2017Unpublished venueRequires access

Evaluation of Security Mechanisms for an Integrated Automotive System Architecture

Joakim Karlsson, Pontus Khosravi

Open publisher page 0 citations

Abstract

Prior research on security mechanisms for the connected vehicle focuses on the current heterogeneous distributed embedded electronic architecture (EEA). However, due to increased demands on data-processing and bandwidth from future functionality, such as automated driving and sophisticated infotainment systems, vehicle manufacturers are now moving towards centralisation of functionality in one or more integrated electronic control units (ECUs). In this thesis, we identify security issues related to centralization of functionality in the automotive domain, and investigate mechanisms which mitigate these issues. Further, we investigate the support, both in hardware and software, for these mechanisms in the current and future automotive embedded systems. To identify security issues related to the integrated EEA and to identify mechanisms which mitigate these issues, a literature review of other domains was performed. The AUTomotive Open System ARchitecture (AUTOSAR) development partnership has specified a software architecture with the aim of standardizing software development in the automotive industry. We have investigate whether the identified mechanisms are specified in the AUTOSAR framework by interviewing AUTOSAR specialists. Some form of hardware feature is typically needed to be able to use the mechanisms. Therefore, hardware requirements for each mechanism were identified and used to evaluate the support present in ECU hardware. We have found that there was lacking support for most of the identified mechanisms, in both the present ECU hardware as well as in the current AUTOSAR platform. However, the hardware targeted to be used in future ECUs, along with the next-generation software standard Adaptive AUTOSAR, show support for all the presented mechanisms. This shows that the automotive industry have realized that cyber-security will be a major challenge in the development of future connected autonomous vehicles and that is aims to adopt a strong security posture.

About this research paper

What this paper is about

Prior research on security mechanisms for the connected vehicle focuses on the current heterogeneous distributed embedded electronic architecture (EEA). However, due to increased demands on data-processing and bandwidth from future functionality, such as automated driving and sophisticated infotainment systems, vehicle manufacturers are now moving towards centralisation of functionality in one or more integrated electronic control units (ECUs). In this thesis, we identify security issues related to centralization of functionality in the automotive domain, and investigate mechanisms which mitigate these issues. Further, we investigate the support, both in hardware and software, for these mechanisms in the current and future automotive embedded systems. To identify security issues related to the integrated EEA and to identify mechanisms which mitigate these issues, a literature review of other domains was performed. The AUTomotive Open System ARchitecture (AUTOSAR) development partnership has specified a software architecture with the aim of standardizing software development in the automotive industry. We have investigate whether the identified mechanisms are specified in the AUTOSAR framework by interviewing AUTOSAR specialists. Some form of hardware feature is typically needed to be able to use the mechanisms. Therefore, hardware requirements for each mechanism were identified and used to evaluate the support present in ECU hardware. We have found that there was lacking support for most of the identified mechanisms, in both the present ECU hardware as well as in the current AUTOSAR platform. However, the hardware targeted to be used in future ECUs, along with the next-generation software standard Adaptive AUTOSAR, show support for all the presented mechanisms. This shows that the automotive industry have realized that cyber-security will be a major challenge in the development of future connected autonomous vehicles and that is aims to adopt a strong security posture.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Prior research on security mechanisms for the connected vehicle focuses on the current heterogeneous distributed embedded electronic architecture (EEA). However, due to increased demands on data-processing and bandwidth from future functionality, such as automated driving and sophisticated infotainment systems, vehicle manufacturers are now moving towards centralisation of functionality in one or more integrated electronic control units (ECUs). In this thesis, we identify security issues related to centralization of functionality in the automotive domain, and investigate mechanisms which mitigate these issues. Further, we investigate the support, both in hardware and software, for these mechanisms in the current and future automotive embedded systems. To identify security issues related to the integrated EEA and to identify mechanisms which mitigate these issues, a literature review of other domains was performed. The AUTomotive Open System ARchitecture (AUTOSAR) development partnership has specified a software architecture with the aim of standardizing software development in the automotive industry. We have investigate whether the identified mechanisms are specified in the AUTOSAR framework by interviewing AUTOSAR specialists. Some form of hardware feature is typically needed to be able to use the mechanisms. Therefore, hardware requirements for each mechanism were identified and used to evaluate the support present in ECU hardware. We have found that there was lacking support for most of the identified mechanisms, in both the present ECU hardware as well as in the current AUTOSAR platform. However, the hardware targeted to be used in future ECUs, along with the next-generation software standard Adaptive AUTOSAR, show support for all the presented mechanisms. This shows that the automotive industry have realized that cyber-security will be a major challenge in the development of future connected autonomous vehicles and that is aims to adopt a strong security posture.

Key concepts: AUTOSAR, Automotive industry, Embedded system, Software, Computer science, Engineering, Software architecture, Software deployment

Related papers

Back to paper searchBrowse research topicsOriginal source
Evaluation of Security Mechanisms for an Integrated Automotive System Architecture — Research Paper | ScholarLens