WiFi Lab C

Post Reply
User avatar
bernard
Administrator
Administrator
Posts: 2498
Joined: January 5th, 2014, 3:12 pm
Division: Grad
State: WA
Pronouns: He/Him/His
Has thanked: 185 times
Been thanked: 789 times
Contact:

WiFi Lab C

Post by bernard »

"One of the ways that I believe people express their appreciation to the rest of humanity is to make something wonderful and put it out there." – Steve Jobs
knightmoves
Member
Member
Posts: 629
Joined: April 26th, 2018, 6:40 pm
Has thanked: 4 times
Been thanked: 108 times

Re: WiFi Lab C

Post by knightmoves »

Do you think it's fair to say that there's no reasonable way to test the WiFi lab device in a satellite format?
User avatar
RasmitDevkota
Member
Member
Posts: 54
Joined: October 23rd, 2020, 6:27 pm
Division: C
State: GA
Pronouns: He/Him/His
Has thanked: 14 times
Been thanked: 36 times
Contact:

Re: WiFi Lab C

Post by RasmitDevkota »

knightmoves wrote: September 17th, 2021, 8:47 am Do you think it's fair to say that there's no reasonable way to test the WiFi lab device in a satellite format?
In what sense? As a fair competition? Probably not as good as an in-person competition, considering how the environment plays a role, unless they let you take data over many, many trials in order to reduce error. The event can still be run though, you'll just need a volunteer to set up the transmitter unit, though I guess that could be difficult to ask for since it's not as easy as preparing weighted loads like what was required in Machines.
knightmoves
Member
Member
Posts: 629
Joined: April 26th, 2018, 6:40 pm
Has thanked: 4 times
Been thanked: 108 times

Re: WiFi Lab C

Post by knightmoves »

RasmitDevkota wrote: September 17th, 2021, 8:02 pm In what sense? As a fair competition? Probably not as good as an in-person competition, considering how the environment plays a role, unless they let you take data over many, many trials in order to reduce error. The event can still be run though, you'll just need a volunteer to set up the transmitter unit, though I guess that could be difficult to ask for since it's not as easy as preparing weighted loads like what was required in Machines.
The score for WiFi lab (the build part) is all driven by how far away a team can achieve a connection, compared to the other teams competing. If you do it in person, everyone uses the same test equipment. If you run it as a satellite event, each school provides their own WiFi router, which won't all be the same. I strongly suspect that the output power on a random selection of cheap WiFi access points is quite variable, placing the school that happens to own a more powerful transmitter at an advantage.

The receiver is also different when run as a satellite competition - each school will provide some kind of computer with WiFi - or I suppose a phone - as a receiver, and the power recorded must also depend on the gain of the WiFi antenna in the receiving computer. I'll bet they're not all the same, either.

You're probably also right about the environment - I didn't think about that, but probably everyone's "long space" is a corridor, and perhaps it matters what your walls are made of.

For Machines, every school owns standard masses, and decent scales that are more accurate than needed for the event. Probably no school owns calibrated radio test equipment. Which doesn't matter if you're all using the same equipment, because you have a relative measurement, and it doesn't matter if someone's calibration is off by 50%. But if you're using different equipment, then everyone's equipment has to be calibrated to a level that's significantly better than the gap between teams. And I don't think that's possible for this event.
User avatar
RasmitDevkota
Member
Member
Posts: 54
Joined: October 23rd, 2020, 6:27 pm
Division: C
State: GA
Pronouns: He/Him/His
Has thanked: 14 times
Been thanked: 36 times
Contact:

Re: WiFi Lab C

Post by RasmitDevkota »

knightmoves wrote: September 18th, 2021, 9:37 am
RasmitDevkota wrote: September 17th, 2021, 8:02 pm In what sense? As a fair competition? Probably not as good as an in-person competition, considering how the environment plays a role, unless they let you take data over many, many trials in order to reduce error. The event can still be run though, you'll just need a volunteer to set up the transmitter unit, though I guess that could be difficult to ask for since it's not as easy as preparing weighted loads like what was required in Machines.
The score for WiFi lab (the build part) is all driven by how far away a team can achieve a connection, compared to the other teams competing. If you do it in person, everyone uses the same test equipment. If you run it as a satellite event, each school provides their own WiFi router, which won't all be the same. I strongly suspect that the output power on a random selection of cheap WiFi access points is quite variable, placing the school that happens to own a more powerful transmitter at an advantage.

The receiver is also different when run as a satellite competition - each school will provide some kind of computer with WiFi - or I suppose a phone - as a receiver, and the power recorded must also depend on the gain of the WiFi antenna in the receiving computer. I'll bet they're not all the same, either.

You're probably also right about the environment - I didn't think about that, but probably everyone's "long space" is a corridor, and perhaps it matters what your walls are made of.

For Machines, every school owns standard masses, and decent scales that are more accurate than needed for the event. Probably no school owns calibrated radio test equipment. Which doesn't matter if you're all using the same equipment, because you have a relative measurement, and it doesn't matter if someone's calibration is off by 50%. But if you're using different equipment, then everyone's equipment has to be calibrated to a level that's significantly better than the gap between teams. And I don't think that's possible for this event.
Yeah exactly, I didn't even think about the WiFi router or anything like that, so there's definitely a variety of advantageous/disadvantageous factors, probably more than we've even mentioned, making this seem all the more impossible.....
knightmoves
Member
Member
Posts: 629
Joined: April 26th, 2018, 6:40 pm
Has thanked: 4 times
Been thanked: 108 times

Re: WiFi Lab C

Post by knightmoves »

So I was thinking a bit more about this.

"Connection" is defined as more signal power than obtained with a 3.1cm monopole antenna at a distance of 3.0m, and the receiving device used. So as long as the 3.1cm monopole antenna is made correctly by each school (and it's easy enough to ask for this to be displayed in a video), you have a reasonable calibration between different test sites by doing this. The "max distance" is a bit awkward, because each site will have a different length room, and the dB value tiebreaker isn't portable between sites.

But I suppose each site could take care to find a large room / do it in a car park, so as to avoid being limited by the length of their corridor, and so the max bonus, and impound bonus, disappear.
pg2006
Member
Member
Posts: 1
Joined: March 21st, 2022, 11:10 am
Division: C
State: CA
Pronouns: She/Her/Hers
Has thanked: 0
Been thanked: 0

Re: WiFi Lab C

Post by pg2006 »

What is the distance you are getting where the constructed device is better than the monopole?
knightmoves
Member
Member
Posts: 629
Joined: April 26th, 2018, 6:40 pm
Has thanked: 4 times
Been thanked: 108 times

Re: WiFi Lab C

Post by knightmoves »

The monopole is an awful antenna, particularly in the specified orientation.

Our best connection in testing was around 50m, although that's a bit marginal. Best achieved in a competition was 40m. Our data for received signal vs distance shows systematic rises and falls: I think what we're seeing is interference between the direct path and a reflection from the floor.
Post Reply

Return to “2022 Lab Events”

Who is online

Users browsing this forum: No registered users and 0 guests