[LINK] Broadband powered by ... power
rchirgwin at ozemail.com.au
rchirgwin at ozemail.com.au
Mon Nov 29 17:50:31 EST 2004
Robin Whittle wrote:
>Data over powerlines (PLC Power Line Communications or BPL Broadband
>over Power Lines) is a can of worms, in every respect. Not only does it
>cause interference, but it is expensive, hard to deploy in a
>standardised way, and covers short distances with no proper isolation or
>even the ability to direct the flow of signal in one direction or
>another along a cable.
The system is also vulnerable to DoS via signal injection; quite minor
variations in cable impedance send the emitted RF interference through
the roof; and the terminal equipment is very difficult to protect
against spikes (eg lightning strikes).
Consumer grade surge protectors are merely very deep low-pass filters;
and if you filter out the high frequencies, you kill communications
(this is why it's also nearly impossible to buy ADSL lightning
protectors, and I don't mean the Dick Smith/Tandy surge protectors,
which are approved only up to ISDN frequencies).
And it amazes me that people who are perfectly well aware of the number
of blackouts Sydney has suffered this year - every single storm - will
quite happily imagine that some magic will keep their communications
happening when the lines are downed by trees.
Powerline >may< be more feasible where the supply has been buried
underground, but in Australia, above-ground dominates.
>No major equipment or chip manufacturer is really pursuing this.
Except Cisco (via Linksys), and it's complaining that it has way too
much unsold HomePlug on the books (on the record, this month; the CTO,
whose name I can't remember).
>I think the promoters of this technology - except perhaps for limited
>data rates within the home (and then, how does this remain isolated from
>next-door's system?) - should be challenged if it looks like they could
>suck in investors or get government support.
Let's posit a home which has three electrical circuits (fuses) for power
points. If the NTU is at the point of ingress, it can communicate with
all three circuits - except that it also creates an electrical bridge
between them which must be dealt with, or safety is compromised (for
example, earth leak detectors would not like any electrical continuity
between different circuits). If the NTU is confined to one circuit, then
you have a crippled network.
As for your concern about isolating households; as far as I can tell,
the PLC systems use frequency division multiplexing; so the carrier
frequencies to your house aren't the same as next door.
Powerline == snake oil. But because its proponents can wave some nice
big numbers, hyperventilation overrides consideration.
>Otherwise, I think they
>should be ignored.
> - Robin
>Link mailing list
>Link at mailman.anu.edu.au
More information about the Link