Monkey Brains

That would create the ultimate single point of failure. For example, if the favored monopoly of the day built such a network on 60GHz, and San Francisco had an earthquake, most of the dish antennas involved would go out of alignment and out of service until a mythical army of technicians could be deployed to put things back together, assuming there are no aftershocks. Of course, motorized mounts that compensate for building or ground movement would fix the problem, but at this time, only cellular providers use them, primarily to avoid paying for expensive truck rolls and tower climbers.

Another problem is if you try to build a mesh network at 60GHz. Mesh requires omnidirectional or sector antennas. However, to get any kind of range out of such antennas, one would need to trade antenna gain for a wide horizontal radiation angle. That can be done, but the resultant vertical radiation angle will end up very narrow, probably on the order of a small fraction of a degree. Something like this

2.4GHz sector antenna: That means all the neighboring mesh nodes, with identical equipment, will need to be located at exactly the same altitude as your antenna. In my disorganized talk on the topic, I presented this drawing of a mesh network, and asked the audience "what is wrong with this picture"? The answer is that with a narrow vertical radiation angle antenna system, rooftops of different heights cannot "see" each other.

So, let's pretend that all this and related problems are solved by flying an aerostat balloon antenna over every building so that everyone's antenna is at the same altitude. That would probably work for a 60GHz mesh network. However, such a network needs to connect to the internet at some point. Let's pretend that your new offices are located fairly close to a major network wireless hub for the 60GHz wireless network. You would have great latency because you're only one hop away from the internet, but you also need to pass all the traffic originating from all the other outlying users. Your corner of the mesh would suffer from major congestion problems. (I'll spare you my rant on the futility of exotic routing algorithms and bandwidth management).

Think about what might happen if something fails. Is there a fall-back available? With the current tangled mess of different ways to connect to the internet, there are alternatives that can be made to work temporarily. When my home DSL fails, I point my wireless antenna to the neighbors Comcast router. Some people use their cell phone as a wi-fi hot spot. Others drive to the nearest coffee shop or library with a working internet.

I can rant on forever on the topic, but that should suffice for now. I suggest that you be careful of what you ask for. Just because one particular technology is superior in one application, doesn't make it superior or even suitable for all other related applications.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Most of those problems are mitigated by directional antennas and short-range, high attenuation links. And applying sheer bandwidth and signal processing.

Wireless is ideal for the last mile, or eventually the last 50 meters.

Our new security system is all wireless. That saved a fortune in wiring. We can add sensors and cameras ourselves, whenever and wherever we want.

Our telephones work over the microwave link, too.

We do have two land lines for the fire system. That's required.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

How far is that hop? How does weather affect it?

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

Not really. The common home nanocellular range extender will provide cellular coverage over a 50ft radius. If you want VZW coverge indoors, just plug one of these into an internet connection, and your smartphone now works. It's literally a small LTE cell site in a box.

Small Cell":

Femtocell repeaters:

The only real limiation on how small a cell can be made is financial, which is mostly in the cost of the backhaul. At the limited range of these devices, mutual interference is a minor problem.

--
Jeff Liebermann     jeffl@cruzio.com 
150 Felker St #D    http://www.LearnByDestroying.com 
Santa Cruz CA 95060 http://802.11junk.com 
Skype: JeffLiebermann     AE6KS    831-336-2558

Only to some extent; there are significant limits to such effects. And importantly they are site specific, can be difficult to predict, and (depending on the cellular standard) vary more or less over time. Here today gone tomorrow is not something a user want to experience!

One of the standard comparisons of how different modulation schemes performs in a cellular systems is to express their capacity in bits/s / Hz / km^2

The bits/s/Hz is obvious; the /km^2 adds in the effects of co-channel and adjacent channel interference.

Not really, no. It can push a limit out, but there will still be a limit. Given that over time bit/s will expand to fill the available bandwidth, it can only delay the inevitable.

But before the inevitable occurs, there may be some benefit.

Interestingly that is where the 60GHz ISM band might help. It is pretty useless for cellular systems due to the "oxygen hole", but that is a benefit w.r.t. co-channel interference.

60GHz is also nice w.r.t. making highly directional antennas; it is amusing to realise that ordinary thicknesses of plastic can act as anti-reflection "coatings".

OTOH, the last I heard (in the late 90s) there was a hard limit on the transmit power, defined by the transmitter transitor geometry and thermodynamics.

Not sure about weather, I'm not on it. Island is ~14 miles wide by 7 miles 'tall' . I think they have a few repeaters across the island as it is very hilly in parts.I guess it's affected by rain storms as is any other microwave link...No idea what the link frequency is but they use small dishes maybe 15" dia.

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus

We don't get lightning here.

If they fail, we can replace them. It's easy to get on the roof.

That's our EMI hardness tester in the fog.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

The power would be out too. Unless our building collapsed, I suspect our dish would still be aimed OK. A dinky little dish like that isn't super directive, and we have 170 dB to spare.

I'm thinking that the super-net backbone would be fiber, run to micro-towers every few blocks. Microwave could be used when fiber is hard to run. Some sort of Ricochet mesh could work a couple of hops in places where the aggregate bandwidth is moderate; upgrade as needed.

It's just that I have noticed that things keep changing, and most people assume that nothing will change.

Of course, motorized mounts that compensate

If a micro-tower can reach its customers, it can be arranged to hit adjacent towers. Beats trenching streets.

Cell phones work pretty well.

That can be done, but the

Or maybe nothing will ever change.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

This is it:

Amazing technology for $300.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Actually, they cost $70 each on Amazon.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

$298 is the list price. beware, this is the LHG 60G ("wireless wire dish") which is the 60 GHz model, not the original LHG which is 5 GHz but will not be able to do 300 Mbit/s fullduplex under realistic conditions.

Actually those MikroTik dishes are "self aligning". The feed has a square array of antennas and the optimal configuration is continuously determined. So there is no exact pointing required.

There also exists a bare feed version ("wireless wire") that is usable on shorter distances and actually uses the antenna for beamforming. I think it can sweep the beam over a 100deg angle or so. For the dish version, it is of course less.

Spectrum business service, and they are charging me $119 / mo right now for 100 mbits/sec down and 4 mbit/sec upload. The download is great, but I'd like to get more upload. I get a HUGE runaround from Spectrum about it, but mostly they seem to say if I go up to $175/mo, I can get 100/10. (These prices include one business phone line, too.)

So, I've been looking for alternatives, but in our area, there are VERY few options.

Thanks,

Jon

Ah, looks like San Francisco area, only. TOO bad!

Jon

I don't believe that an electronic device every 4cm all around us is workable physically or financially. It might one day carry a bit of the wanted data, but much it can't. The hard limit is physical, the practical limit any year soon is financial.

NT

only upto a point. 100x as many transmitters can only get you in the ideal case 100x the data throughput.

NT

Imagine sticking a tiny little thing to most everything you own, and always being able to find it.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

Even if it only covered 5' it's still a hard physical limit. I know it's hard to imagine wanting massive data throughput, but we've all lived through enough previous decades to know how that goes.

NT

rote:

rnet

=1

=1

ssic

hing*

ersus

rica

kable physically or financially. It might one day carry a bit of the wanted data, but much it can't. The hard limit is physical, the practical limit a ny year soon is financial.

To act as a data hub it would only work on every thing I own with power. Ho w many are powered at once? Dunno, could be over 100 indoors, not many outd oors. That's a hard limit. That limit won't bother us now, but soon it will . 300 baud didn't bother me at first.

NT