Integração de padrões de ataque, fraquezas e modelagem de ameaças para suporte à análise de segurança no domínio automotivo

Abstract

The growing interconnectivity among computing systems, embedded devices, and communication networks has intensified concerns regarding cybersecurity, particularly in critical domains such as autonomous vehicles. In this context, the ISO/SAE 21434 standard was developed to provide guidelines for Threat Analysis and Risk Assessment (TARA) throughout the automotive system lifecycle. To support this process, techniques such as the STRIDE model (Spoofing, Tampering, Repudiation, Information Disclosure, Denial of Service, and Elevation of Privilege) and attack tree analysis have been widely adopted to support the standard’s methodology. Complementarily, structured databases such as CAPEC (Common Attack Pattern Enumeration and Classification) and CWE (Common Weakness Enumeration) offer comprehensive catalogs of attack patterns and weaknesses, with the potential to enrich threat identification and security requirement definition. However, the lack of a systematic integration between these databases and normative models remains a gap in the literature, hindering their practical application. To address this issue, this work proposes a methodological approach to support cybersecurity risk analysis in autonomous automotive systems, in accordance with ISO/SAE 21434. The proposal aims to structurally integrate the STRIDE threat model, the CAPEC attack pattern catalog, and the CWE weakness database, supporting the identification, evaluation, and treatment stages of the TARA process. For this purpose, a data model was developed to represent the main attributes and relationships among these bases, along with a document-oriented database implemented in MongoDB, enabling efficient and contextaware queries. Attack feasibility and impact rating are inferred from CAPEC’s qualitative attributes, while the risk treatment is enhanced through mitigation strategies extracted from CWE, complementing operational aspects not explicitly addressed by the standard. The proposed methodology was evaluated through a case study simulating attacks on a vehicle’s headlight system via Bluetooth, demonstrating its practical applicability and its ability to provide traceability, technical foundation, and consistency to the cybersecurity risk analysis process.