Internet at Competitions

[EricH]

Originally Posted by yash101

First of all, do you mean FMS (Field Management System)? Also, you say, don't use WiFi in the competitions, but you don't really explain why. Also, yes, they will get angry at you, but it isn't a bad thing, like how you are making it sound. That is to just keep the fields working properly. WiFi is a thing that is quite useful in competitions because it will network devices together, allowing them to communicate.

Interference. That's why no WiFi. Field working properly is HUGE! You weren't around in 2012, I don't think--get your hands on a copy of the Einstein Report from that year, and see just what can go wrong if either the field or the robots aren't working properly.



FYI, all: From the 2014 Administrative Manual, Section 4.15, 3rd bullet:

Do not arrange for Internet access or phone lines on the site or attempt to connect to the Internet.

I'm guessing this replaces the WiFi restriction from 2013 and prior years.

[philso]

Originally Posted by yash101

I mean a small difference, just in the antenna, like 2.6GHz instead of 2.4!

How would changing the antenna will change the bandwith that the radio operates in. In the radio systems that I worked on (precursor to WiFi), one would have to make a change in the radio itself to make that happen.

The FCC, a US government agency, manages the radio frequency spectrum in the US. The ISM bands that WiFi operates on are well defined by FCC regulations. By straying outside of the defined bandwidth, you may be interfering with some other radio system, possibly with very serious consequences for the operator of that system. Do it for long enough and someone from the FCC may visit you and slap you with a fine.

Originally Posted by antimatter_john

Well it's a good way to scout with the possibilities of it.

Also it allows teams to make something cool with their app or web app using the internet.

Why is it a superior way to scout?

Is the purpose of the team to compete at a tournament or to make cool phone apps or web apps?

Originally Posted by EricH

Interference. That's why no WiFi. Field working properly is HUGE! You weren't around in 2012, I don't think--get your hands on a copy of the Einstein Report from that year, and see just what can go wrong if either the field or the robots aren't working properly.

The radio band can carry only so much data at any particular time. Would you rather it be data for controlling your robot on the field or data for scouting?

[saikiranra]

Originally Posted by philso

Is the purpose of the team to compete at a tournament or to make cool phone apps or web apps?

It opens a whole world of possibilities. If you have internet in the pits and in the stands, you can seamlessly and effortlessly transfer data to create comprehensive match strategies. In addition, you can talk to the scouts via chat and ask them for opinions and suggestions all from within your pit. It also helps with collaborative scouting, for something like CowScout, where you want multiple sources of data from different events.

[techhelpbb]

I agree with the others:

FIRST made a decision to use WiFi.
Way back when....before the cRIO...I recall many conversations about how great WiFi would be.

Okay so we have WiFi.
On the field side you have Cisco, on the robot side to date: D-Link and monitoring for the lot on 5GHz spectrum.
There is monitoring on the 2.4GHz spectrum but it's hard to enforce it because that spectrum is typically a mess.

On the high side of the 5GHz spectrum you have bands assigned to radar.
Luckily I've yet to see a FIRST team deploy 5GHz radar on a robot .

That said: if there are multiple fields and you are in the 5GHz WiFi bands you should stay out of range of the fields (read - just don't do it). There are only so many channels in that 5GHz WiFi band and it's *extremely* likely with anything that can bond 2 or more channels (for 300Mbps or greater) you will short change the field merely by existing. Plus that can cause chaos with the robot radios as they are capable of discovering that interloper network.

Now the issue starts turning into this: Apple devices are increasingly happy to use the 5GHz WiFi spectrum. So ad-hoc networks between Apple users on 5GHz are point and click (and mind you that's not even for Internet). Per the Einstein report more and more mobile devices are using chipsets with 5GHz WiFi support. All of those can click their way into becoming an issue and per that report have become an issue (the details of that have been covered at length).

All that said: when I helped propose a 2015 control system I intentionally proposed a secondary low frequency radio for field related traffic leaving all the WiFi spectrums available. The entire point was to prevent the field safety mechanism from disabling robots because of radio link level issues. I proposed to do this using 3rd party modules with FCC approval attached to them. This meant that you could swap the hardware modules and the sockets were attached to the MCU in such a way that a very wide selection of frequencies were available below 1GHz. That meant you could accommodate radio regulators everywhere on Earth merely by swapping those modules. Mind the point that whatever was on the WiFi spectrum was no longer my concern. If a team wanted to use 2.4GHz on their robot for their vision systems it would not matter. If a team wanted to use 5GHz on their robot for their crazy Theremin controller not my issue.

As it stands FIRST (speaking as a 3rd party) has a vested interest in not altering the 2014 field system because they have end-of-life Cisco gear that is expensive and likely a pile of D-Link radios they can waste. Not to mention I noticed that last year antennas were changed which added some cost. I also noticed some additional changes in the way WiFi spectrum use was monitored.

I have, in the past, had to ask teams to turn off their 5GHz WiFi near the field. I will not say whom. There was a WiFi router with an open network sitting literally next to the field. Please just do not do this. You have bluetooth and other choices. Heck if you are clever you have infrared. Do not risk making issues on the field merely because you have zeal to do something interesting. Situations like the Einstein report need to never happen again and now that it has been investigated I doubt anyone will be able to plead ignorance of the risk.

In the past at the Mount Olive district event WiFi in the 2.4GHz spectrum was available but it was swamped by people with their phones and just shear demand. In reality the same bandwidth throttling technology now in use on the fields would help with that...but I bet very few events have any Internet available that actually has controlled bandwidth usage because that equipment adds cost (yes it can be done with a PC but most people just buy the finished product and have no interest in hacking it themselves). So with that all said: I personally am in favor of offering teams Internet access - by twisted pair ethernet in the pits. This has the distinct advantage of consuming zero radio spectrum unless the event coordinators decide to backhaul the traffic with WiFi. Still bandwidth controls should exist. However it should be much easier to reliably cap any team's Internet traffic if they are wired to the network as there should be much less risk of packet loss. For the teams this would mean they could use bluetooth or cellular near the field and dump their data at zero risk to the venue at the pits (these days there is way more than enough cheap storage to buffer up data in a smartphone).

[philso]

Originally Posted by saikiranra

It opens a whole world of possibilities. If you have internet in the pits and in the stands, you can seamlessly and effortlessly transfer data to create comprehensive match strategies. In addition, you can talk to the scouts via chat and ask them for opinions and suggestions all from within your pit. It also helps with collaborative scouting, for something like CowScout, where you want multiple sources of data from different events.

What you are saying is true. It does not answer my question regarding what your priorities are. As I stated earlier in my post, the WiFi bandwidth can carry only so much data. Do you want to use it to control your robots or do you want to use it chat and pass around data about robots that would kill on the field if they could be made to operate? This is an engineering competition. In engineering, one has to make tough choices based on one's priorities.

Originally Posted by techhelpbb

I personally am in favor of offering teams Internet access - by twisted pair ethernet in the pits. This has the distinct advantage of consuming zero radio spectrum unless the event coordinators decide to backhaul the traffic with WiFi.

This is also true. The data rates could then much higher than what you could achieve with WiFi. It would also take a lot of work to set up properly. The type of equipment used to create a such network is not typically meant to be moved around and I suspect that after some time, some of the connections would become flaky and teams would complain loudly here.

[yash101]

I think that that's when you build a backpack with a router (with wireless off), connected to a battery. Then, you have ethernet cables connected to the router, coming out of the bag. Also, use the Computer as a router. Set up ICS in Windows and just use the router as a switch.

Otherwise, just buy an ethernet switch!

[Andrew Schreiber]

Originally Posted by yash101

Also, you say, don't use WiFi in the competitions, but you don't really explain why. Also, yes, they will get angry at you, but it isn't a bad thing, like how you are making it sound. That is to just keep the fields working properly

Here's why - FIRST has asked teams nicely NOT to use wifi at events as it interferes with the Field. Yes, ticking off the FTA is a bad thing. They are already stressed enough without your punk team deciding the rules don't apply to them for WHATEVER reason you are using to justify it.

Frankly, I'm appalled that you're so easily discussing interfering with the field. But I'm more annoyed at how easily you discount the impact your actions will have on the volunteers who will face the repercussions. Because when the field doesn't work nobody is going to come to you, they'll go to the FTA and then the FTA will have to spend an hour of their time hunting down your stupid wifi network while listening to people blame them/NI/Cisco.

[Koko Ed]

FIRST is deadly serious about the Wifi issue. They had Wifi sniffers at the champs tracking down people who were setting up Wifi hotspots. And every event I went to they tell teams repeatedly to turn off any wifi they set up. After the fiasco of 2012 they are not in the mood to have that happen again and trust me you do not want to have happen to you what happened to the dude who was causing mischief that day.

[coalhot]

Originally Posted by yash101

Also, you say, don't use WiFi in the competitions, but you don't really explain why. Also, yes, they will get angry at you, but it isn't a bad thing, like how you are making it sound. That is to just keep the fields working properly. WiFi is a thing that is quite useful in competitions because it will network devices together, allowing them to communicate. That is the reason why you use Bluetooth in competitions!

This topic has been beaten to death in numerous other threads. Do a search, you'll find them.

As for the WIFI, I agree that offering it would provide many benefits, not limited to teams being able to use the internet, but also to less interference with the field because teams aren't setting up their own connections.

That being said, it's not on the teams to bring their own 4g hotspot and router and set up their own network. It's on FIRST and the event staff to allow it and set it up. Please don't break the rules, the FTA and field staff have a very difficult job to do, and they don't need teams and individuals causing them numerous delays to the event schedule because they decide to do something unwise.

[saikiranra]

Originally Posted by philso

What you are saying is true. It does not answer my question regarding what your priorities are. As I stated earlier in my post, the WiFi bandwidth can carry only so much data. Do you want to use it to control your robots or do you want to use it chat and pass around data about robots that would kill on the field if they could be made to operate? This is an engineering competition. In engineering, one has to make tough choices based on one's priorities.

All of that is true, IF you use Wi-Fi. As many other posters have noted, there are plausible Wi-Fi alternatives that works just as well, if not better. In engineering, one has to make tough choices, or be smart about the design and have all the features you want.

To answer OP's original question, I would try sanddrag's method. It has successfully worked with us in the past and is fairly easy to set up. If you can't use the internet, which might be the case this year, try an offline server running on WAMP.

[techhelpbb]

Originally Posted by philso

This is also true. The data rates could then much higher than what you could achieve with WiFi. It would also take a lot of work to set up properly. The type of equipment used to create a such network is not typically meant to be moved around and I suspect that after some time, some of the connections would become flaky and teams would complain loudly here.

Considering we move the actual electricity in the pits over and over I fail to see how this is any worse.

In point of fact you need a managed switch or switches with adequate ports. It does not need to be faster than 10Mb. Maybe VLAN everyone from each other. Get a load balancer if you can or make one. We already move the load balancer and switches in the Scorpion cases with the field (over and over). Maybe it needs a UPS like the Scorpion case or maybe not.

Then we just need transit to plug into which from the perspective of the field crew is no different than what they and the streaming folks need.

Originally Posted by Andrew Schreiber

Because when the field doesn't work nobody is going to come to you, they'll go to the FTA and then the FTA will have to spend an hour of their time hunting down your stupid wifi network while listening to people blame them/NI/Cisco.

True but long before that happens the field crew should see this if they are paying attention to the gear they have. Plus when I work a field I bring my own gear and put it around.

They are still causing an impact and it shouldn't be going on: otherwise they'll see me wandering around like a hunting dog fox hunting them. I used to provide engineering work to a WiFi security company so I won't be easy to shake.

I have seen FTAs so busy they don't have time to deal with this. The issue with not stepping on it is that the more people do it the less they perceive the consequence and clearly real risks to all of FIRST exist.


On a different note:

I noticed last year that FIRST started off with a single channel on each radio effectively limiting each robot. This limitation did not matter because the load balancer gave each robot less than that for bandwidth on the field side anyway. As the 'christmas tree' issues continued I noticed FIRST started using both channels again on the premise that would resolve this issue (the idea apparently being that at power-up the D-Links were flooding the available channel and then failing without channel bonding available).

FIRST must realize at this point that, just like 2.4GHz WiFi, the days of using 5GHz WiFi like this are limited. Between the number of access points and client network cards from a vast array of manufacturers there are far too many permutations between the teams, the venues and the spectators to patrol away all the risks. At this point I've yet to see an example of the field WiFi security not seeing 2.4GHz spectrum filled with far too many sources of interference at a given venue. Sooner or later the 5GHz spectrum which by default is depended upon will look just like that. At some point FIRST will start to encounter situations where infrastructure wireless systems from nearby sources are using 4 channels or even more for reasons that FIRST has no control over and there just are not enough channels as the number of fields at use at one time increases. I figure FIRST can expect reasonable compliance from teams. Maybe some compliance from the venues. I think it is going to be *very* difficult to control non-aligned spectators. Probably impossible to control nearby infrastructure wireless users. That was the reason I proposed sitting on radio frequencies that were easier to patrol and harder for every 'joe blow' to just wander into.

I would hate to see a situation where some unsuspecting person brings in a device that manages to impact the field access for all or some robots at a single event and cause a problem difficult to reproduce elsewhere. I am not confident at all that such a circumstance will be caught in a way that does not impact team rankings (even if for a short while).

[philso]

Originally Posted by Andrew Schreiber

Here's why - FIRST has asked teams nicely NOT to use wifi at events as it interferes with the Field. Yes, ticking off the FTA is a bad thing. They are already stressed enough without your punk team deciding the rules don't apply to them for WHATEVER reason you are using to justify it.

Frankly, I'm appalled that you're so easily discussing interfering with the field. But I'm more annoyed at how easily you discount the impact your actions will have on the volunteers who will face the repercussions. Because when the field doesn't work nobody is going to come to you, they'll go to the FTA and then the FTA will have to spend an hour of their time hunting down your stupid wifi network while listening to people blame them/NI/Cisco.

Yash

Think this through. Please! What if the interference you produce causes one or more robots on the field to malfunction? What if an alliance lost a match because of it? What if an alliance lost Einstein because of it? What if your team, or one of your friends teams, was on that alliance? These are NOT hypothetical questions!

Did anyone notice that the robots on the practice field are required to run with ethernet tethers? Maybe the do that to prevent interference on the actual competition field. Please don't make arguments about "being far enough away not to cause interference" unless you are able and willing to prove it by doing a comprehensive survey of the competition venue like what the cell phone companies do to select cell sites.

Originally Posted by techhelpbb

Considering we move the actual electricity in the pits over and over I fail to see how this is any worse.

It is not impossible. It is just even more work for the people setting up. It is not too hard to find volunteers to help run power cords and check that they work.

[brennonbrimhall]

To get back to the OP's question:

We simply run our stuff locally, hosted on a laptop (WAMP, Node.js, etc.). We have old D-Link bridges (pre-2012) that are no longer legal/in use, so we use them in bridge mode (the same as if you were using a tethered robot) with wired ethernet connections with static IP addresses. We can link up to 4 scouting laptops in this fashion (four clients if you're not using a dedicated machine to host).

If you're using data in your app from the Internet, see if there is a way to cache it in your application. If you're using PHP/Node.js, there are easy ways to serialize objects--PHP has built-in serialization and supports JSON, Node.js has JSON, etc. I'm not familiar in any way with Ruby, but it doesn't seem too hard to serialize there either.

[yash101]

Just install DD-WRT x86 on a laptop with a powerful antenna! That'll hold all that bandwidth! You'll have an actual processor instead of a 240MHz ARM!

Doing that will say good bye to the possible hiccups in the network infrastructre if you don't want to pay a thousand dollars for a router!

[Abhishek R]

I agree with what some others have been saying. It hurt to see robots from the stands on Einstein 2012 (Einstein 2012 official report) just sitting there, doing absolutely nothing, to see a season's worth of effort and work end anticlimactically. And I'm sure it was even worse for the teams on the field. I would much rather see robots working to their potential on the field than have chat messages between the pit and stands. Could it be useful, yes, but I don't think it outweighs the dangerous possibilities of interference.

Besides, I can practically guarantee that no matter how effective a web based app is (we had developed one that we could use last season) you will still have to have a face-to-face strategy meeting before alliance selections anyway, because it's just difficult to communicate full ideas and tactics through. I feel like taking that part out would be killing half the fun and excitement of the strategic aspect of competition. I just don't see too much advantage to it. It could lead to miscommunication, etc.

tl;dr Internet at competitions is something that needs some time before it can develop into something that can SAFELY be widely used to benefit. FIRST has asked nicely, let's honor that.