Slashdot Mirror
Electric Plug 14Mbps Spec Agreed On
Tei'ehm Teuw writes: "From this article
on EDTN the effort to establish a standard for
power-line-based home networking will take a step forward
this week when the HomePlug Alliance announces it will adapt
technology from Intellon Corporation for its specification.
The 36-member alliance will release a complete specification based on Intellon's
technology, with its 14-Mbit/second raw data rate.
In Europe, meanwhile, the HomePlug Alliance has established formal liaisons
with two groups working on power line
home networking: the European Telecommunication Standards Institute and the International Powerline
Communications Forum. Neither has defined a technology to date,
but it would be possible for them to adopt the same technology as the HomePlug Alliance, even though the European power line access technology is
different there than in the United States. (The European power line delivers 220 volts at 50 Hertz; in the United
States, it's 110 V at 60 Hz.)
The overall forecast for power-line-based home networks is now beyond the 32 million
nodes initally projected."
I have no idea where you're getting this 1.5k ohms as body resistance
1.5 kohm sounds about right. That's *body* resistance, measured with a good connection to internal fluids. Of course, most of us don't implant electrodes to measure this. ;-) You were measuring *skin* resistance, which depends on many things, including saltiness, moistness, skin thickness, and skin composition. With small electrodes on dry skin, it measures anywhere from 100 kohm to >100 Mohm.
Try repeating your measurements, but with reduced skin resistance. Make each electrode a big wire, one of which is squeezed in each hand. And drench the contacting skin with salt water. I just did this and got 20 kohm. And the skin is *still* a significant impediment to current. So a 1.5 kohm body resistance is reasonable.
Nope. The skin effect (unrelated to human skin, BTW) is mostly negligible at 60 Hz. At 60 Hz, it's only important for huge generators dealing with thousands of amps of current.
"Put your arm in it!" When I did so I didn't feel anything at first ... all the muscles in my arm (fingers, wrist, forearm, all of them) started to flex and unflex because of the current flow.
Stupid, stupid, stupid. It's things like this that earn Darwin Awards. A portion of the current was certainly flowing through your heart. It just wasn't enough to cause immediate cardiac arrest. And this is the most dangerous situation possible, because your heart may have been silently damaged. Plenty of people have gotten shocked, counted their blessings for not being killed, gone home at the end of the work day, and quietly died in their sleep from delayed heart failure. A quick trip to the friendly emergency room cariologist will show any subtle injury to the heart -- the heart's electrical waveform usually changes significantly when it is seriously injured (AFIK).
-- ;-)
Kuro5hin.org: where the good times never end.
Technical and security challenges aside for the moment, the really killer app for this technology is the seamless networking of otherwise non-networkable appliances.
Not everybody wants an ethernet jack on their toaster, but _everything_ has a power jack.
Imagine the following:
- buy a new VCR/DVD Player/Alarm clock? Plug it into the wall, and watch it set the time on itself to the same time as all the other devices in the house.
- Self-monitoring appliances that are syslog() capable (or something similar) and report faults to a central logging facility
- Appliances that export network APIs to provide scriptable control
...and a host of other Really Neat Stuff possible if you have a standardized network in every home.
Want to learn about race cars? Read my Book
According to the technical specs, it meets FCC Part 15 emission standards. Still, it is transmitting in the 4.3 - 20.9 MHz frequency band over non-twisted, non-shielded, copper power wiring. This has the potential of trashing important shortwave broadcast and amateur radio frequency bands. Widespread deployment of these devices could be a disaster for HF radio users. Noise is already a severe problem in many places.
Mea navis aericumbens anguillis abundat
How hard would it be to eavesdrop on something like this? Easier than a cat 5 connection? Would you have to worry about your neighbors potentially tapping into your data connection.
I'm trying to teach myself to set people on fire with my mind... Is it hot in here?
but only 14mbps? this may serve okay for gaming over home-based networks, but as we've seen, electric-plug networks can't touch good old 10baseT/cat5. i know electric plug is neater, but it costs a lot more, is not user-serviceable, and is held back by lack of bandwidth.
That I can make my own personal organic echelon by plugging my fingers into the electric sockets? ;)
-Stskeeps, http://unrealircd.com
Don't know what the bandwidth was but they were modulating the carrier to send data over power distribution lines.. company was Energis I think. They failed for a number of reasons.. EMF emissions being a security risk and the sheer' dirtiness' of the carrier signal were the main problems they didn't manage to overcome.
Oh.. and it's 240v at 50 Hz in the UK.... BTW.. which, as *everybody* knows, ain't strictly Europe.
t o b e
How well does it scale? How many nodes will we be able to have on a single circuit? How well will it deal with electrically noisy circuits? Will it be possible to isolate my network from my neighbor's?
This sounds pretty sweet, but... There is no heaven on earth... Things are not usually as perfect as they seem...
From what I understand the trial systems in the UK ended up retransmitting the network signal using any meta object. For example every street lamp on the road.
This is not good for EMF and for security of course.
The reason that an ISP->user connection will never be feasible across powerlines is due to transformers - the moment you put a signal through a transformer you get garbage out the other end.. that's the downside of transformers - and why you can't use load coils with xDSL. The second problem I see with this is that because of the high voltages involved, it is quite possible to kill yourself.. well, the voltages don't kill, but use alittle ohms law and you'll figure out why high voltages are a problem (for reference, your body is about a 1.5k resistor and your max safe current is 5mA with lethal at 30mA).. anyway.
I also think the technology will be limited SOLELY to the home market - if I was IT manager I wouldn't let my company even *think* about deploying it.. you have all kinds of nasty things in commercial/industrial settings on those wires that just make it totally unreliable - a blown circuit breaker takes out your network, phase shift from flipping on the refridgerator, all those flourescent lights severely throw the phase out of whack - you're left with anything *but* clean energy in a commercial setting.. this is why power strips are so VERY VERY VERY VERY important.. and UPS' on anything worth a damn.
So, uhh, don't expect this to be any kind of "long term" tech - it'll be around for about 10 years tops.. most new buildings these days have cat5 and coax drops just per default.
Only if you want the outside world to be able to talk to your lightbulbs and fridge!
:-)
Good point. Very few household devices would need to talk to the outside world. Although perhaps more than at first glance. Your fridge might want to talk to its manufacturer (or vice versa) for software updates or service calls if it detects a problem, for example. Your blender might want to conduct an email love affair with that cute toaster oven it sat next to on the store shelf. That sort of thing.
Meanwhile, how long before somebody builds a webcam into a light bulb?
-- Alastair
Raw data.. do they define what they mean by this?
You wonder why I ask....
In 10Mbps ethernet, the 10Mbps refers to the capacity of the ethernet as a single, baseband channel. at perfect 100% usage, the channel will contain 10 million bits/second.
In practice, the maximum amount any single host can transmit, (full sized 1518 byte frames, smallest legal inter-frame gap (9.6uS, or 96 bits) equals about 9.9Mbps. Accounting for ethernet framing, fcs, plus ethernet headers, plus ip & tcp headers, and accounting for tcp acks... the max throughput on ethernet between two hosts doing ftp is about 9.8Mbps.
Fine you say... close enough. True.
T1 = 1.544Mbps, raw data rate. This translates to near 1.3 mbps (or higher, I forget) after PPP framing and whatever else is in there.
However... take many wireless networking protocols. Using whatever proxim's protocol (whatever they use in their rangelan-II radios)... a raw data rate of 1.6Mbps translates to a max uni-directional broadcast of about 800Kbps. About 600-700Kbps in normal TCp operation.
I've seen 11Mbps wireless gear that only does 5Mbps in useful throughput....
so.. in other words... be wary when someone says 'raw data rate' or 'throughput'... you (and they) will probably be unsure of what you mean.
Also..for any given medium, one has to take into account latency due to (c), the fact that there really IS no such thing as throughput.. a more relevant profile of a link is...
How many of what size packet can our setup move with what latency?
10Mbps ethernet again... it's a good example, because aside from the (frame+header)/(data) ratio changing with the size of (data).. it's efficient.
Many wireless devices I've tested aren't so lucky. Some of them perform just great trying to bridge 1518 byte packets, but then you get down to 64 byte... and they drop *way* off (processing bottleneck or something..). In other words... small packets take up much more than their fair share of resources in some devices... so in a common office ethernet, where 75% of the packets are 200bytes or smaller (not 75% of the total data.. 75% of the packets)... perhaps you don't get the throughput you think?
Why am I on this rant?
Cheers.
D
Hmm... Just like Compuserve used gifs based on Unisys' technology, or mp3[*] is based on Fraunhofer's technology? I would be very careful to find out if the technology is encumbered by any current or pending patent claims before making it the standard. Whether or not the company says that it won't charge for the use of the technology, once enough people move to the new standard it starts looking real attractive to a corporation to go back on that agreement. Don't get fooled again.
[*] apparently Fraunhofer believes that their mp3 compression patent is broad enough that it is impossible to create mp3s (using any algorithm) without infringing on their patent.
Your right to not believe: Americans United for Separation of Church and
The overall forecast for power-line-based home networks is now beyond the 32 million nodes initally projected.
Time for IPv6? Or do we just put every toaster, refrigerator and light bulb in the house on an unroutable domain and handle the translation in the router/firewall/proxy built into the fuse box?
More seriously, this ain't bad. Speed is comparable to 10base Ethernet, but I'll still run Cat 5 cable in my new house (faster, and for now cheaper, since I already have the NICs and hub). One nice thing about signal-over-AC is that you don't need a separate cable to the device -- the power cord is the network cable, very convenient for appliance-type devices.
-- Alastair
1. Inari (formerly Intelogis) has been shipping a 350 Kbps powerline networking kit for 2 years. You can purchase it at Fry's, Office Max, CompUSA etc. It works in about 80% of the outlets in an average home. There are GPL'd Linux drivers for it on SourceForge. It's good for no more than about 10 nodes. It uses encryption to keep your neighbors from sniffing your data. Sure it is slow, but it's faster than your dial up connection.
2. 14 Mbps is really impressive on a power line. (Lots of reflections, lots of noise, dynamic line conditions). I wonder if it really runs that fast? Has anyone seen a demo?
3. Intel's home networking product is a phone line product based on the HomePNA (Home phone network association) spec.
4. Wireless is still more expensive than powerline and it has its own set of problems.
They look promising in this area. Peep this article for more info.
Just a random thought, but what happens if you're using a technology like this in, say, an apartment building? Is it even feasible? Assuming that it is feasible, two things that immediately come to mind are:
Security might not be a big deal to most people, but I'm sure bandwidth would be. Things to think about....
At least, in the Netherlands, that is.
Every expression is true, for a given value of 'true'
All those ping packets of x volts combined - boom!