idea - wireless distributed home alarm system

[rowan194]

Instead of passive sensors reporting back to an active control
box/panel with a siren, why not a distributed system with each sensor
containing its own siren?

If one sensor is tripped then it sends out a signal and all of the
other sensor locations squeal. There is no one point that can be brute
forced to kill the system; you would need to crack the message sequence
used to command the sensors to arm or disarm.

Maybe this has already been done...?

 

[Caliban]

Instead of passive sensors reporting back to an active control
box/panel with a siren, why not a distributed system with each sensor
containing its own siren?

If one sensor is tripped then it sends out a signal and all of the
other sensor locations squeal. There is no one point that can be brute
forced to kill the system; you would need to crack the message sequence
used to command the sensors to arm or disarm.

Maybe this has already been done...?

Cost of duplicated circuitry would be a factor, all that repeated power
and control (how do you turn them off?) wiring too.

You could kill it by removing the power - unless each sensor/siren had a
backup battery (more expense).

I can't see this being viable n a very competitive market.

 

[rowan194]

Cost of duplicated circuitry would be a factor, all that repeated power
and control (how do you turn them off?) wiring too.

You could kill it by removing the power - unless each sensor/siren had a
backup battery (more expense).

I'm talking about a wireless system, thus there is no single point that
you can disconnect (or damage) to disable the system.

Each sensor box would have a battery, microcontroller, sensor and piezo
speaker. Possibly a small solar panel also, since most of them will be
mounted on windows. Most of the time would be spent in sleep mode with
the uC occasionally waking up to broadcast its status, or flash a
deterrent LED.

False alarms are an issue, so is ongoing development (I can't imagine
that having to reflash 20 window sensors and a couple of PIRs stuck on
the ceiling will be fun)

I can't see this being viable n a very competitive market.

Probably not, but I'm intending to build it for use in my own home.

 

[Caliban]

had a



I'm talking about a wireless system, thus there is no single point that
you can disconnect (or damage) to disable the system.

Each sensor box would have a battery, microcontroller, sensor and piezo
speaker. Possibly a small solar panel also, since most of them will be
mounted on windows. Most of the time would be spent in sleep mode with
the uC occasionally waking up to broadcast its status, or flash a
deterrent LED.

False alarms are an issue, so is ongoing development (I can't imagine
that having to reflash 20 window sensors and a couple of PIRs stuck on
the ceiling will be fun)




Probably not, but I'm intending to build it for use in my own home.

Have a look at the Zigbee wireless modules.

The chipcon CC2420 is a good start:
http://www.chipcon.com/index.cfm?kat_id=2&subkat_id=12&dok_id=115

Also Microchip have a Zigbee development board but it's not cheap:

http://www.microchip.com/stellent/idcplg?IdcService=SS_GET_PAGE&nodeId=1406&dDocName=en021925

 

[dmm]

I'm talking about a wireless system, thus there is no single point that
you can disconnect (or damage) to disable the system.

Each sensor box would have a battery, microcontroller, sensor and piezo
speaker. Possibly a small solar panel also, since most of them will be
mounted on windows. Most of the time would be spent in sleep mode with
the uC occasionally waking up to broadcast its status, or flash a
deterrent LED.

False alarms are an issue, so is ongoing development (I can't imagine
that having to reflash 20 window sensors and a couple of PIRs stuck on
the ceiling will be fun)


Probably not, but I'm intending to build it for use in my own home.

It'd probably be pretty spiffy if you got it all to work though.

Have a look at Zigbee RF comms. Its a cheap, low power, low data rate,
networked system. It should be ideal for your use. There's a description
of it in February and March 2005 issues of Circuit Cellar. They even have
a contest going using the Zigbee devices.

http://www.circuitcellar.com/
http://www.zigbee.org/en/index.asp

Nordic have some cheap low power transceiver chips as well.
http://www.nvlsi.no/index.cfm?obj=product&act=display&pro=83

 

[Scott B]

Instead of passive sensors reporting back to an active control
box/panel with a siren, why not a distributed system with each sensor
containing its own siren?

If one sensor is tripped then it sends out a signal and all of the
other sensor locations squeal. There is no one point that can be brute
forced to kill the system; you would need to crack the message sequence
used to command the sensors to arm or disarm.

Maybe this has already been done...?

I had thought of this a while ago but using standard home alarm
components, using a code hopping UHF car alarm in the sensors. This
would send a secure signal back to the normal home alarm panel inputs.
Depending on how many ZONES and devices you need, each ZONE would
require a different receiver.
Oatley Electronics have a receiver unit that can accept up to 15
transmitters.

http://www.oatleyelectronics.com/remote.html

This means each ZONE can have up to 15 sensors connected to it.
You would then use the alarm output to trigger the sirens/lights etc,
which could be hard-wired or remote(via another transmitter/receiver).
Throw in the $50- Jaycar Nokia SMS kit and you would have one hell of
a home alarm/notification/automation system...

Just a thought.
Scott B.

 

[Chris]

"rowan194" <[email protected]> wrote in message

I had thought of this a while ago but using standard home alarm
components, using a code hopping UHF car alarm in the sensors. This
would send a secure signal back to the normal home alarm panel inputs.
Depending on how many ZONES and devices you need, each ZONE would
require a different receiver.
Oatley Electronics have a receiver unit that can accept up to 15
transmitters.

http://www.oatleyelectronics.com/remote.html

This means each ZONE can have up to 15 sensors connected to it.
You would then use the alarm output to trigger the sirens/lights etc,
which could be hard-wired or remote(via another transmitter/receiver).
Throw in the $50- Jaycar Nokia SMS kit and you would have one hell of
a home alarm/notification/automation system...

Just a thought.
Scott B.

Yes, that notification point can be a pain - having driven across the city
recently to attend a false alarm. My current thinking is to duplicate the
few sensors I have. ie position 2 at different locations in a room or
hallway so that both would need to be active at the same time to cause an
alarm. Has anyone tried this?

Cheers.

 

[David]

Yes, that notification point can be a pain - having driven across the city
recently to attend a false alarm. My current thinking is to duplicate the
few sensors I have. ie position 2 at different locations in a room or
hallway so that both would need to be active at the same time to cause an
alarm. Has anyone tried this?

Cheers.

You could arrange the logic so that if the one sensor tripped twice
within say one minute, or two different sensors tripped then activate
the alarm. Cheaper than two sensors per room. Also, you can use dual
detection sensors. These use IR and microwave sensors built into one unit,
and both sensors need to trip to activate the output.

David