Another case of a certain industry lagging behind, I mean come-on – who seriously still using proprietary cryptography algorithms in 2010? Especially only 40 or 48-bit protocols, with the processing power available on hand now and new techniques like GPU based cracking – that just doesn’t cut it.
The latest discovery of such implementations was in the immobiliser technology used by car companies to secure their expensive vehicles. A researcher Karsten Nohl has exposed these weaknesses at the recent Embedded Security in Cars conference in Germany.
Weak cryptography means that car engine immobiliser technology has become easy for crooks to circumvent.
Nothing weaker than 128-bit AES is considered sufficient protection for e-commerce transactions, but car manufacturers are still using proprietary 40-bit and 48-bit encryptions protocols that are vulnerable to brute force attacks. Worse still, one unnamed manufacturer used the Vehicle Identification Number (VIN) as the “secret” key for the immobiliser.
The weakness of the technology was exposed in security research by ethical hacker Karsten Nohl of Security Research Labs, who links the weakness of the technology with a growth in car thefts in Germany last year, following years in decline.
Nohl outlined preliminary findings from his research at the recent Embedded Security in Cars conference, in Bremen, Germany. His research covers the communications between card immobilisers and engine electronic systems in dozens of cars. For example, Nohl was able to crack the Hitag 2 car immobiliser algorithm used by Dutch firm NXP Semiconductors in around six hours.
And using the VIN number as the secret key? Well, that’s not very secret is it? It’s akin to using the MAC address of a computer as the SSH secret key, no one in their right mind would do that. I guess that’s what happens when you leave the engineers to implement cryptography schemes without having anyone around handy with the cluestick.
I’d imagine some of these systems are protecting extremely expensive cars, so some basic equipment, some strong crypto knowledge and 6 hours and you can land yourself a $100,000 car. Not bad for a days work.
The research builds on work by other computer scientists and encryption experts dating back at least five years. In 2005 Ari Juels of RSA Labs and researchers at Johns Hopkins University in Baltimore, Maryland, circumvented the encryption system used by Texas Instruments.
Manufacturers of car immobiliser technology have defended the robustness of their technologies.
“To our knowledge the direct causal link between the failure to adopt AES systems and the rise in car theft cannot be drawn,” Thomas Rudolph of NXP told New Scientist.
Texas Instruments claimed its proprietary cryptographic systems might be stronger than AES. Nonetheless both firms are in the process of phasing out their home-cooked crypto tech in favour of industry standard encryption systems based on 128-bit AES.
And what it is with TI claiming their system MIGHT be stronger than AES? When did ‘might‘ ever give anyone confidence? In all honesty, there is no reason at all for using proprietary algorithms or implementations. Those out in public like AES have been tried, tested and approved by the greatest crypto minds in the World, I don’t care how smart you think your employees are – but trust me they aren’t as smart as the people scrutinising AES.
I hope to see all companies using weak proprietary protocols in any industry phase them out and switch to tried and tested industry algorithms.
Source: The Register
- The Logjam Attack – ANOTHER Critical TLS Weakness
- BitTorrent Bleep – Encrypted, Decentralized Voice & Text App
- Google Revoking Trust In CNNIC Issued Certificates
- Researchers Hack Mobile Calls On GSM Network
- Awareness of Phishing is on the Up – But so are Monetary Losses
- Stoned Bootkit – Windows XP, 2003, Vista, 7 MBR Rootkit
Most Read in Cryptography:
- The World’s Fastest MD5 Cracker – BarsWF - 47,197 views
- Hackers Crack London Tube Oyster Card - 43,061 views
- WPA2 Vulnerability Discovered – “Hole 196″ – A Flaw In GTK (Group Temporal Key) - 32,140 views