This is not going to be my usually full of pictures and graphics fanpost. This is one to discuss future concepts for hockey while also presenting a couple of brainstorm ideas of my own. I do explain a few things before hand to help understand some of the tings I am talking about, but since this is brainstorming concepts, it is not heavy in big pretty pictures. We as fans know what we would like, so lets see what type of realistic ideas we can come up with.
In a previous Fanpost I talked about how the Goal-Line technology used by FIFA, Hawk-Eye, is not ideal or ready for Hockey.
This did get me to thinking (dangerous I know) as to what could work?
The biggest drawbacks of Hawk-Eye, is it requires a clean Line of Sight (LOS) from numerous cameras, and as range from the camera increase, accuracy, which is vital to determining goals, decreases. Cameras in the net do not work simply because the net moves, so the cameras for Hawk-Eye have to be at much more distant fixed points.
On the drive to work one day, I was thinking about the dilemma, on how to track and map a puck in space and time. Since in order for the pucks to be directly observed somehow 100% of the time in order for a system to be effective, this eliminates anything that uses visible light, including the extend regions at infrared and below, and ultra violet. Those spectrum's have the same limitations as visible light, requiring a clean and visible LOS. But when you get into higher energy Electromagnetic Radiation, like Radio and Microwaves, those wavelengths have the ability to penetrate solids and liquids, a required aspect to adequately track and map a puck in motion.
So how can radio waves be used to track a puck? Everyday I go to work, I have to present my Common Access Card (CAC) Military ID to get through numerous checkpoints. And on my ID is a possible answer, a Radio frequency Identification Device (RFID) chip. It is used to store information about me, so it can be quickly read by an electronic device, such as a CAC Card reader. The thing is when you are issued this, you are either given or offered a special copper lined envelope to carry hold it in. The reason for the envelope is, the RFID can be accessed remotely, even while still inside my wallet in my pocket simply by using radio waves. The copper lined envelope acts as a Faraday shield, to prevent radio waves from accessing my RFID.
The point of this is the RFID with a transmitter and reciever can be used to measure distance using time, more specifically time difference between transmission and reception. Just like RADAR does, it would use the same concept that the difference in time from transmission to arrival will determine range, the big difference is bearing and altitude is NOT needed like in RADAR. Using numerous Radio sources, an RFID chips location can be determined using Trilaterartion, a form of trigonometry using spheres with known radii. The GPS receiver in your phone (if you have one there) uses Trilateration with GPS satellites to determine the receivers location. The GPS receiver receives signals from the GPS satellites. These signals are embedded with a time code from the satellite. Your phone uses the difference in the received time from the satellite with the time code in the receiver (both are synced to the nanosecond level) to determine how long it took the signal to travel from the satellite to your phone. This time difference tells how far away the receiver is from that satellite, the satellites exact position is a known point (also embedded in the signal) making it possible to triangulate the receiver using multiple satellites.
By knowing the distance from on satellite (using the time difference) you know you can be at any point on a sphere that is that distance from that satellite.
When you have distance known from two satellites, you know you will be at a point that intersects the two range spheres. This intersection will form a 2D circle, the receiver is somewhere on this circle.
When three Satellite distances are known, this will reduce the circle, to two points on the circle, where the third sphere intersects them. You now know the receiver can only be in one of two places, the points. (*note: an actual location may be determined at this point if one of the two points is in an impossible location, like in the mantle!)
With a fourth satellite, the receivers position will be known, since only 1 point in space will be common to four separate satellite spheres.
That is how trilateration works and how GPS determines your position (fix) on the Earth, or above it.
My thought was, could this same concept used in GPS location be used in RFID location? Could a constellation of RFID transmitters and receivers be placed above Hockey Rink to locate and map a moving puck, even if a clear LOS obscured? Instead of the puck being a receiver like your phone, the receivers are paired with a transmitter, and measures the time difference from transmission and the receiver measures the time difference from reception from the RFID tag on the puck. Basically using the exact same concept as how GPS works, only the transmitter and receiver work together to determine range to the RFID to generate the range spheres. Also is it possible to place 3-5 RFID tags in a puck to determine the pucks orientation for use in a goal line technology?
I then did some research and found, the answer for if you can track a puck and map it is not only yes, but in concept it is actually already being done. But the problem with it is how it is being done, and where. It is being used in factories and warehouses to track product and material. Finite accuracy is not required, so errors of 0.5-9.0 feet are not uncommon, depending on the several factors involved. So this is not looking good for the concept. Because of this error, it is not practical in any form, and also prevents any form of reliability in determining puck orientation, something that would need fractions of an inch accuracy to do. Even GPS has errors in range, some due to satellite position error, others due to good ole' Albert Einstein and his theory of General Relatively. The satellites are traveling so fast, they are actually going through time at a slower rate relative to us on earth. It is a measurable rate and the satellites time Codes have to be routinely adjusted to compensate for this time effect, if unchecked it will lead to progressively larger position errors.
Upon reading numerous studies and tests on RFID tracking I found that the ranges typically being measured from the transmitter/reciever were from many dozens of yards to/from the RFID tag. This would be pretty much right in line with the same ranges you would expect from a system using the rafters to hold the transmitters and receivers. But one big thing that stuck out to me was, the smaller the distance to the RFID was and the more power throughout the transmission-reception cycle, the more accurate the measurements were.
So this really gets me to the meat of the point of the discussion I would like to have, perhaps some here have an expertise in these concepts I have. Are they possible? Are they feasible? Are they practical? Or am I a total nut-job?
If as range increases and/or power decreases, accuracy in position decreases, what if we made it so range and power were at the closest and highest possible?
The purposes of Goal line technologies is to determine goals, not track a puck across a 200' x 85' sheet of ice, only did it fully cross the goal line. With a goal being the only concern, instead of sensors around the rink, what about13 transmitter-receivers in the goal posts and cross bar (4 in the posts, 5 in the cross bar) with 5 inlaid into the ice as well, or any number that is really sufficient for a high degree of accuracy and precision.
Will this number (or any other number) spacing and proximity to the puck and goal line, lead to the level of precision required to determine exactly the pucks location in space and time?
In the Fanpost about Hawk-eye, I mentioned cameras for that could not be in the net due to the net moving, and a goal is a puck completely passing the goal plane which is defined by the goal line, not the goal posts. The same issue applies here, if the goal moves or pivots, the apparent RFID goal plane based on the posts will also shift, though by rules the true goal plane itself does not. Could a system like Hawk-eye, at the same time map the goal to determine its points to properly compensate for its movement and accurately map the pucks position based on the Goal frame RFID data?
I also suggest placing 5 (or however many needed) RFID in the ice, that are separate from the goal RFID sensors, since these are fixed and cannot move. These would be used primarily for when the puck is being obscured from being visible from overhead or in the event too many bodies over the puck prevent accurate mapping from the goal posts RFID sensors. If technology warrants are can can it, they could also be integrate with the goal post data.
I tried reading deeper into the RFID studies, but it started using maths with all sorts of funny looking letters and shit with some squiggly lines here, hooks there, occasionally a number and letter too, but no equal signs.........I think. Needless to say, the math was a LITTLE past Multiple-Divide-Add-Subtract.
So my big questions or desired discussion points for the committee are these.
1- Can 3-5 passive RFID tags be placed at fixed-known points inside a puck.
2- Can RFID transmitters-receivers be placed into a metal goal frame and still have the power required for accurate and precise measurements.
3- Can RFID transmitters-receivers be placed under 1" of ice and still function properly and effectively without effecting the ice above it, and also prove the precision needed?
4- Can two technologies, Hawk-eye for Goal Frame location and RFID be successfully married into one to quickly and automatically determine both goal frame location, goal plane location, puck location and if the puck fully passed the goal plane?
One final thought. After this I went back and thought about the constellation idea again. even though it may only be accurate to 2-3 feet, perhaps this approach may also be one to consider for real time player tracking, in addition to one for goal line technology. Passive RFID tags a cheap and can easily be embedded in any uniform item, like the name tape on the back of the sweater. Perhaps both concepts are something for hockey and the fans?