I'm just playing around with setting up a "mesh" wifi access for my entire neighborhood, and would like to discuss theory if possible.
Am I right in that using a router as an AP only (or bridge) would cut the bandwidth in half?
And, for experiments sake, I'd like to get an idea of what kind of hardware to buy to allow access for, say, 100 homes, in a 6 block area.
I know I'd need to convert the AP for outdoor use (I'd like to play with cheaper consumer stuff), which I can do using an oudoor box and power over the cat5, right? Although I'd probably design a conduit to carry the signal and 12V seperately for easier maintenance. Any idea the wattage necessary? I'm thinking of using PSU's from scrap computers.
Then, how about the repeaters? I need the same brand usually, correct?
I know about the LOS issues, and plan on doing all roof mounts, and USB client adapters. Also, any idea of the bandwidth necessary and how to limit it to each client? (100 users, typical home use).
Why use a mesh? The only real benifit to a mesh network is the cost savings on the wired backhaul to the ISP. Since you're probably only going to have one or two backhauls, there's not much savings. Instead, you get the deal with store and forward packet repeaters, that eat airtime, mutual interference problems, bandwidth bottlenecks, complex routing algorithms, and performance issues. More specifically, you only use mesh when you absolutely must use mesh.
Light reading:
Reality. Be sure to read this one carefully, especially the part where there's typically 50% or less probability of delivering a packet intact, and the reliability even at 1Mbit/sec.
The commercial version of Roofnet:
Not quite. One hop will cut the *MAXIMUM* thruput in half. Let's pretend that you're very lucky and get a 36Mbit/sec association (that's a wireless connection). Maximum thruput is about half or
18Mbits/sec for TCP (assuming no interference, etc). However, if you shove that through a mesh network or store-n-forward repeater, it gets cut in half again to perhaps 9Mbits/sec. I say perhaps because it's usually more than half if there are a number of mesh nodes nearby. If you add a 2nd hop, it gets cut in half again or 4.5Mbits/sec *MAXIMUM* TCP thruput.
What you need is a good overdose of reality. Setup a mesh network in a closed room, where everyone can hear everyone else. A dedicated repeater or WDS bridge will suffice. Now, try to measure performance from end to end using IPerf (instructions on request or Google this newsgroup for my posting on IPerf). If you turn off the repeaters, you should get something near the maximum theoretical performance. Turn on the repeaters, and watch the thruput drop radically along with a corresponding increase in retransmissions and errors. Try it.
Ummm... 100 customers? Every time it goes down, you want 100 phone calls? Have you perhaps been snorting, injesting, or smoking controlled substances? Never mind the technology. Do you think you can handle the complaints and service calls? Whatcha gonna do when some clueless user comes home with a virus or worm and eats all your bandwidth? How about the not so trivial problem that customers will tend to call you first, before calling their computer guru (who charges money), simply because you're cheaper? Do your really need this headache?
Anyway, the loading is easy: 100 light email and web users 10 business users 1 file sharing user It doesn't matter how many access points or backhauls. One user can hog the whole system unless you have monitoring and traffic management in place.
Ummmm.... How big is a "block" in feet or meters? Is that a city block or country block? Tall apartments of single family suburbia?
Right. It's called PoE (power over ethernet) or 802.3af. There are vendors that will sell you the complete package, or just the pieces. For example:
The delivered power is limited by the resistance of the power wiring. That's why PoE is normally done at 48VDC, so that the effects of the wire resistance is minimal. Junk power supplies are only going to supply 12VDC, make a bunch of noise, probably smoke the cable if you short it, and are not terribly reliable. Look into real PoE adapters.
Repeaters are not very well defined in the IEEE 802.11 specs. The result is substantial incompatibility among vendors and versions. That's one nightmare I suggest you avoid. Repeaters are also a basic component of mesh networks, which I previously suggested are a bad idea.
Keep planning. You'll find that USB is limited to 16ft maximum cable. You can get amplifiers that will go farther, but then you have to supply power at the destination end. Use ethernet and PoE instead.
Look at the real bandwidth managers:
and see what you might need. Also note that it's not enough to simply configure a bandwidth manager and let the system free run. You gotta monitor the traffic to look for abuse, changes, hackers, idiots, and failures. You'll need to know the performance of each user and which one's are hogging the system. Check out various traffic monitors.
In case it's not obvious, you're going to need EVERYTHING that a typical wire line ISP uses, with the added enjoyment of an unreliable means of delivery. If you want to do this for free, that only eliminates the billing system. Everything else still has to be there.
Nope. I'm busy buying test equipment, updating up my lab, dealing with the ultimate messy house and office, loafing, and playing hard to find.
Well, ok... just one idea. Start small. Find a Linux based wireless router and install it on your roof (using PoE) as an access point. Deliver shared service to a few friends and neighbors. Keep it small so that when you do something stupid, you don't have a torchlight parade of irate neighbors banging on your castle door. Learn system admin, bandwidth management, and abuse mitigation on a small scale. Once you've made all the basic mistakes, grow the system so that you can move on to making bigger mistakes.
Well, nice links, Jeff. One more question, please.
These guys:
formatting link
Seem to already make what I had in mind. Seeing as I don't have to re-invent the wheel, are you saying that this (apparently a mesh for hobby types) has the inherent problems you just mentioned?
So, the result is what I want, but I should consider another design? What do you have in mind? (gotta be wireless for the ususal reasons).
Maybe a simple POE with long shots? Or what?
(BTW, this is going to be a lab experient similar to the Roof Net guys....)
Yep. Overly simplified, you have several problems:
You're building a WISP (wireless internet service provider) but are not very aware of how much work that really involves.
Mesh networks can be made to work and certainly have their place. Servicing non-paying neighbors is not my idea of one of those. Others (such as Earthlink, Meraki, and various manufacturers will disagree).
Delivering reliability and quality service via Wi-Fi is a dubious proposition mostly due to self-interference.
Mesh systems do not scale very well. The problem is that you discover this only AFTER you spend considerable money. Anything can be made to work on a small scale.
Also, I can tell you did NOT read the MIT Roofnet article or you would have some questions regarding thruput and reliability limits.
Well, you could just join Meraki and try their system. It's not that expensive and certainly cheaper than a do it thyself system. If mesh is what you want, you'll certainly get a clue on how it works very quickly. If that's too much trouble, go unto the Google (free) Wi-Fi complaint and help message board, and see what manner of problems a real mesh system can generate. Remember, as system operator, owner, and chief technologist, *YOU* get to answer these type of questions.
As for what I had in mind, I'll answer your questions after you answer mine. For example, how big is 6 blocks in feet, meters, cubits, or furlongs? That's not intended to irritate you. It's because wireless networks are designed to operate in specific environments and topologies. What's suitable for a small coffee shop hot spot, is not suitable for "illuminating" an office building or apartment complex. Give me some numbers and I'll gladly spend your money for you.
Dream on. Sounds like you're thinking of an omnidirectional antenna. Yeah, that might work for starters, but has some rather severe limitations, especially when it comes to minimizing interference. I'm a big fan of "sector antenna", where you have multiple access points and multiple 90 or 120 degree sectors per access point. However, I can't tell if this is suitable for whatever you're doing.
Yep. Learn By Destroying (tm). Nothing like repeating their mistakes on a small scale. I recall the definition of insanity was doing the same thing multiple times and expecting different results. As I mentioned before, you're problems are NOT going to be technical. They're going to be administrative and system management. Simply identifying the cause of problems, separating them between user and system problems, and doing something about it at 4AM, is going to be a major challenge. That's why I wanted you to start with a small scale system. You'll get all the experience with irate customers immediately and will not have to wait until the system grows to get the experience. Also no need for a mesh with a small system. I have a neighborhood wireless and wired LAN that I run. I often regret doing it simply because of the weird problems that tend to appear at odd times. For what it's worth, I tried to use old 802.11b junk radios and am now replacing them with much better and more reliable
802.11g equivalents.
There are other mesh networks out there being used. We have one locally:
Some light reading from my bookmark mess:
An Experimental Scaling Law for Ad Hoc Networks. (Also applies to mesh)
Doing mesh the right way, using multiple radios per node:
Here's a good start. How many wireless connections do you think a typical WAP can handle before something pukes in the firmware? It's not as many as you think. See list at:
Because you're the guru around here. Plus, you let me pay your consulting fees in antique Japanese relics. (old stereos, tube radios, wrecked Datsuns, etc.)
Oh, I read it all right. Just some things I may get a bit later. Plus, I'm the persistent type.
I'm not irritated in the least. To answer your questions, the mileage is exactly 584' 7" on one side, and 956' 2" on the other side. A slight rectangle. LOS clear from the roofs. Mean temp is 75*F, no snow, no freezing in winter. An earthquake once in awhile, not much else in the way of topography to talk abut. I'll start with 25 or so clients, and thanks for the link to max supported per router.
No, I have the admin all figured out. I'm not going to do any:) It's gonna be free, and free means no support. Later, I'll just jerk it all out.
AS you say, wifi is not designed for some applications. However, the need, especially in third world places is enormous. I'm thinking that off the shelf parts may not be perfect, but eventually they will. Remember, a computer was not designed for long distance phone service either, but it works reasonably well. Most technology grew from the tinkerers, me thinks.
So, yeah, it may not be optimal. What really is in this world? Windows Vista? (sacarism dripping........).
Thats even worse. From my experience when you undervalue or give away your services people don't appreciate it. And since everything is free they will expect lots of extra free stuff. You are in for a real experience about human nature if you become the free ISP on your block.
Naw, I have all the experience with human nature I'll ever need. I'm already despised, so a few more won't hurt my chances for re-election.
My plan is to just offer it on a trail basis (that means not very long), to see how they like it (means I want to see how it works), and becasue it's USB, I'll just confiscate them and give them their deposit back:)
That's what I was going to say. I don't charge my neighbors although I do expect some help when it comes time to do some digging or tree climbing. All are appreciative, considerate, and tactful when they call me on the phone at midnight, on a holiday weekend, to ask if the system is down. Never mind that I have a local web server with traffic and status information for them. Never mind that I showed them all how to use ping to check for connectivity. Never mind that most of them have more than one computah and can easily check the 2nd machine. Never mind that it's a brand new laptop the kid got from the grandparents for skool and is now desperately trying to get it online so he can impress his friends at skool tomorrow. Too bad it's not in my MAC address filter table so it won't connect. So, at midnight, I get to do sysadmin for a new computah with the help of a kid that can't read the MAC address off the bottom of his laptop without getting at least two digits transposed.
The theory is that if one is very polite and respectful, offers the appropriate compliments and familiarities, one can get away with just about anything. That's amazingly true except that most wireless users haven't figured out that this method of getting free support becomes really old after a while.
Two days ago, it was early in the morning instead. I had apparently lost the wireless link to one of the neighbors. Everything at that end of the neighborhood was dead. I spend an hour preparing a replacement bridge, drive over, and find that their refrigerator is sitting on the deck defrosting directly in front of their wireless antenna in the window. I make a mental note to move their antenna.
It goes on and on like this almost continuously. If I had 100 neighbors connected with wireless, it would probably be a full time job dealing with the questions and problems. I was in the hospital last month for a few days some luser called with a computer question. I was too drugged to answer or remember the conversation.
I've heard the zero support theory of free networks before. I have yet to see it work or last. That's why I suggested that one start with a small number of users. so they can get a feel for the support and admin load.
God, grant me the serenity to accept users that will not change; the courage to educate those that are willing and able; and the wisdom to spot the difference before trashing the whole day.
If you have time to spare and want to contribute to the community why not help folks who really need it and will appreciate your efforts? Example, do stuff for people who are homebound or get involved with a reputable charitable organization.
On 2007-09-03, Jeff Liebermann intrigued me by typing:
Jeff, those of us in the computing business know very well the frustrations that comes with trying to correct someone else's mistake. We love and hate it at the same time and despite the fact that we tell our clients how to do certain simple troubleshooting tips, very few actually listen and try them. Trust me, sir, I feel your pain. Now, for the person that called you while you were in the hospital, now dang, that is just downright inconsiderate but typical of today's mentality. Did the...as you put it...*luser* even ask how you were doing?
It's frustrating, but I've become tolerant of the situation. There are those that simply are not going to learn anything new and happily throw money at me to do it for them. I call them customers. Some of these are justified. One rather wealthy individual mentioned that his working time was worth perhaps $1,000/hr. Why should he waste it doing a job that I can do better and faster for only $75/hr? I also help out at a senior retirement home, where many of the residents are borderline senile. Others have simply not had the experience of having to learn anything since they escaped from skool. I've often been tempted to yell "WAKE UP" at them, but have learned that it's often too late.
One observation is that the ones that are really trying to learn, usually take notes. Those that I'm wasting my time educating, usually do not. One of the locals has an IQ of perhaps 80 with the added bonus of dyslexia. He realizes that he must compensate, so he carries a digital voice recorder around with him to help remember things. I was really impressed when I saw his computer full of rather well organized audio clips with some of my rants on different topics. He drives me nuts, but I can't fault him for trying.
Ummm... I don't remember if they asked. I don't even remember who it was. All I know is that the nurse mentioned that the person I talked to earlier in the morning after surgery called *AGAIN* to ask if I was available. Excruciating pain has a way of reducing my curiosity, so I graciously declined. I never bothered to find out who it was.
Incidentally, I apparently had a rather lengthy conversation in the recovery area with the surgeon, none of which I can remember. One of the nifty things about modern anesthesia is that there's no weird dreams between pre-op and post-op. One minute I was lookup up at the IV bottle wondering when it would start working. What seemed like seconds later, but was actually 18 hours later, I was awake in a hospital room. I also suspect that there were some psychotropic mood enhancers mixed in with the cocktail, as I felt minimal pain and quite uncharacteristically optimistic. Of course, that didn't last as the drugs wore off, but it sure was great while it lasted.
Speaking of sneaky, one of my customers called me erratically asking how I was doing and offering advice, assistance, and sympathy. I'm sure it was genuine, but she always managed to sneak in a computer problem or question afterwards, which made me rather suspicious.
Also, I was posting answers to this newsgroups almost immediately after surgery. What wasn't obvious was that I was doing it rather differently. I would write a suitable answer, save the answer, and not post it immediately. A few hours later, I would re-read my answer to see if it was reasonable coherent. Usually it wasn't and required a re-write. I would repeat this ceremony until it seemed suitable for public consumption. I had to abandon several messages as no amount of re-writing would have helped. A few with problems managed to sneak through anyway. Remind me not to write anything while drugged.
Look into a low cost wimax base station, or even a wisp type of base station. As for the mesh, we haveused meraki at some of the camps around here and they work ok
As well as CALEA compliance...which doesn't make sense. If someone wants to conduct activities, they'll simply use free Wifi whereas a "for fee" WiFi has to have a government mandated wire tapping process in place.
Just to comment on this, you really DONT need to spend your life chasing abusers. We've been recommending per-IP controls for several years now. But using burst controls, you can identify those that are abusive automatically and implement effective controls that don't need to be monitored. As tunnelling and encryption become more and more prevalent in "abusive" protocols, you'll eventually have no choice but to monitor overall usage, so you might as well start now and save yourself a lot of headaches.
As an ISP you are a reseller of bandwidth. I've always has philisophical problems with ISPs deciding which protocols are good and which are bad. The truth is, that "good" usage is identifiable by the fact that its not a continuous use of bandwidth. Web browsing pulls pages and stops for a while. Even downloading movies stops after awhile. Defining a fair slice, and then setting controls that enforce that slice, is a generic way to manage a network without having to care what specific users are doing with the bandwidth.
Just to comment on this, you really DONT need to spend your life chasing abusers. We've been recommending per-IP controls for several years now. But using burst controls, you can identify those that are abusive automatically and implement effective controls that don't need to be monitored. As tunnelling and encryption become more and more prevalent in "abusive" protocols, you'll eventually have no choice but to monitor overall usage, so you might as well start now and save yourself a lot of headaches.
As an ISP you are a reseller of bandwidth. I've always has philisophical problems with ISPs deciding which protocols are good and which are bad. The truth is, that "good" usage is identifiable by the fact that its not a continuous use of bandwidth. Web browsing pulls pages and stops for a while. Even downloading movies stops after awhile. Defining a fair slice, and then setting controls that enforce that slice, is a generic way to manage a network without having to care what specific users are doing with the bandwidth.
Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.