PDA

View Full Version : Beating the burglar alarm

edd
31st August 2011, 10:13 AM
The probability thread reminded me of a question I never came up with an answer to:

My parents have a burglar alarm that disarms on entry of a random 4 digit number. The alarm disarms regardless of any previous key pushes as long as the right code has been entered. For example, if the code was 7823 then entering 527823 would disarm it.

What sequence do you enter to get to the random number in the shortest number of key presses on average?

phunk
31st August 2011, 11:07 AM
I suspect it would be a De Bruijn Sequence (http://en.wikipedia.org/wiki/De_Bruijn_sequence), which would mean a 10000 digit sequence to cover all possibilities.

ETA: The uses (http://en.wikipedia.org/wiki/De_Bruijn_sequence#Uses) section of that wiki covers your question as an example.

Hellbound
31st August 2011, 11:14 AM
Actually, you'd rely on human psychology first (assuming users can specify their own codes). If you've researched the home owner, then things like birthdates and anneversiary dates should be first...along with the last fours of SSNs and phone numbers, keypad letter variations of 4-letter names, and similar.

After that, physical configurations of a keypad are often common: 4 corners, diagonals, cross, and similar.

Also, a lot of people won't reuse a number, even if the system allows it (or at least, not use a double-number together...1391 is more likely for someone to choose than 1139).

Of course, I suspect what you actually meant was treat this as a mathematical hypothetical exercise, rather than a practical one ;)

BowlOfRed
31st August 2011, 11:37 AM
I wonder if something similar to a gray code generator would work here. Thinking about 4 decimal digits was too hard, so I thought about 2 digits of 4 positions.

There are necessarily 4^2 or 16 unique codes: 11,12,13,14,21,22,23,24,31,32,33,34,41,42,43,44

Brute force would require 16*2 or 32 keypresses. Better would be a string where no 2-digit combination was repeated within (such as 11 being repeated at positions 1 and 2). If such a string were to be constructed, then it would be 17 characters long.

I have a feeling there's a way to do this in a mechanistic manner, but I just tried constructing a walk and seeing how it went.

11223344132421431

That seems to work. I don't think I'd want to do this by hand for 3 or more digits.

ETA: I *knew* someone would have a real answer before I finished. But once I started, I didn't want to peek...
Thanks, phunk

I Ratant
31st August 2011, 12:35 PM
I worked where we had locked safes for data.
In my department, all one needed to know to open the combo lock was the first number.
The second was twice that, the third 3 times, and the fourth its square.
Wasn't a particularly high threat environment though.

edd
31st August 2011, 01:06 PM
Thanks phunk! That's brilliant.

Toke
31st August 2011, 01:15 PM
I would be thinking of whether the solenoid in the lock were normally on or normally off, NO or NC, that makes a difference to how it would react to a large hammer.

Having installed some of these systems I know that the NC/NO is usually not programmable, whereas the continuous entry vs. # between each entry is.

Lensman
31st August 2011, 01:24 PM
I'd just check which numbers are the most worn - that won't get the correct order though, but if it's only 4 digits, you've got 4 tries to get it right.

Hellbound
31st August 2011, 01:56 PM
I'd just check which numbers are the most worn - that won't get the correct order though, but if it's only 4 digits, you've got 4 tries to get it right.

There's more than four combinations, though. IIRC it's 24 (4!):

1234
1243
1324
1342
1423
1432
2134
2143
2314
2341
2413
2431
etc.

Soapy Sam
31st August 2011, 03:50 PM
I only use imaginary numbers for PINs.
Much more secure.

Speedskater
2nd September 2011, 12:02 PM
Back to the original problem, with 0000 to 9999 we have 10,000 combinations and 40,000 digits. If we enter the digits serially in the correct order, we only need to enter a little over 10,000 digits (or maybe a lot over).
But it would be quicker to enter all 10,000 combinations than trying to follow a list.

2nd September 2011, 12:12 PM
Meanwhile, back in the real world, a 4-digit home burglar alarm gives you 30 to 45 seconds to enter the Disarm code. How many combinations can you try before it goes off?

not_so_new
2nd September 2011, 12:55 PM
Meanwhile, back in the real world, a 4-digit home burglar alarm gives you 30 to 45 seconds to enter the Disarm code. How many combinations can you try before it goes off?

Party Pooper.....

:D

JWideman
2nd September 2011, 01:24 PM
Meanwhile, back in the real world, a 4-digit home burglar alarm gives you 30 to 45 seconds to enter the Disarm code. How many combinations can you try before it goes off?

Who says you have to enter them manually? Bypassing the keypad and entering the numbers electronically from a laptop wouldn't take more than a few seconds.

Blue Mountain
2nd September 2011, 01:26 PM
I only use imaginary numbers for PINs.
Much more secure.

Cool, having to rotate your alarm panel by ninety degrees every time you use it must get tiring after a while.

Toke
2nd September 2011, 02:55 PM
Who says you have to enter them manually? Bypassing the keypad and entering the numbers electronically from a laptop wouldn't take more than a few seconds.

Keypads are made to be difficult to bypass, often with a anti tamper micro switch.
Or the keypad itself can be the micro-controller, inside an epoxy block.

PbFoot
2nd September 2011, 09:40 PM
I find the fastest way to defeat a numeric keypad is to closely examine the keys. By looking at which ones are the most worn or dirty will vastly decrease the number of combinations. If the keypad is too clean, or a visual inspection is inconclusive, press each button. The most used ones may have a slightly different "feel" due to wear of the switch mechanism.

-PbFoot

yomero
2nd September 2011, 10:37 PM
I hope I'm not too far off-topic. Does anyone know if the motion sensors of a burglar alarm could be activated by any of the following?

A) Ceiling fans being left on.
B) Moving images on the TV screen.
C) Large dogs (Irish setters) moving in the yard in front of large windows.

Yes, it's a stupid question, but I need the answer. Every time we leave the house, I have to check all rooms and turn off fans and TVs and lower the blinds. I suspect that's not necessary, but I'm not sure.

Toke
3rd September 2011, 01:42 AM
It would depend on whether it was IR or ultrasound detectors.
Ceiling fan would trigger both.
Tv none of them, but leaving it on when out is silly.
The IR may well trigger on something seen through the window.

Do you have any way to test it, like a system central with a diode for each room?

Soapy Sam
3rd September 2011, 01:48 AM
Cool, having to rotate your alarm panel by ninety degrees every time you use it must get tiring after a while.
Cartwheels.
I'm selling an exercise video.
That's why I need the safe.

JWideman
3rd September 2011, 02:22 AM
Keypads are made to be difficult to bypass, often with a anti tamper micro switch.
Or the keypad itself can be the micro-controller, inside an epoxy block.

On a system that weak? This is the electronic equivalent of a \$2 combination bike locks.

Toke
3rd September 2011, 02:44 AM
On a system that weak? This is the electronic equivalent of a \$2 combination bike locks.

I don't know how weak they come, but some measures are quite simple to implement.
Like a delay in the electronic between each input.
Pouring epoxy in the keypad case to cover the electronics may be beyond the cheapest systems.
You will need to know the schematics and software for each system in order to try to bypass with a laptop, those are somewhat guarded. :D

TjW
3rd September 2011, 06:47 AM
You may also want to watch out for the "duress" feature, which does disarm the system, but immediately calls the comm center to report disarming under duress.

DrDave
3rd September 2011, 06:52 AM
You could refer to an online list of codes such as the following:

http://www.guzer.com/videos/everyones-pin-numbers.php

Works for PIN numbers too

3rd September 2011, 12:37 PM
Now for another visit to reality:

1) What does the average home burglar alarm protect that is worth the effort it would take to defeat it?

2) What kind of burglar would be willing to take the effort to defeat the average home burglar alarm, in order to burgle the average home?

Tony
3rd September 2011, 07:29 PM
I find the fastest way to defeat a numeric keypad is to closely examine the keys. By looking at which ones are the most worn or dirty will vastly decrease the number of combinations. If the keypad is too clean, or a visual inspection is inconclusive, press each button. The most used ones may have a slightly different "feel" due to wear of the switch mechanism.

-PbFoot

Modern keypads have a short-cut sequence to arm the system. Plus, key fobs that can disarm or arm the system. Our code almost never gets entered. As I'm sure you can guess, this results in almost no wear.

WildCat
3rd September 2011, 08:09 PM
Who says you have to enter them manually? Bypassing the keypad and entering the numbers electronically from a laptop wouldn't take more than a few seconds.
But the burglar won't have a laptop until after he breaks in your house.

4th September 2011, 11:22 AM
I worked where we had locked safes for data.
In my department, all one needed to know to open the combo lock was the first number.
The second was twice that, the third 3 times, and the fourth its square.
Wasn't a particularly high threat environment though.

I forget which film, but in regards to this kind of situation the scene always sticks in my head.

Older woman is getting involved in a sketchy situation, after a while, her and a criminal type that is helping her, discuss her password for her computer ( which contained sensitive information. ) she informs him it is the birthdays of her children.

He proceeds to snap on her.

" I bet you think that is real cute, don't you? That no one could possibly guess that you would use your kids birthdays. Don't try and be cute, don't use your birthday, your anniversary , anything that has anything to do with you. Why? Because that is the first thing someone like me is going to try."

I honestly blame the movie " Hackers" on people's lack of intellect when it comes to passwords. They made the concept of knowing what someones password is so absurd people never really took the core of the advice they give ( Love , sex , god, for those who have not seen the film. ) , seriously.

Toke
4th September 2011, 12:47 PM
correct horse battery stable (http://xkcd.com/936/):D

Dave_46
4th September 2011, 01:31 PM
<snip>
The IR may well trigger on something seen through the window.

<snip>

I had an IR detector which although it could not see the window directly, it reacted to sunlight it could see that was coming through the window. Cured by re-siting the detector.

Dave

ETA This detector also gave problems when a spider took up residence inside it.

fuelair
4th September 2011, 02:01 PM
I hope I'm not too far off-topic. Does anyone know if the motion sensors of a burglar alarm could be activated by any of the following?

A) Ceiling fans being left on.
B) Moving images on the TV screen.
C) Large dogs (Irish setters) moving in the yard in front of large windows.

Yes, it's a stupid question, but I need the answer. Every time we leave the house, I have to check all rooms and turn off fans and TVs and lower the blinds. I suspect that's not necessary, but I'm not sure.By sad experience, the problem you are much more likely to have is a spider or insect crawling over the movement sensor - no dogs/cats, etc., no kids, no fans in the areas, do not leave tv on, but happened to us (with police notes
four times).

I Ratant
4th September 2011, 03:21 PM
My boss out-of-town would smoke a cigar in his office.
Sometimes it would trigger the smoke alarm, usually after he'd left the office.
I got where I would call the FD to tell them the cause before they got all excited and came over.
There was a motion detector situated high in a corner of the room, which I could get to blinking while sitting at my desk, and waving an arm.
One of the tamish feral cats would trigger motion detectors in the hangars when we'd left for the weekend.
Security finally captured it and took it clean across the test site.
A few weeks later I saw drag its weary ass back into a hangar.
Made it across the nuclear test site unscathed by any of the animules there... didn't check if it glowed, though. :)
One of the guys took it all the way home.

Toke
4th September 2011, 03:29 PM
Ships have dead-man alarms on the bridge, so you have to touch e.g. the radar every 12.5 min, sometimes it is supplemented with a motion detector so you can wave at it from a chair.
I have seen a detector equipped with the paper and string from a teabag, waving gently in the air-con breeze. :boggled:

TjW
4th September 2011, 07:18 PM
I had an IR detector which although it could not see the window directly, it reacted to sunlight it could see that was coming through the window. Cured by re-siting the detector.

Dave

ETA This detector also gave problems when a spider took up residence inside it.

I worked for an alarm company for several years.
A client had an alarm system that had a motion that would false regularly on the same three days of the year. It was fine the rest of the time. Finally tracked it down to a high window that illuminated a mirror, which illuminated a wall that the sensor saw, only on those three days. We offered to move the sensor, but the client elected to move the mirror.

BobK
4th September 2011, 11:22 PM
Get the police to ignore the alarm by setting off numerous false alarms over a period of time. Their response time will gradually increase. Donuts should be savored and not rushed.

Alternatively, if your PD charges for responses. A few false alarms will be sufficient to make the homeowner decide the the system is crap and they will stop setting it.

I've burgled many a ham that way.

Signed,

The Fallen Serpent
5th September 2011, 04:41 AM
I honestly blame the movie " Hackers" on people's lack of intellect when it comes to passwords. They made the concept of knowing what someones password is so absurd people never really took the core of the advice they give ( Love , sex , god, for those who have not seen the film. ) , seriously.

I know people wh o had to change their passwords after that movie. Though in my experience PASSWORD, PSSWRD, PASSWRD or PSSWORD or the most common. Yes, in all caps. Mostly older people prone to forgetting.

I think the bigger one is that people don't password protect their computer log in and do not realize how many spots might remember passwords even if you so "No." Personally I do not have the skills of even a moderate computer professional but I have seen backdoors before. I bet a friend once (after discovering which browser with a rather large backdoor he used) that if he gave me 30 seconds with his computer I would have all of his logins and passwords. He had a mini-meltdown when I started reading aloud in front of other friends what his adult site logins were. Then I showed him how to access it over the network our computers were sharing.

After watching Mythbusters defeat a finger print lock with scotch tape or printed paper the same group of guys and I decided to test out the experiment ourselves. A friend had a laptop with a thumb print key on it. Both lifting the print from a glass with tape and backing it with paper and just scanning his print in our cheap home scanner to print off worked. Lifting the print and scanning that however did not.

Fun stuff.

ponderingturtle
5th September 2011, 06:50 AM
Keypads are made to be difficult to bypass, often with a anti tamper micro switch.
Or the keypad itself can be the micro-controller, inside an epoxy block.

Or you really want fun, use a touch screen that randomly assigns the numbers to different locations.

fls
5th September 2011, 07:08 AM
This all reminds me of a story I read where a woman is kidnapped and held hostage for something like 7 years in a shed in a yard secured by a four-digit keypad accessible from the outside and inside of the shed. There is a description during the story of her occasionally trying a bunch of four-digit numbers on the keypad, but no mention that the keypad locks out after a certain number of tries or that it alerts the kidnapper that she is trying to escape. I suspect the author didn't think this through and figured "it would take a lot of tries" was sufficient to make her escape impossible. The other people in my book club just rolled their eyes when I brought that up.

Linda