TY - JOUR
T1 - T-Box
T2 - A Forensics-Enabled Trusted Automotive Data Recording Method
AU - Lee, Seungho
AU - Choi, Wonsuk
AU - Jo, Hyo Jin
AU - Lee, Dong Hoon
N1 - Funding Information:
This work was supported in part by the Korea Ministry of Land, Infrastructure and Transport, and in part by the Korea Agency for Infrastructure Technology Advancement under Project 18TLRP-B117133-03.
Publisher Copyright:
© 2013 IEEE.
PY - 2019
Y1 - 2019
N2 - Modern vehicles are equipped with numerous electric control units which exchange vehicular status data, providing drivers with convenience, efficiency, and safety. In addition, the autonomous vehicles adopt various sensors that produce high volumes of high-speed data to process and assess internal and external situations. This data is particularly useful to automotive service providers such as car insurers, rental companies, and manufacturers. One way to understand how this data is used is to imagine the scenario in which an automobile insurer would provide a discount to a customer with an accident-free or near accident-free driving record. However, it is still possible that a less than the honest customer could manipulate their driving data in order to receive premium insurance services at preferential rates. To prevent this and similar scenarios, it is then critical to ensure that all data generated in a vehicle upholds integrity, continuity, and non-repudiation. Unfortunately, no such trustworthy data recording system of this caliber exists in any manufactured vehicle to date. This paper attempts to respond to this need, and we present a reliable automotive data recording system that satisfies these requirements and detects malicious manipulations from data deletion, replacement, replaying, and truncation. The proposed method additionally satisfies forward integrity of message authentication keys and is designed to utilize recorded data as automotive forensic evidence. Finally, the evaluation results demonstrate that our system can manage bandwidths of up to 64 MB/s.
AB - Modern vehicles are equipped with numerous electric control units which exchange vehicular status data, providing drivers with convenience, efficiency, and safety. In addition, the autonomous vehicles adopt various sensors that produce high volumes of high-speed data to process and assess internal and external situations. This data is particularly useful to automotive service providers such as car insurers, rental companies, and manufacturers. One way to understand how this data is used is to imagine the scenario in which an automobile insurer would provide a discount to a customer with an accident-free or near accident-free driving record. However, it is still possible that a less than the honest customer could manipulate their driving data in order to receive premium insurance services at preferential rates. To prevent this and similar scenarios, it is then critical to ensure that all data generated in a vehicle upholds integrity, continuity, and non-repudiation. Unfortunately, no such trustworthy data recording system of this caliber exists in any manufactured vehicle to date. This paper attempts to respond to this need, and we present a reliable automotive data recording system that satisfies these requirements and detects malicious manipulations from data deletion, replacement, replaying, and truncation. The proposed method additionally satisfies forward integrity of message authentication keys and is designed to utilize recorded data as automotive forensic evidence. Finally, the evaluation results demonstrate that our system can manage bandwidths of up to 64 MB/s.
KW - ARM TrustZone
KW - Forward integrity
KW - audit trail
KW - digital forensics
KW - event data recorder
UR - http://www.scopus.com/inward/record.url?scp=85065203147&partnerID=8YFLogxK
U2 - 10.1109/ACCESS.2019.2910865
DO - 10.1109/ACCESS.2019.2910865
M3 - Article
AN - SCOPUS:85065203147
SN - 2169-3536
VL - 7
SP - 49738
EP - 49755
JO - IEEE Access
JF - IEEE Access
M1 - 8689029
ER -