Here's an idea - Electronic Circuits Projects Diagrams Free

0mega · 15th February 2004, 12:42 AM

Hello,

I was thinking about how unsecure wireless remote controls are... Most remote controllers are simple one-way devices that transmit a frequency / code to a car to open it. Anyone with a wideband recorder/playback system can "listen in" and repeat the code to open your car.

My idea works very much like modern-day network logins:

1. Client (remote) sends "want to login" token
2. Server(garage/car/whatever) sends random 16-bit hash.
3. Client XORs the hash with the 16-bit login code and sends it back to server.
4. Server unhashes (XOR again) to extract code
5. Compare codes
6. If codes match, open door / whatever.

Of course, this makes for a clunki(er) remote, but it is far more secure. Simply repeating the code will have no effect at all.

Of course, if my idea is complete bullshit, feel free to tell me.

Cheerz,
JB

john1 · 15th February 2004, 02:31 AM

yes its a good idea.

It took me a while to comprehend what you were saying,
cos you weren't using normal language.

The remote shouldn't be much different in its effect.
I dont think it would be notice-ably affected.

Chips to do this are cheap enough.
Maybe it is already done on some cars.

It occurs to me that its only the signal from the
key-fob that is suspect.
The signal from the vehicle to the key-fob would not
be suspect.

That strikes me as an ideal situation for a 'high-encryption' one way, to a 'high-decryption' set
on the car.

Dunno ... its all getting too hi falutin' for me.
And i cant figure out where to vote ....
Its time i gave up

John

john1 · 15th February 2004, 02:39 AM

found the vote bit,
it comes up after ...

John

Roff · 15th February 2004, 03:22 AM

Now you have to have a sensitive receiver in the remote, with some processing power, in addition to the transmitter. This might be tough to fit into a key or fob.

**Nigel Goodwin** · 15th February 2004, 11:06 AM

I thought some remote keyfobs always transmitted different codes, in order to stop them being copied?. Both the transmitter and receiver use the same pseudo random sequence - so every time you operate it a different code is used.

pwollner · 16th February 2004, 09:39 PM

idea is ok

yet still not secure enough and too complicated

two timers (one in remote and one in car/garage) could also do it- they produce identical keycodes every few mins. that would make the remote cheaper since it would only have to be one-way

View Poll Results: Is this idea BS
No	2	28.57%
Yes	5	71.43%
Voters: 7. You may not vote on this poll

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

15th February 2004, 12:42 AM	(permalink)
0mega	Here's an idea Hello, I was thinking about how unsecure wireless remote controls are... Most remote controllers are simple one-way devices that transmit a frequency / code to a car to open it. Anyone with a wideband recorder/playback system can "listen in" and repeat the code to open your car. My idea works very much like modern-day network logins: 1. Client (remote) sends "want to login" token 2. Server(garage/car/whatever) sends random 16-bit hash. 3. Client XORs the hash with the 16-bit login code and sends it back to server. 4. Server unhashes (XOR again) to extract code 5. Compare codes 6. If codes match, open door / whatever. Of course, this makes for a clunki(er) remote, but it is far more secure. Simply repeating the code will have no effect at all. Of course, if my idea is complete bullshit, feel free to tell me. Cheerz, JB

15th February 2004, 02:31 AM	(permalink)
john1	yes its a good idea. It took me a while to comprehend what you were saying, cos you weren't using normal language. The remote shouldn't be much different in its effect. I dont think it would be notice-ably affected. Chips to do this are cheap enough. Maybe it is already done on some cars. It occurs to me that its only the signal from the key-fob that is suspect. The signal from the vehicle to the key-fob would not be suspect. That strikes me as an ideal situation for a 'high-encryption' one way, to a 'high-decryption' set on the car. Dunno ... its all getting too hi falutin' for me. And i cant figure out where to vote .... Its time i gave up John

15th February 2004, 02:39 AM	(permalink)
john1	found the vote bit, it comes up after ... John

15th February 2004, 03:22 AM	(permalink)
Roff	Now you have to have a sensitive receiver in the remote, with some processing power, in addition to the transmitter. This might be tough to fit into a key or fob.

15th February 2004, 11:06 AM	(permalink)
Nigel Goodwin	I thought some remote keyfobs always transmitted different codes, in order to stop them being copied?. Both the transmitter and receiver use the same pseudo random sequence - so every time you operate it a different code is used.

16th February 2004, 09:39 PM	(permalink)
pwollner	idea is ok yet still not secure enough and too complicated two timers (one in remote and one in car/garage) could also do it- they produce identical keycodes every few mins. that would make the remote cheaper since it would only have to be one-way __________________ pwollner