Slashdot Mirror
VoIP As a Solution To Rural Broadband
boyko.at.netqos recommends his article up at Network Performance Daily, which notes the recent reports that up to 30% of households do not have a landline telephone, preferring a VoIP or cell-phone based solution. What to do with the miles of last-mile phone line infrastructure already in place in almost all the homes across the country? Maybe there's a solution to rural broadband by using the high-reliability frequencies reserved for voice purely for data — and using VoIP to make phone calls. From the article: "Repurposing the broadband of 0-25kHz would result in... speeds of around 14.4 kBytes/s (or 115.9 kbits/s) upload and 28.8 kBytes/s (231.3 kbits/s) download. That's not much of a speed boost. Still, if you've been plodding along on a '56.6k' modem, at speeds of 7.2kBytes/s, this would be like an oasis in the desert. And what about those phone calls? Well, if you make the same phone calls with VoIP that you were with the standard 0-4kHz landline, it would only take about 20.8kbits/s using the G.723.1 codec — that still leaves you with 80% of your broadband capacity when on the phone — and 100% of your broadband when you're off it." Only the US FCC calls 231K "broadband," but as noted it does beat dialup.
Not to be pedantic (and I understand the general drift of the article), but from wikipedia: "... the US FCC used 200 kbit/s in their definition until march 19th 2008 after which it was scaled up to require a minimum of 768 kbit/s to be defined as broadband and at that time the FCC introduced new tiers in their definition as follows: 1) 200kbit/s to 768kbit/s ("first generation data"); 2) 768kbit/s to 1.5Mbit/s ("basic broadband"); 3) 1.5Mbit/s to 3 Mbit/s; 4) 3Mbit/s to 6 Mbit/s; and 5) 6Mbit/s and above." http://en.wikipedia.org/wiki/Broadband_Internet_access
That was much like the overall premise and promise of ISDN BRI - "high speed" digital access over voice grade plant, which failed miserably due to a number of technical, political, and corporate reasons.
Granted, the OP's proposal is somewhat different, as I assume he was referring to using DSL-like technology in the full voice band. But, there are also limitations on how much data can be carried in a given amount of spectrum using various modulation and encoding schemes.
This sounds pretty much like IDSL. The problem has never been technological - the problem is getting your telephone company to implement it at a price that's reasonable. Instead of breaking up the low frequencies into two 64 Kbit/sec ISDN BRI channels and one 16 Kbits/sec D channel for signaling, IDSL just uses all 144 Kbits/sec (symmetric) for data. The suggestion is asymmetric ISDN based broadband, but that's a minor difference. ISDN goes much longer distances than ADSL or SDSL due to the lower frequencies. In the early 1990s I had ISDN and it worked fine, except the phone company charged $ 250 a month for unlimited 128 Kbits/sec. Great technology (at the time), but insane pricing.
911 works on VOIP -they call it e911. It works as long as your provider has your address. Many providers have it, including Acanac ( http://www.acanac.com/Phones.htm )
"Knowledge is the only instrument of production that is not subject to diminishing returns" -Journal of Political Econom
It gets 144kbps - ISDN has two 64kbps B channels and a 16kbps D channel, and is typically used for a 128kbps bonded circuit.
The big advantage of IDSL is distance - it typically gets about 30,000 feet, compared to about 18000 for most DSL flavors.
Bill Stewart
New Fast-Compression-only CPR http://preview.tinyurl.com/dy575ks
There isn't too much information in the article, particularly what problems may be encountered. The amount of data that can be delivered will vary greatly due to certain technical considerations. Politically, giving everyone in rural areas the 'same lousy service' is a minefield.
The outside cable plant and distance to the central office is everything:
* "Wires on poles" can degrade bandwidth 10x or more, particularly if there is industrial or broadcast interference. Modern underground cable plant can provide several Mbit/s up to 30km or so.
* Loading coils, commonly used in the past to maintain 600 or 900 ohm line impedance, limit the bandwidth of the lines to not mush more than 4kHz. They must be removed which is allot of tedious labour. Once removed, POTS may not work properly. Since some lines will need them and others definitely not, this gives a great excuse to 'take forever' to install the service.
* COTS DSL-modem/routers, common in many areas, may not work on large runs. Slightly modified units can put out greater signal and have better echo cancellation. This looks like a lock-in and higher prices. Higher transmission levels, lower received levels and longer runs invite crosstalk in a big way. It may be that many systems start out really good, but quickly degrade as more subscribers are added.
* Some rural cable-plant is "hollow-sounding" with voice and will simply not work with DSL. I'm no expert on US rural phone systems, but its fair to say most will get the pitiful 256kbit/s rate. This is what can be achieved with above-ground cable-plant at 30km in a city environment. The actual case I use example is Buenos Aries.
Any cable-plant that doesn't support 25kHz should be recycled! Otherwise, most will probably do much better, so limiting service to below 256kbit/s is deceptive. All told, there are a number of technical hurdles, which can be overcome, but the politics will go on forever.
This isn't a nice comparison to make, but in England there is more 'broadband' (there is a somewhat higher standard to the definition there) in the country than the city. Of course, like most of Europe, all wires are underground.
Power goes out. Landlines still work. Weather gets crappy. Landlines still work. Not much that can fail on the user's end. Start tossing VoIP in the picture and you're adding a whole bunch of equipment that has lot of ways to fail.