Wireless scoring equipment

[BarryTice]

Stryder --

You're close on the second question: There is just a lot of background (radio-frequency) noise that we can't help picking up. However it won't cause interference if you're using the common understanding of interference.

Instead, your foil-antenna system zeros itself on this background RF noise (Yes, Mergs. It would have to be smart enough to determine what its stable state is.) As the fencer moves or the blade bends, the length of the antenna changes. But these changes are relatively gradual and continuous -- Your 35-inch foil, during the bending transition to become a 27-inch foil, passes through 34, 33, 32, 31, 30, 29, and 28.

On the other hand, when your blade contacts your opponent's lamé, the effective length of the antenna makes an instant change (a quantum leap, for the physicists in the group) from one length to the other. This sudden change is the one that the circuitry detects as an indicator that your blade is now touching something metal (with the tip depressed).

From here, it would have to make some determination about whether it is touching a bell, lamé, or strip, and send an encoded signal (on a frequency entirely unrelated to the RF background noise that the system is "reading") to the scoring equipment.

Something you mentioned could help with this determination: Just as my circuitry judges the antenna nature of my foil, your circuitry could judge the antenna nature of your lamé. If my foil antenna increases in size at the same time your lamé antenna increases in size, the odds are pretty good that I've just hit you. Otherwise, if my foil antenna size increases but your lamé antenna size is unchanged, I've hit your bell or the strip.

With this arrangement, my circuitry would have to be able to send three distinct signals to the scoring apparatus:
1) My foil tip is depressed. (Send it for therapy?)
2) My foil antenna just grew in size.
3) My lamé antenna just grew in size.

If I send signal 1 but neither of the others, I've hit you off target.

If I send signals 1 and 2, and your circuit sends signal 3, I've hit you on target.

If I send signals 1 and 2, but your circuit does not send signal 3, I've hit your bell or the strip.

I hope this is a bit clearer. If not, keep posting.

oiuyt --

I don't know what the answer to your question is, regarding an off-target hit with a lamé contact farther down the blade. But presumably, that's already an issue with wired scoring. That's why you're required to have 15 cm of insulation on the end of your blade. Perhaps new rulings will require longer tip tapings? I certainly don't know.

-- b.r.t.

[Stryder]

Barry-
That is probably about as close to understanding as I am going to get. But James made a good point: Hasn't someone already developed a less-complicated wireless system?

Is this design you are talking about just some crazy new idea you are thinking of?
No offense meant, of course. What is confusing me is that your system seems to involve a lot of relative intangables that I never expected. Wireless communication happens every day and there is no adjustment for signal in the length of one's antenna.
(Girls say size doesn't matter.)
Is your system the low-cost analog version of what I was expecting? My cellular phone transmits data infinately more complex over ridiculously greater distances with immeasurably higher security. All without any of the worries you are suggesting, and for less than the cost of curent scoring equipment.
What's the deal?

------------------
www.geocities.com/strydermike

[Mergs]

Barry,

Since the size of the antenna you are really worried about is the lame`, why not have the system "learn" the size of the lame`/weapon combined antenna during the test touch? Thus, all it now is looking for is an antenna that is that size (+/- some percentage, say 5%). In other words, I know the size of the weapon antenna (at rest), I know the size of the combined weapon/opponent's lame antenna, anything else is extraneous and should be ignored.

This could then also be applied to the determination of whether or not you are blocking your opponent by placing your weapon on your own lame`.

Symbolically:

W1; L1 = Fencer A's weapon and lame antennas
W2; L2 = Fencer B's weapon and lame antennas
P = Piste antenna

Valid lights:

W1 + L2 (A's light)
W2 + L1 (B's light)
W2 + W1 + L1 (A's weapon grounded on his lame`; B's Light)
W1 + W2 + L2 (B's weapon grounded on his lame`; A's Light)

Invalid, no light
W1 + W2
W1 + P
W2 + P

Invalid, White Light
W1 or W2

In all cases it is assumed that the tip is depressed.

Interesting approach. Anxious to see how it plays out. What about Saber?

[This message has been edited by Mergs (edited 05-31-2001).]

[oiuyt]

BarryTice:

a quantum leap, for the physicists in the group

You mean for the NON-physicists. Physicists know that quantum leaps are the SMALLEST possible jumps.

You're right of course, the insulation would be doing a different thing (preventing touches from being registered rather than preventing them from being grounded out), but the same solution should work for the situation I mentioned.

Mergs- Sabre is easier, don't have to worry about the tip being depressed, don't have to worry about off-target. The only thing that needs to be taken into account is that touching your own lame will look the same to your sabre as touching your opponent's.

Maybe use the B/C lines to send a signal (different for each fencer) and the A line carries signal back to the box on your back hip (or whereever) the signal, if it's the same signal that the box is sending out (through the B/C line) it knows that the growth wasn't from hitting the opponent.

-B

------------------
"Oh but you can't expect to wield supreme executive power just because some watery tart threw a sword at you!"

[Mergs]

Good, OK guys, here it comes................................point. Or lack there of.

[james bird]

Barry and Mergs, thanks for formulating the idea that's been circulating - I think what we've got now is a possible sollution.
Interestingly for all the hype about wireless scoring equipment and the fact that it has now been approved by the world body, there seems to be little knowledge or information about it in the real world.
I'll look at this further but my assumption/ knowledge is that here in the UK I can get 'type approved' low power 27MHz self contained transmitters (meant for garage door opening etc.) for about £7 ($10) each, plus the exciter circuitry and detector circuitry for say another £7 a side Plus the master receiver for another £7 i.e. £35 ($50) on top of building a conventional scoring box. So this $50 replaces the cost of 2 spools which seem to cost in the region of £150 EACH in the UK.
Stryder, you ask about the complication of this sollution, remember we're not talking about the radio link here (the equivalent part to your cell phone radio), that's the bit I think I can get for around $10 a pop. In reality we are only looking to send 3 or 4 different signals from each player, a mobile is sending around 1000. The complication is in the sensing - This is the equivalent to the chip in your phone that uses predictive speech rules to compare what you say with what it thinks you'll say and sends the difference to the other end, where a similar process takes what you should have said and compares it with the difference and says what you said. (Listen to your mobile when you get 'music on hold' it starts of OK then slowly disappears into a 'waterfall', this is where the codecs get confused because music doesn't follow the rules of speech and therefore the predicted differences get to large). Mobile 'phone companies have this complication handled in a single chip that they've spent 15 years perfecting - and buy globally in millions. We are at the bottom of the learning curve, and I predict (don't quote me on this) may never see wireless fencing boxes sell in the same quantity as mobiles. The key complication we load on ourselves by trying to do this wirelessly, is that we've no direct return path between the tip and the opponents lame, therefore we're probable going to have to do some sort of 'look at both ends simultaneously' trick to get an unambiguous result.
Thanks for all your inputs, keep me confused!
All the best

------------------
James Bird
sa-fc.org.uk

[BarryTice]

oiuyt --

I stand by my use of "quantum leap." True, in physics it generally refers to an electron making a leap from one shell to another, and absorbing or emitting energy accordingly. But "quantum leap" refers not to the size of the step, but to the discontinuous change of state from one discrete energy level to another.

Webster's Dictionary shows quantum leap as:
"an abrupt change, sudden increase, or dramatic advance."

The sudden change of antenna length falls into that category, especially as compared with the gradual changes resulting from the extension of the fencer's arm or the bending of the blade.

Stryder --

Let's try to get your understanding a little bit closer by using a different analogy. Forget everything I've said about any antenna and start from scratch with this model.

Pretend for a moment that your foil is covered with solar panels, and that it is constantly generating some small voltage as you fence (or just stand there, for that matter). Pretend, also, that your opponent's lamé is covered with solar panels, as is the strip. The amount of voltage generated by each of these things is proportional to their surface area.

Now imagine that you have some circuit checking the voltage signal you're getting from your foil. When your foil hits your opponent's lamé, that voltage signal changes accordingly. Likewise, the signal changes if your foil hits your opponent's bell (as his/her foil is also covered with solar panels) or the strip.

The circuit at your end must recognize when your tip is depressed and be able to differentiate between the four states: 1) Just your foil, 2) Your foil and opponent's lamé, 3) Your foil and opponent's bell, 4) Your foil and strip.

Assuming your tip is depressed, the last two signals (opponent's bell and strip) can be ignored as no action. The first signal indicates an off-target hit. The second signal indicates an on-target hit.

Your circuitry then takes this data and uses some small radio transmitter to send a signal back to the scoring apparatus, presumably as an encrypted signal to prevent people from cheating by having someone near the strip generate fake signals. To make things perfectly clear here, the antennae used by this radio transmitter and receiver are fixed and constant, and do not ever change.

So far, I'm going to assume that all this makes sense.

Now, take that circuitry and that radio transmitter, and let's say that it is essentially finished. (It will actually have to be changed to interpret a different, um, flavor of incoming signal, but we'll get to that.)

So let's revisit the weapon-lamé-strip environment.

Realistically, the solar-panel thing couldn't work that way. (I know that. This is merely an analogy.) So instead we need to come up with another way of determining when one piece of metal touches another.

Think of a harp. It has dozens of strings of different length, each precisely tuned to a particular note. If I take a trumpet and blow a D (re) next to the harp, the D string on the harp will begin to vibrate "sympathetically", as the air around it vibrates at it's harmonic frequency. (You can try this with a guitar, too. Finger the fret to make two adjacent strings play the same note. Pluck one string, and the other will begin to vibrate, too.)

All around us, all the time, there is electromagnetic radiation "noise." Some of this electromagnetic radiation, within a certain frequency range, we sense as visible light. At other parts of the spectrum, the radiation's frequency falls into the range we use for radio waves. Other ranges fall into microwaves, X-rays, etc.

A piece of metal, in the midst of all this electromagnetic radiation noise, responds similarly to the harp strings. When there is electromagnetic noise of the right frequency to match the harmonic of the piece of metal, the metal develops a sympathetic electromagnetic signal. Which frequency is harmonic for the piece of metal is determined by the length of the piece of metal, just as (given a certain weight and tension) the aural harmonic of a harp string is determined by its length.

If we have your circuitry look at what electromagnetic signal your foil normally harmonizes with and watch for quantum changes in that signal, we'll know that your foil has touched another piece of metal.

Of course, there is still the matter of determining which piece of metal your foil has touched. But that is covered elsewhere in this thread.

Does this help any more?

Also, this isn't just some crackpot theory I came up with. It comes from my (limited) understanding of RF circuitry, from my 12 hours of physics and 6 hours of analog circuits classes in college before I switched from electrical engineering to photojournalism, and my conversations with Walter Triplette about likely directions for how things would be developed. In other words, it's someone else's crackpot theory.

Realistically, it's not as complicated as it sounds. Determining what electromagnetic frequency a piece of metal harmonizes with isn't that complicated, in theory. And sensing an abrupt change in that signal is certainly no more difficult.

If we, at rest, track the natural harmonic frequencies of both the foil and the lamé, then a change in my opponent's lamé harmonizing frequency that occurs at the same time as a change in my foil harmonizing frequency indicates that I've hit my opponent on target (assuming there's also an indication that my tip is depressed).

Personally, I think this notion would be easier than having the system "learn" the signal it would receive from the test touch. What if a fencer accidentally touches the strip during the test touch? But this notion is my opinion and speculation, and may not be how things are actually done. (Frankly, all of this may not be how things are actually done. But I haven't been able to think of any other way to do it that makes more sense.)

[This message has been edited by BarryTice (edited 06-01-2001).]

[Mergs]

Barry,

As for the touching the strip during the test touch, something as large as a 32 square meter piste (as opposed to a 1.5 sq meter lame`) would definately be outside what the system would be looking for, and ignored, just as hitting the opponent's weapon, as it would be too small.

I am curious about frequency management at a tournament, though. Do the tournament organizers do this because the relays to the scorebox are a part of the overall system, or do the individual fencers show up with it and are assigned a frequency for the duration of the tournament? Logistics and human factors questions are just as interesting as the technical ones.

[james bird]

Mergs - I'd guess that from an administration point of view, each scoring box and pair of players kit will be assigned a unique frequency band or perhaps coding (just like the newer generation of domestic cordless 'phones, where unless you want to use more than 50 or so in a room they will all coexist haply together).
Each player would put on the equipment from each set of scoring equipment each time they played on a particular pieste - same way as they now have to wire themselves into a spool each time.
Don't think it'll be to difficult to 'identify' each set of kit radio wise - The newer generation of domestic cordless 'phones do it, and with a combination of a few frequency bands and a few bits of encoding (8 bits gives 256 combinations, 16 gives 64,000) should be possible to make it robust. In the limit, it may be possible/ desirable/clever to have a system 'sniff' at power up and choose a unique event code, i.e. one that's not being used in the current environment. This'll also make it more difficult for someone to 'listen in' and either corrupt good signals, or introduce spurious ones. (I'll speak to the NSA, I'm sure they'll have lots of clever scheams they'd be happy for us to use.
Barry, love the analogy using solar panels, I've now got a lovely vision of 2 fencers wearing what looks like shinny armor, glinting and glistening as they stagger towards each other covered in glass tiles. - Solves the battery problem though.
All the best

------------------
James Bird
www.sa-fc.org.uk

[BarryTice]

Mergs --

True, it would be pretty easy to scale out the strip as an accidental touch. But I think differentiating between the opponent's foil and lamé might be harder than you imagine.

The factor that changes the harmonic frequency for an antenna is not its surface area, but its length. True, a lamé has much greater surface area than a foil and body cord.

But consider that a crouching fencer (as if kneeling to have a weight test at the strip) will have a foil antenna reaching from (approx.) his or her butt (where the body cord connects to the touch-sensing circuitry) to a little above his or her head (the tip of the foil). Compare that with the opposite extremes of right hip to left shoulder, for example, on the lamé, and the foil line is as little as 30%-40% longer.

True, with an extended arm while testing bells and lamés there will be more difference. But the circuitry also has to be flexible enough to recognize the foil as a foil in a crouch like that, with, say, 1.4 meters from butt to foil tip AND know that it's still the foil with a full lunge, with more than 2 meters from butt to foil tip.

When you add in there that a crouching fencer can still score a touch in the center of his or her opponent's chest, and that from that direction of hit, the lamé is not going to add an effective meter to the length of the scoring fencer's antenna, it becomes a much more difficult task. It's still a quantum change, but might end up tuned to a frequency still within the range of what the foil could be alone.

(Note that the change in length of antenna during a hit is the change in length from farthest ends of the antenna, essentially a straight line through the torso to the far spot on the back of the lamé. The one meter (-ish) antenna extension that you could get from hitting squarely into the hip or shoulder (from the right direction) is a maximum, not a practical liklihood.)

Furthermore, unless you score a touch on the tip of your opponent's weapon, weapon lengths aren't additive. If you make a bell hit, the length you've added to your antenna is the distance from where you hit (assuming roughly parallel blades) to your opponent's butt (roughly). If your opponent's hand is hear his or her body, that's pretty close to the length you could get from the lamé on many body hits.

Within these constraints, could you build a circuit that "learns" the size of your weapon and your opponent's lamé, and can distinguish between that and the size of your weapon with your opponent's weapon? Yes, you probably could. Would it be easier to key off the harmonic of the opponent's lamé instead? I think so, but I don't know for sure.

-- b.r.t.

[Stryder]

Barry-

Thanks for the afterschool attention.
I understand basically how all of these things work, what I don't understand is why it is necessary for the wireless scoring system to be based on such things.
I assumed that the wireless system would work just like the current system.
The fencers sending unit would interpret the signals sent to it through the lame as well as those sent and returned through the weapon.
That seems to be the real difference I didn't expect. Is there a reason that we cannopt interpret our opponent's weapon through our own lame'?

[BarryTice]

Stryder --

Wireless fencing can't be like the current system because there is no current (electrically speaking).

Electricity always moves is circles (well, loops, actually). Somewhere, you start with an atom with all its electrons. Something adds energy to that atom and breaks a few electrons off, creating an (positively charged) ion and loose electrons, which carry a negative charge. It is the desire of the electrons to get back to where the positive charges are that makes electricity work.

But in order for this to work, you must have a complete circuit. In the case of the foil, this is one lead from the score box along the floor cord, through the reel, up the body cord, up the wire in the foil groove, down the foil blade, through the body cord, reel, and floor cord, and back to the scoring box.

Concurrently, as you are scoring a hit, you have the electricity from the scoring box, all the way up through the foil, then making a return trip to the scoring box through your opponent's lamé. In either case, you have a complete circuit back to the scoring box.

With a wireless system, you have a circuit starting at the end of your body cord, up through your foil, into your opponent's lamé, and then having nowhere to go.

It's something like the science fair projects you did in fifth grade with a battery, some wire, and a light bulb from a flashlight. Even if one wire connects one side of the battery to the bulb, the bulb won't light unless another wire connects the bulb to the other end of the battery. Only then do the electrons start moving, because only then do they have anyplace to go. That's why batteries always have two terminals and electrical outlets always have (at least) two plugs.

In other words, without a common connection (or a common ground of some sort), the fencer's sending unit couldn't interpret signals sent to it through the lamé because there would be no signal sent through the lamé, because the electrons wouldn't be moving through the lamé, because they wouldn't have anywhere to go. That's also why the solar-panel analogy above couldn't be implemented, because there wouldn't ever be a complete circuit.

Theoretically, you could overcome this by generating a charge on the weapon, a charge on the lamé, etc., and then testing for a change in the charge density. But this opens itself up for lots of static shocks (ouch!). It's also an inherently slower and more gradual change than checking for RF interference.

Does this clear it up?

[Stryder]

Barry-

I understand how electricity works.
(Geez, am I coming off sounding this stupid? I am going to keep my mouth shut.) I am really smart guys, I promise. I can perform complex tasks, I can learn new tricks. I am multi-tasking! Watch me walk and type at the same time:

OK, so that was a bad example.

Anyway, back to the lesson.

There is a current for the weapon, and couldn't the scoring boxes be designed to wait for and accept a charge from the "A" line? Once that charge was recieved they could signal that a "hit received" to the central unit.

------------------
www.geocities.com/strydermike

[james bird]

I think in the end, the issue about using some sort of charge detection is that of proximity vs. contact. e.g. 2 warlike talkie radios can communicate with each other, they don't have a direct physical return path, they rely on each antenna having it's own ground plane or counterpoise from which the antenna can invoke enough potential difference to radiate a signal differentiated from this 'pseudo ground', when received at the other end, the receiving antenna, can see something changing in the ether relative to it's own 'pseudo ground' and interprets same as signal. Problem we may have, is that if we radiate from the weapon, the lame may be able to detect the weapon, but will there be enough difference in received signal between the 2 cases of 1. sword hits just off target (button pressed, sword knows it's hit something), lame picks up large signal from sword, 'cos it's only just (1mm) of target and 2. Sword hits on target?.
This is all getting a bit theoretical - Reality is there are now some approved units being sold commercially, which obviously must work and in doing so have addressed these 'on target/ off target' issues. We've got 1 or 2 possible hypotheses, think I need to get on and build something and see just how complicated it is to make a reliable system (remember why I asked the question) - I believe this is a cheaper way of building a complete system - If the reality is it needs so much electronics and processing horse power, it may in the end be less expensive to find a way of replicating the spools. I'm not worried about the complexity as such, just the relative costs.
Keep exercising those grey cells
All the best

------------------
James Bird
www.sa-fc.org.uk

[neevel]

Regarding the wireless prototypes that are already out there:

The Ukranian-developed system which was being tested prior to the Sydney games relied on the metal piste and contacts affixed to the bottom of the fencers' shoes as a common ground. There was some provision for a short-term 'touch buffer' in the fencers units for those instances when both feet were momentarily out of contact with the strip, as well as a potential rules change to prohibit jumping actions. It didn't perform up to par at the Epee World Cup it was tested at, and wasn't used at the Olympics.

-Dave

[Purple Fencer]

I think this thread has set some sort of record...all the posts total are probably larger than any 5 threads put together! I've certainly seen many more LARGE posts than normal.

However, since electronics ain't my bag, I'll just leave it to the experts.

Eric...you want me to cram two wires into a foil blade now? Oy! I have a hard enoughtime putting in one wire, amd redoing an epee always gives me fits (gotta do one tonight, as a matter o' fact).

------------------
Sam Signorelli -- Boldly going forward...'cause I can't find reverse!

[james bird]

Dave - metal contacts on peoples feet - I have enough trouble keeping up right at the best of times, with out metal to metal 'skids' to slide about on - I've now got this image of people sliding about like fair ground bumper cars (I think you call them dodgems) - All they need now is a poll sticking out the top of their mask running to a net above the piste and the picture is complete (may stop me falling over as well).
Thanks for the info about the system though.
All the best

------------------
James Bird
www.sa-fc.org.uk

[Stryder]

I saw the "metal slippers" system demo'd at Palm Springs.
I have two words for it:
Lame.
It was so lame it didn't even get the second word.
The tin foil booties would tear if I even thought of fencing in them, and the buffer idea meant that if you scored while jumping (or moving fast enough to break contact) your light would go on a seconfd later. This would ruin the directors sense of tempo.
He would call the attack "no" and a remise yes.
Silly Ukrainians!


------------------
www.geocities.com/strydermike

[Stryder]

http://olympics.smh.com.au/fencing/2...2000Jan19.html

Has a bit of info about an upcoming pre Sydney test event. But nothing afterwards of course.
They mention "wrapping the arm to ground/earth it"?

Any of you Aussies hear abou this?


------------------
www.geocities.com/strydermike

[james bird]

Wonder if this "They put some kind of wrapping on your arm to earth it." is creating the counterpoise for the sword/ antenna to radiate against? Yet another imponderable!
All the best

------------------
James Bird
www.sa-fc.org.uk