Ask Slashdot: Ad-Hoc Wireless Mesh Network For Emergency Vehicles?

First time accepted submitter Texaskilt writes "I am looking to put together a mobile mesh network for my volunteer fire department and would like some recommendations from the Slashdot crowd. Ideally, the network would consist of cheap wireless routers (Linksys WRT-type) mounted on each vehicle. From there, tablets or other wireless devices could connect to the router. When the vehicles are in the station, the routers would auto-connect to the WiFi network to receive calls for service and other updates. When out on a call, the router would form an ad-hoc network with other vehicles on the scene. If a vehicle came into range of an Internet 'hotspot,' it would notify other vehicles and become a gateway for the rest of the 'ad-hoc' networked vehicles. I've looked at Freifunk for this, but would like some other options. Recommendations please?"

Comments

Get ready for it!

[Frosty Piss]

Of course at least 1/3 of the posts will try to knock you down with the blather "if you have to ask, you're not the right person for the job".

Re:Get ready for it!

[fm6]

In this case, that seems a valid criticism. Messing around with technology you don't understand is a harmless, and even educational, pastime for the hobbyist/hacker. But when lives are on the line, a more conservative approach is called for.

Re:Get ready for it!

[HangingChad]

But when lives are on the line, a more conservative approach is called for.

You know nothing about being a first responder, especially out in rural areas where radio coverage may be spotty to non-existent.

Lives are on the line whether you have working comm or not. There were times I would have settled for two tin cans and a string if I could call for mutual aid on it. During emergencies ad-hoc networks could be a lifesaver.

There is a big need for self-discovering networking between emergency response vehicles. You won't find any commercial solutions in the budget of most departments.

Maybe drag your fat butt out and pull some volunteer shifts before you start telling people in the field what they need.

Re:Get ready for it!

[mdfst13]

I think that you are framing the question incorrectly. The question is not if it would be better to use a more robust networking method. The question is if this would be better than no network at all. Is a network that only works 50% of the time better than a network that works 0% of the time? The original poster doesn't share the purpose of this network, but for many applications, sometimes working is better than never working.

Another issue here is that if they can build an ad hoc network that sometimes saves lives, it will be easier to then get funding for a more robust networking option that can consistently help them save lives.

It's also true that there are some applications where intermittent connectivity is worse than no connectivity. It's possible that this is one of those situations. You're right to call out that possibility. However, shouldn't we at least consider the possibility that the poster has thought that through and really would be better off with intermittent connectivity?

Re:Get ready for it!

[fm6]

Well, this conversation has convinced me that I don't know much about how first responders might use an ad-hoc network. They say they need it, and the rest of us can only nod and agree, though more specifics would be very helpful. But from what little I've learned so far, it doesn't sound as if a solution that breaks down unexpectedly is an acceptable one.

I'd also suggest that this problem needs to be attacked on a higher level. I know that there's been a big effort to create standards for emergency radio networks using common frequencies. Part of the justification for the move to digital TV was to free up analog frequencies for this purpose. But broadcasters have ended up holding onto their old frequencies even as they moved to new ones. Obviously this problem needs some serious leadership.

Re:Get ready for it!

[AK Marc]

If you have working radio, what good would a WiFi connection be? What is added to the response if you have everything working, WiFi, radios, and mobile phones?

I still don't get what a WiFi connection would do to help. None of the fire trucks here have computer on them, other than what's running the radio and the engine. Most of what a pump operator does could be replaced by a computer and do a much more efficient job (juggling water flows and pressures), but would never be used because of reliability and safety. A human will think before taking an action. If you ask for 1200 kPa on hose one (properly calculating it for the length of the hose to get the 700 kPa required at the branch), The human will not only fail to comply (as a computer should as well), but tell you why, and make sure to tell the OIC that you are an idiot in need of retraining.

Project Byzantium?

[Intrepid imaginaut]

http://project-byzantium.org/

I have to wonder though, what's wrong with good old fashioned radios.

Re:Project Byzantium?

[slimjim8094]

I am an emergency responder, and frankly I can't come up with much I'd use internet for. Medical history databases? Like what? Even at the hospital they need it sent from other hospitals if they don't already have it , there's no world-wide database of medical history, and even if there was can you imagine the nightmare of hooking every EMS agency up to it? And the security involved in handling patient data on such a scale? No thanks. In any case, it doesn't do me much good. Either the patient can tell me their medical history, or we've got much bigger problems. If they're unconscious, their history is secondary to keeping them alive, and you've got plenty to do on that front.

As for conferencing with doctors - that's crazy. We already have medical directors (physicians) we can call on the phone or over the radio, and it works fine when we need it. Plus, it won't give them any more information than what you can tell them over the phone anyway. I don't much want to fidget with Skype and a webcam when we're supposed to be deciding on a course of action. They can't interact with the patient anyway, and crappy wireless webcam video wouldn't be sufficient to notice something subtle that we missed.

The paramedics that I work with have CAD for tracking status, location, nature, etc - but they don't use it past dispatch. They can also send telemetry (specifically EKGs) from their monitors via their cellphones to the receiving hospital, so a heart attack can be diagnosed from the trace before we get there. That's pretty cool, but it's about the limit of what we've ever felt like we needed.

Re:Project Byzantium?

[AK Marc]

You don't need a general purpose computer to get to the event. A GPS (non-networked, assuming it has updated maps) is sufficient for that purpose, and if you want something for one and only one purpose. a PC will make it harder, in almost all circumstances (longer boot, less reliable, more confusing).

The description was more of something to use after arrival, which there is no process that would be enhanced by computers.

In typical slashdot fashion, he's looking for a technical answer to a question he doesn't know. The question should start:

If tech were free and every appliance had a 10 Gbps fiber run between it and the station, what would they do differently? If the answer is "nothing" then there's nothing to fix. If there is an answer, then the next question should be something like "why aren't they doing that now?"

Re:Project Byzantium?

[girlintraining]

I know few places have such things now, but it's happening, gradually. Try to be forward-looking.

Yes. Poster would like you to experiment with configuring a wifi mesh instead of saving his life. He'll understand because he wants people who care for him medically to be "forward-looking", not "prudent."

BS. If you could have at your fingertips their recent medical history, current medications, etc. on the way to the site, you would be much better prepared even if they aren't responsive.

An EMTs job is to stabilize your vitals, not to diagnose and treat your condition. They don't need to be prepared for anything except keeping you breathing, your heart beating, and, since you're unconscious in the above scenario, not much else.

But if you COULD have a doctor there, without messing with Skype or a webcam, would you think that's a bad idea?

The doctor is at the hospital, treating the other patients who may have life-threatening injuries. You're suggesting the doctor step away from those duties to help the EMTs perform... basic triage?

Don't take this the wrong way, but I think you're being shortsighted. You are fixated on what the current systems do for you, but you don't seem to be very receptive to what improvements in the technology COULD do for you.

He's fixated on the only thing that matters: Keeping the patient alive. Who the fuck cares what systems he uses? Unless they contribute to Job #1, they're worthless. I don't want someone googling "bleeding to death" or trying to skype or webcam to someone else to tell them what to do when I'm taking the ride, I want them trained in keeping my ass alive until someone with the right qualifications to fix whatever put me in that ambulance can see me.

You're coming at this from the perspective of someone who's spent too many years in technical support -- treat the EMT like he's some kind of moron or puppet, to be directed about by the guy on the other end of the line. Medicine isn't like that. They work as an integrated team, and they depend on their training and experience, not their google-fu, to do the job.

Re:Project Byzantium?

[Will.Woodhull]

No, parent post is dead wrong.

When a series of lightning strikes begins a complex forest fire, many of the first responders are separated from the guys who might be starting to do the hand drawn maps by miles of flames and roadless terrain. The responders are coming in from all directions; there is no central point that they are going to funnel through. In the Bilger Creek Complex of fires in 1987, a number of the first responders were woodsmen with bulldozers and chainsaws in the middle of it all, who had to guess where to set up fire lanes since although they were all in radio contact with each other, at that time there was no way to communicate where the fires were or where they were heading.

Lives were lost. If it had been possible to cobble together a mesh and communicate even rudimentary map data between the groups, there would have been fewer losses, and probably the fires would have been contained more quickly at less cost.

Tornados and hurricanes present similar situations: the first responders come in from all the margins. There is no central point they would funnel through. From what I have seen of earthquake news and when Mt St Helens blew its top, the same is true of tectonic events.

Generally, it is probably true that the first responders are on the ground, doing the work, before the map makers have got their magic markers out of the box. It would be very nice if these guys who are already in the thick of it could become a part of a mesh network and benefit from the second wave of responders who are building the maps and beginning to coordinate activities that are already in progress.

Internet connectivity would be useful to such a mesh, but even if that is broken, the ability to send low quality photos, penciled diagrams, rough maps, and plain text lists between the first responders and the first coordinators would be damn useful.

easiest solution

[starblazer]

While ambitious, this is the wrong path to go down. Great for hobbyists, but is NOT what emergency services needs. Emergency services needs reliability. If your department can't afford a few mobile broadband units, you should seriously look into throwing a couple more raffles or asking for more money from the city/county/township/state.

Re:easiest solution

[TwineLogic]

This post makes a very good point. What is going to happen to you, personally, when a call is lost for some reason, a person dies, and a lawsuit follows?

Re:easiest solution

[ThatsMyNick]

You mean they have no money? How do they get firetrucks (which are very very expensive, if I may add), and everything else need to support their volunteer fire dept? The mobile devices are really really inexpensive. I have one from ting, the device cost me $70 (a bit more expensive than a wireless router, but not much), and for my usage I pay $15 a month (and all devices get pooled into the same plan, and will share data, making it very cost effective)

Re:easiest solution

[AK Marc]

For most first responders, that's not true. They are protected while on the job. http://en.wikipedia.org/wiki/Jose_Guerena_shooting Some ex-marine got his gun while the police were staging an unnanounced home invasion (he was on edge because a relative recently was killed in a home invasion). He never took his gun off safety, but the police fired 72 shots at him, hitting him with 22 of them and killing him. He never fired a shot and the initial reports all indicate the police claim he fired shots first. Yet, not a single charge was filed, even for the guy who shot blindly over the shoulder of another cop, who was in no danger and couldn't have seen or heard any shots from within. And here, firemen have protection under law from just about any liability, criminal or civil. Short of raping an unconscious smoke inhalation victim, I can't think of anything that could get a fireman in trouble.

Re:easiest solution

[DarthBart]

My old department got hand-me-down trucks from the Texas Forest Service that we converted to brush trucks. We also got a 5000 gallon flight-line fuel truck from them that we converted to a mobile hydrant. The one "new" truck we bought, we bought as a used truck from a department in Chicago and had to take out a loan to pay for it. The bi-annual fundraising BBQ we held covered operating expenses, but that was just about it. Everything else came from handouts from the government.

The radio system? Patched together with stuff my dad & I bought at hamfests.

And there was more than a few times during the summer and we were fighting multihundred acre brush fires that I wish I knew exactly where each truck was, how much fuel and water they had onboard, and be able to set a waypoint for them to drive to for their next task.

That may not have all been able to have been done with an ad-hoc wireless system, but that would have helped immensely.

What is it for ?

[SomethingOrOther]

What is it for?

No really......
You have told us how you *think* you want to communicate, but not what information you are communicating.

The first step of any IT problem is to adapt your ideas to fit users needs........... not adapt users needs to fit your ideas.

Re:What is it for ?

[fm6]

Actually, that's the second step. The first step is to find out what users need. This step is often skipped, alas.

Have a look at Mikrotik

[Jimbookis]

http://www.mikrotik.com/ devices might have what you want. They are inexpensive, very flexible and have interesting mesh modes I have yet to try out and will run directly off your fire engines battery system with some power filtering and clamping. Whatever you do in general you should have a play, write a clear specification with all sorts of test cases and run a small trial for a while. Make the devices/solution meet your requirements, not the other way around or you will be sorry.

Good ol' HAM radio?

[guruevi]

It works well, it won't give you much throughput but if all you need is some text and voice-based systems this should be plenty (it's about 300-9600 baud for IP so a slow serial link).

The issue I see with your approach is that when the vehicles are within range of each other they will also be within range of the same hotspot. So mesh is simply overkill. Mesh is intended for lots and lots of nodes in dense areas to connect to each other to a single (large?) uplink for either anonymizing or places where you cannot place (either due to economic or ecologic reasons) multiple antenna's. This works well for the GSM range because they are intended to cover literally miles (2W) at a frequency that is licensed to cooperate with each other and able to penetrate a lot of structures so two cell phones can technically talk to each other and extend the range of the original tower another mile or so (given the battery usage to do so is acceptable).

The 100 mW you get out of a WiFi router close to the hydrogen resonance frequency is simply not enough to cover a mile of random area which may have other compatible and incompatible broadband sources (microwaves, garage door openers, bluetooth ...) that could overpower the signal.

You're better off using the professional systems for this. WiMax base stations can be had for $1500 and a receiver is ~$200 and it will cover about 50km. Otherwise get a free cell phone plan for your volunteer fire department (I mean, some local corporate overlord MUST be benevolent enough) or set up your own transmitter (HAM or otherwise).

Re:Good ol' HAM radio?

[drwho]

This is not appropriate for ham radio. Ham is supposed to be for amateurs and emergency comms only. Emergency comms as used by an auxiliary force. It is not to be used for commercial purposes, military purposes (even though tthe military acna overrid this and do whatever they want, it would be frowned upon), and not for police/fire/rescue. Seriously, the emergency services have enough equipment and bandwidth of their own, they shouldn't be trying to compete with all the signals on 2.4 ghz and the ham frequencies. They shouldn't be using crap equipment either.

This should be obvious - emergency services communications need to be reliable - not the ad-hoc stuff life mesh and ham radio that works when the sunspots are in the right position. That means VHF and big heavy radios, high bandwidth gear, that has been proven reliable and yet costs a lot of money because of it. There are no rusted out old pickups being painted red and used as fire trucks. Same principle applies here.

Since it's been posted before...

[sbrown7792]

Here's the article from less than 2 weeks ago about the same thing. They had a few interesting ideas.

Consumer grade routers? Not a good idea!

[toygeek]

While I applaud your willingness to use technology where you see a need for it, the consumer grade routers just aren't up to the task. I've seen routers die simply being moved from one side of the desk to the other. All it takes is a cold solder or a flaky chip and *poof* that router is history. You'll be troubleshooting weird problems constantly and will be replacing routers just as often. If your solution depends solely on these routers, then I think its not much of a solution at all.

3G

[Great Big Bird]

What about using some level of 3G wireless access that can be low bandwidth attached to other hardware?

Hammers and Nails

[abarrow]

Yeah, this definitely feels like a case of "When the only tool you have is a hammer, every problem looks like a nail".

WiFi meshes like crap. Your first responders will spend valuable time just trying to get their devices to work. While your volunteer situation is well understood, and your budget is probably pretty low, don't ask people to depend on consumer stuff for this sort of thing. A trunk radio system (and one that is not too highly loaded) or something similar is highly recommended.

VANET

[ArhcAngel]

So you want to set up a VANET (Vehicular Ad-Hoc Network)...a subset of the MANET (mobile ad-hoc network). There's even a proposal for a secure fire truck communication protocol via VANET. Perhaps you can find more information by reaching out to some of the agencies working on this protocol.

Saw a presentation on MESH for emergency response

[bdwoolman]

Keven Whipp of the Montgomery [County Maryland] Amateur Radio Club gave a presentation last April on a very similar topic to the Columbia Area Linux User's Group (CALUG). The radio club has been working with Montgomery County to test various setups of MESH networks on Linksys WRT54GL routers running custom firmware to be used in emergency situations. They have been testing distances and reliability using different frequencies using high gain antennas (which require a license). As I recall the deployments they tested faced a lot of technical and regulatory obstacles. And they were looking at simple static deployments, not mobile. If, say the infrastructure went down after a flood, their objective was to provide basic internet services to Emergency Response Teams working in the area.

Anyway, here is a link to a PDF summary of the presentation. My take away was that even after pretty extensive testing the system was not ready for prime time, but was very promising. To be useful in the situations to which they aspired the Mesh had to be reliable and robust. It was not. I am sure they would be happy to share their experience with you. And I bet they made progress over the summer.

How is this going to be more useful?

[NimbleSquirrel]

What exactly do you need this system for? Seriously, if you are needing internet access to save people, there is something wrong. Sure, there may be a situation where an Incident Commander may need to look up something like an MSDS on a hazardous material, but in that situation a tablet with 3G access is all you need.

If you're wanting this for comms, then you really need to think again. For Emergency Services, any comms system need to be robust (ie. not built on cheap consumer grade hardware), reliable (ie. able to work when parts of the system fail, and it must be easy to fix or replace) but most importantly it need to be able to work with the systems of other Emergency Services. If you go ahead and do your own thing, it could potentially cripple your response capability. If your Fire Department was first to attend a Mass Casualy Incident, would you be absolutely sure that your system, built on 'cheap wireless routers', would be 100% effective? Would you be prepared to stake your life, the lives of other firefighters, and the lives of multiple casualties on this system working? If the answer is not an absolute yes, then walk away now.

Don't get me wrong, it is a cool idea, but it is not something that you or your Volunteer Fire Department should be looking into as a deployed solution. You cannot go from "hey, this sounds cool" to putting it into operational situations without doing some serious research and development as well as thorough testing. This may seem over the top, but this is for an Emergency Service: people's lives will depend on this.

Not appropriate

[drwho]

I've been doing wifi mesh networks for over ten years. As much as people try, these just aren't reliable or secure enough to be used for such things as military and emergency services networks. Emergency services have more radio spectrum than they know what to do with, and access to lots of other resources. Use technology which is appropriate to these advantages, taking into account the demand for very high reliability.

There are standards for this.

[Animats]

It's not standardized across the US, but many states have standards for emergency radios. Find out what's standard and go with it.

One of the more useful projects of Homeland Security is to get all the agencies that have first responders connected in emergencies. It's hard, because each agency has their own system and they don't interoperate. Here's the Texas plan. And the Florida plan.

Most of the hard problems have to do with too many people on the air in urban areas. If you're a volunteer department, you're probably not in an urban area and don't have that problem. If you want something that will Just Work, get high-powered 700MHz public safety band capable VHF FM handhelds and vehicle radios for your own people and get them fitted into your state plan. A few Iridium satellite radios for command personnel and those who really need to talk to the outside world during an incident are helpful. Here's one suitable for fire truck installation. Iridium airtime costs are high, about $1.29 per minute, but in an emergency that's the least of your problems.)

The Serval Project

[complete loony]

The Serval Project on the Android Market.

Our focus is on providing useful services without any reliance on fixed infrastructure. Phone calls and text messaging via adhoc mesh, and even file distribution in the field.

Though you might find our next release more suitable than the version on the market. It's still in heavy development, but would also allow phone calls to be relayed to the PSTN via an asterisk PBX. We'd be happy to provide an alpha version and help you to get the most use out of it.

We're also working on a separate application that uses open street map data for situational awareness and collaborative mapping.

A few thoughts

[KenDiPietro]

I can see several applications that would make this type of network incredibly useful. Having the ability to distribute situational awareness video in real time would be awesome. This could be useful independently from internet connectivity and a tablet with a decent amount of storage could keep the video for later review. If it were within the budget, wouldn't a head's up display in the firefighters helmet of something built into the brim of a law enforcement officer's hat be pretty slick? The ability to Wifi locate any of your team could also be quite useful. Perhaps more to the point are disasters like Katrina or 9/11, where the telecommunications network may be down for extended periods of time. This kind of backup network could very well be the difference between life and death. If enough of these radios could be dropped in place with solar and battery backup as entire area could be brought back online in a very short period of time. Year's ago there was mention of a completely independent group of license exempt wireless pros forming an instant adhoc network on the the upcoming anniversary of September 11th to drive this exact point home. As a suggestion, you might want to look if the 4.9GHz band in available in your location. Among others, Motorola's Motomesh uses this band. It's clean, licensed for this use, and some Wifi adapters can be switched to work in that band, lowering the cost for equipment. Certainly, challenges exist and need to be worked through but if the opportunity to experiment is presented, why not go for it?

We are trying HSMM-MESH

[tipo159]

Our local club is playing with HSMM-MESH to supplement our existing ham radio set-ups (two repeaters and an "assigned" ARES/RACES-type simplex frequency (in the last regional drill, hams in the next county were demanding that we get off "their" frequency, which is why "assigned" is in quotes)). Some times it would be helpful in an emergency situation to be able to transfer files or stream video and Wifi speeds are better than TNC speeds.

A bunch of us have purchased a bunch of WRT54Gs and reprogrammed them, but we haven't yet tried to get them to mesh.

Bad idea

[slimjim8094]

First of all, let's imagine the technical stuff isn't an issue. Imagine all the trucks have cellular modems and can just communicate over the Internet as usual. What are they using it for? I do volunteer EMS (not fire, admittedly, though I work with fire agencies a fair bit) but I can't figure out what it'd be used for, aside from CAD (computer-aided dispatch) which is outside the scope of our volunteer agency, and likely outside the scope of yours. Large-scale incidents (MCIs) do require a lot of information sharing that might be well-served by a data network of some sort, but interop is already a huge problem just with bog-standard FM radios. What sorts of computer data are they going to share without the internet? Keep in mind, this has to be data that's either already available digitally (in which case, why the network?) or created on-scene and then digitized. I can't think of any, honestly, except for perhaps pictures? And they're not really necessary.

So assuming there's a use for wireless data sharing, what justifies not dropping a few hundred a month (which is nothing for even a volunteer FD) on cellular broadband? It's a mature technology, reliability is high, and it doesn't require any customization - just logging into a VPN or something, if even.

Finally, your solution can't be finicky or unreliable at all. If it doesn't work once, it'll become a liability and nobody will rely on it. People don't screw around with stuff like this, since it can literally get people killed. Public safety has been doing fine on voice radios for a long time, and even if it could be done better, there's no hesitation about giving up your enhancements permanently the first time there's a problem with it.

I'm an amateur radio operator. I get the attraction to playing with this kind of stuff. But I'd never use it in my EMS agency, since "playing" isn't acceptable. That's why we buy $1k-a-pop Motorola radios that do less than my $100 Chinese HT - because there's no fidgeting, and no question about whether it's going to work when you need it. Even if you've dropped it in a puddle, or it's gotten dropped from 6 feet onto pavement, or used it to clobber a drug addict away from you (yes, it happens).

Peer to peer cellphones without cell service

[Sussurros]

There is an Android app developed by a mob of Australians which allows cell phones to talk directly to each other in emergency situations without using any cell service at all. Sorry I have no name either for the app or the developers. It was developed last year, released this year, and is intended for bushfire teams and search and rescue operators. I'm guessing it won't be in the app store and I know it works by dialling the number you want and if both phones have the app and are within range then they connect. I'm sorry I can't provide you with more information but it's just something I read about in the Sydney Morning Herald (www.smh.com.au) some months ago and it stuck in my mind because the main developer spoke at length about the incredible problems of getting the protocols to work.

Re:Easier approach

[icebike]

You can't build a mesh until you have some place for the mesh to connect.
If you are out in the sticks so far that you have no radio coverage, there is not likely any nearby mesh members either.

I can see where being deep in a ravine with your home base transmitter on the other side of a mountain might present a localized situation which might be solved if you could some how get a mesh partner on top of that mountain.

But knowing how long it takes to get something working it would be easier to send a guy out in a support car to some intermediate point and simply relay data back and forth until air assets arrive.

Does someone living near by have wifi you could tap into? Maybe. But is Grandma going to have a clue how to turn off encryption or even what the password is when you wake her by pounding on her door in the middle of the night?

Satellite phones can be had for under 700 bucks, and an annual satphone plan starts Under 500/yr. I would recommend radio mapping the service area and determining those areas where there is no radio service and equip at least one unit in that are with sat phones.

But before you go to that expense take that cell phone out of your pocket and see if it works in these areas. If it does, your mesh is already in your pocket, and just forget about ad-hoc anything in an emergency.

Commercial grade...not hobby grade

[ff1324]

Disclaimer: I work for a paid fire department but started as a volunteer. I understand the financial challenges for both m

Our metropolitan area uses InMotion OMG1000 or OMG1050 mobile routers. The cost just over $2000 each, but they create a man network for emergency equipment, have up to four different WAN connections, are remotely configurable and upgradable, and provide GPS services. They're almost bulletproof.

Wiring up WRT54Gs isn't the answer. You'll spend more time than your time is worth creating a solution than finding funding to implement one. Apply for a FIRE grant or cooperate with multiple agencies to access a larger program. Check with your county or state.

Realisitcally, an air card is all you need otherwise. If enough agencies want in on the project, see if the wireless provider will put you on your own APN.

Good luck.