questions for ESE519
0 views
Skip to first unread message
Philip Asare
Dec 8, 2010, 7:18:00 PM12/8/10
to ese519upe...@googlegroups.com
Pick a question and let me know which one you picked. I will submit mine last.
Philip
Car Hacks
1. If the car network is supposed to allow access for diagnostics and upgrades, but also protect itself from attacks, what stops a malicious attacker action as a "diagnostic tool" from having unauthorized access?
2. Do we need to take another look at the CAN bus since it clearly was designed under the assumption that all devices on the network will play nice? Or should ECUs implement their own defense mechanisms?
3. In most systems, even if we cannot prevent or fix an error, we can at least detect it. Are there any ideas on how to detect these attacks, even if they cannot be prevented?
ICD attacks
1. The authors suggest that there needs to be a balance between security and privacy, and safety and effectiveness. What causes these trade-offs and when can security and privacy get in the way of safety and effectiveness?
2. The authors say that in wireless systems, firewalls may not bee enough, and that protecting the wireless links themselves may be necessary. Why is this the case?
3. The authors used replay attacks to compromise the ICD device. This seems like a simple mechanism requiring listening to the programmer and replaying "messages" recorded. Is it possible to guard against such attacks? If so, how does one guard against such attacks?
Philip
Car Hacks
1. If the car network is supposed to allow access for diagnostics and upgrades, but also protect itself from attacks, what stops a malicious attacker action as a "diagnostic tool" from having unauthorized access?
2. Do we need to take another look at the CAN bus since it clearly was designed under the assumption that all devices on the network will play nice? Or should ECUs implement their own defense mechanisms?
3. In most systems, even if we cannot prevent or fix an error, we can at least detect it. Are there any ideas on how to detect these attacks, even if they cannot be prevented?
ICD attacks
1. The authors suggest that there needs to be a balance between security and privacy, and safety and effectiveness. What causes these trade-offs and when can security and privacy get in the way of safety and effectiveness?
2. The authors say that in wireless systems, firewalls may not bee enough, and that protecting the wireless links themselves may be necessary. Why is this the case?
3. The authors used replay attacks to compromise the ICD device. This seems like a simple mechanism requiring listening to the programmer and replaying "messages" recorded. Is it possible to guard against such attacks? If so, how does one guard against such attacks?
Santosh vattam
Dec 8, 2010, 8:55:48 PM12/8/10
to ese519upe...@googlegroups.com
Hey Philip,
I'm not submitting any questions anyway. So you may go ahead :-P
--
Thanks and Regards
Santosh G Vattam
Sirisha Nalmela
Dec 8, 2010, 9:03:26 PM12/8/10
to ese519upe...@googlegroups.com
Wowow... Thanks Philip.. I told it for fun.. but I will accept your help here ;) I am randomly picking no. 3 in both the papers.
Thanks a ton!
Philip Asare
Dec 8, 2010, 9:43:36 PM12/8/10
to ese519upe...@googlegroups.com
You are welcome.
@Santosh: good that you want to read the paper before you turn in your questions :) I'm sure Sirisha will read it before class tomorrow?
Philip
Sirisha Nalmela
Dec 8, 2010, 10:20:57 PM12/8/10
to ese519upe...@googlegroups.com
Luckily, the deadline of animation assignment I was struggling to get done is suddenly postponed to friday.
So I am back to reading the papers now.
I soooo want next 2 weeks to get over soon!!!
-Sirisha
Reply all
Reply to author
Forward
0 new messages