier frequency, the greater the Doppler effect. It should be possible to operate in one of the ISM bands with an off-the-shelf transmitter. With narrow receiver bandwidth, the receiver will produce a "blip" indicating the arrival of the transmitter with arbitrary precision.

Adding a second receiving antenna, at some distance from the first, perpendicular to the direction of travel of the transmitter, creates opportunities for other algorithms to indicate the transmitter's position.

Brian K1LI
  • 7,808
  • 10
  • 33
  • Brian - yes, there are other details but I offered what I thought was enough to frame the question of using RSSI to measure position/distance. I initially discarded the Doppler idea as being too low in resolution since the speeds are less than 20mph. But I will revisit and run some numbers - thanks. – RayZ Mar 27 '19 at 15:00
  • @RayZ The numbers predict 720Hz frequency offset of a 24GHz signal at 20MPH. Narrow DSP filters would make it possible to discriminate. 24GHz CW/FM transmitter chips are now commercially available. Alternatively, a vehicle radar system might fill your need. – Brian K1LI Mar 27 '19 at 15:19
  • It's an interesting concept but I think the resolution is too low. 720Hz is the offset at the max 20mph. But at 5mph it's only 179Hz. I can't see how I'd determine when it crossed the finish line with any accuracy. – RayZ Mar 27 '19 at 16:09
0

How about a laser or other infrared transmitter/receiver similar to the safety device on your garage overhead doors?

Benjie
  • 29
  • 3
  • I think you're referring to a through-beam sensor. They won't work because there may be 8 or 10 vehicles crossing the finish line very close to each other. – RayZ Mar 27 '19 at 19:54