Wireless proximity Alarm

[HThomas15]

Hi,

I am trying to build an alarm system that can detect when units have been moved more than a certain distance from the alarm unit, and sets the alarm off.

I have designed a simple circuit using quite a few PICAXE chips with BASIC programming to operate the circuit.

HOwever I am really struggling to find a suitable wireless connection between units and the actual alarm unit, so that the unit can differentiate between different units (with different signals or something?).

I have tried IR however that has the major disadvantage that it only works in straight lines of site. I also looked at using a radio garage door opener to create a permanent signal, and the radio transmitter can be attached to the unit that i want to alarm. However this drains the battery of the transmitter incredibly quickly, using one for such a long period of time is against RF regulations, and also I do not know how to get multiple transmitters on the same frequency to be individually identified by the radio receiver so that it can output specific codes for each transmitter and even less so how this can be interfaced to a picaxe circuit.

Any help will be greatly appreciated!!
Many Thanks

 

[(*steve*)]

I am trying to build an alarm system that can detect when units have been moved more than a certain distance from the alarm unit, and sets the alarm off.

Yeah that's a really hard problem.

 

[goldfist]

I know of a great piece of software to handle this and cheap if it were wireless networked devices over IP. AlertPingPro is a program I bought a few years ago that has this capability if it were communicating over IP. If each device had a Raspberry PI in them that communicated over wireless TCP/IP just idle on a DHCP lease and you had Alert Ping Pro running to make sure that the many devices were present and if not present to alert as to which one that it is as option maybe, however, your still in the same situation with batteries draining in each of the detected devices that rest idle active on the wifi. You could go with rechargeable batteries maybe and charge them overnight etc, however you would have to do the math to make sure that the batteries can hold the Rasberry PI with wifi active during the period in which whatever these are installed in you may be limited to a tight space where a large NiCad may not work out.

Going with reg wifi B, G etc you dont have to worry about FCC or RF regulation issues also and gives you a good range which can be controlled by the wireless routers transmission strength and/or devices transmission strength, so you can reduce the radius before the alert is caused so it triggers sooner than later to catch who is stealing something etc.

You also can have this wireless secure so that no one can place a dummy wireless connection in place of it, in which an unprotected radio signal without source confirmation can simply be mimic'ed and the item stolen with a radio transmission mimic device left behind if its as simple as a garage door opener transmission that is not protected from spoofing the system.


I built a notification system around Alert Ping Pro when I worked in IT for 7 years as the main admin. I needed a way to be proactive vs reactive and so I built up a system around Alert Ping Pro monitoring that would execute programs I wrote that interfaced with Shoretel VoIP phones to have the monitoring system call me and when I received the call, I heard in my own voice back to myself from preprogrammed audio files that the C++ program launched and played which was tied to the alert condition of the software, so this audio file played over the VoIP Phone to my cell phone or home phone and called once every 5 minutes until the problem was resolved and I then knew about the problem and where the problem was located in the complex network of 7 sites and multiple networks including critical point of sale network. This allowed me to know that our exchange server crashed at 2:30am and so I was able to receive this phone call and rush into work and get the exchange system back up and running before people stroll in at 6am to start their day and check their important e-mail etc. Made this system for less than $50 parts and software, and about 8 hours labor which wasnt too bad. My boss got a kick out of it and I got a pat on the back and raise for the monitoring system that I built to watch the network and devices on it.

I am also a white hat hacker and so I look at the design not just working, but at how it can be defeated and if it can, make it better so that a black hat cant get around it. I have made donations to www.hak5.org which is a great white hat hacker site if anyone is interested in white hat hacking. Lots of stuff is not as secure as it should be, especially when it comes to BlueTooth devices and lower end garage door openers that can have a signal mimic'ed.

 

[(*steve*)]

Let's go back to the basics.

Determining the distance between two devices.

*THAT* is the hard problem. There are multiple ways to solve it, but you come across problems in practice. GPS in each device is perhaps the simplest in many respects, but that's not going to work at short distances.

relying on low of power with distance, or timing a reflection (e.g. radar/sonar) are also problematic in places that have multiple targets, reflective surfaces, attenuating regions or surfaces, etc.