[Discuss] Home security & automation

[tmetro+blu at gmail.com (Tom Metro)]

Richard Pieri wrote:
>Derek Martin wrote:
>> Whoever is monitoring the home can be fairly sure
>> that a call to the police is warranted.
> 
> Not until after an intruder has left the premises. Read the very first
> item from ADT's FAQ:
> http://www.adt.com/customer-service/home-security-faqs

You mean "How does my ADT alarm system work?" where it says, "Once a
zone has been violated, the alarm system then dials ADT and transmits
the message through a telephone line. Once the signal is received, the
alarm system will wait to send another message if another zone is tripped."

So you're thinking that if the ADT setup requires the intruder to trip
multiple zones before ADT dispatches the authorities, then it's a
pointless setup?

It's rather vague the way they describe it. They don't specify when they
actually call the authorities. Don't they usually have a process of
calling the home owner first?

I'm not sure what panel ADT currently used, but they used to show the
same GE panel I'm familiar with in their ads. That panel doesn't do this
multiple zone alerting thing they describe. (A zone, by the way, is
simply a single wireless sensor, like a door switch or motion detector.)
You trip a zone, alarm goes off after a programmable delay, and then
after another programmable delay, it dials out to the specified numbers.
Subsequently tripped zones are irrelevant.


Derek Martin wrote:
> ...if both the cellular and land line communications are severed, 
> something funny is going on.  Whoever is monitoring the home can be 
> fairly sure that a call to the police is warranted.

Once upon a time alarm monitoring companies used leased lines to the
customer so they could immediately detect when the loop was cut. I think
there might have been a period of time in which the telcos offered that
service on a regular local loop, avoiding the need for a leased line.
(The telco would detect the cut, and notify the alarm monitoring service.)

As far as I know, all modern alarm monitoring services (except perhaps
for some high-end commercial property monitoring services) simply use
regular phone lines. They depend on the alarm panel calling them
periodically to report its status. The timeout for not hearing from the
alarm panel might be days. Not minutes, or even hours. Thus by the time
you get alerted of a line cut its way too late.

The timeout period for not getting supervisory GPRS packets may be
smaller. I'm not sure. But the monitoring company is paying for the cell
data, so they have incentive to have the messages happen as infrequently
as they can get away with.

The only practical way to keep a close eye on a remote alarm system in a
fail-safe way is to use the Internet, where it is practical to have the
monitoring service drive the process and ping the panel on the order of
minutes. But because the Internet link is unreliable, you need some
automated escalation procedure (like calling the alarm panel's modem)
before you bother involving people.

 -Tom

-- 
Tom Metro
The Perl Shop, Newton, MA, USA
"Predictable On-demand Perl Consulting."
http://www.theperlshop.com/