Adaptive and Context-Aware Authentication Framework Using Edge AI and Blockchain in Future Vehicular Networks ↔ A Novel Authentication Systems in Vehicular Communication: Challenges and Future Directions

EARLIER¶0-5

STAP Journal of Security Risk Management Vol.2024, No.1 45 Adaptive and Context-Aware Authentication Framework Using Edge AI and Blockchain in Future Vehicular Networks Received: 05-07-2024 Revised: 14-11-2024 Accepted: 15-11-2024 Published: 16-11-2024 Vol.2024, No.1 DOI: https://doi.org/10.6318 0/jsrm.thestap.2024.1.3 *Corresponding author. Email: aitizaz.ali@apu.edu.my Orcid: https://orcid.org/0000- 0002-4853-5093 This is an open access article under the CC BY 4.0 license (http://creativecommon s.org/licenses/by/4.0/ ). Published by STAP Publisher. A B S T R A C T The rise of connected and autonomous vehicles (CAVs) within intelligent transportation systems has introduced new demands for real-time, scalable, and privacy-preserving authentication mechanisms. Traditional authentication methods, such as Public Key Infrastructure (PKI), are often insufficient in highly dynamic vehicular environments due to their reliance on static credentials and centralized control. This paper proposes an adaptive and context-aware authentication framework that integrates Edge Artificial Intelligence (AI) with blockchain technology to secure vehicular communication. The framework leverages edge- based AI models to assess driver behavior and contextual signals in real time, generating dynamic trust scores for authentication. These scores are verified and recorded through a permissioned blockchain, ensuring tamper-proof identity validation and decentralized access control. The proposed system addresses key challenges including low latency, dynamic trust evaluation, and conditional privacy. Through detailed architectural design and security analysis, this work highlights the potential of hybrid AI-blockchain models to enhance the security, scalability, and accountability of future vehicular networks. Keywords: Vehicular networks, federated learning, V2X security, Edge AI, blockchain authentication, context-aware systems, intelligent transportation systems, real-time authentication, privacy preservation, decentralized identity, future mobility, and 6G networks. STAP Journal of Security Risk Management https://www.jsrm.thestap.com/ How to cite the article Ali, A. (2024). Adaptive and Context-Aware Authentication Framework Using Edge AI and Blockchain in Future Vehicular Networks. STAP Journal of Security Risk Management, 2024(1), 45–56. https://doi.org/10.63180/jsrm.thestap.2024.1.3

EARLIER¶4-8

Keywords: Vehicular networks, federated learning, V2X security, Edge AI, blockchain authentication, context-aware systems, intelligent transportation systems, real-time authentication, privacy preservation, decentralized identity, future mobility, and 6G networks. STAP Journal of Security Risk Management https://www.jsrm.thestap.com/ How to cite the article Ali, A. (2024). Adaptive and Context-Aware Authentication Framework Using Edge AI and Blockchain in Future Vehicular Networks. STAP Journal of Security Risk Management, 2024(1), 45–56. https://doi.org/10.63180/jsrm.thestap.2024.1.3 STAP Journal of Security Risk Management Vol.2024, No.1 46 1. Introduction The evolution of intelligent transportation systems (ITS) and the proliferation of connected and autonomous vehicles (CAVs) have revolutionized modern vehicular communication [1, 2]. Vehicles now routinely exchange information with other vehicles (V2V), infrastructure (V2I), pedestrians (V2P), and broader networks (V2X), enabling real-time data sharing for enhanced road safety, traffic management, and passenger experience[3, 4]. However, this increasing level of connectivity introduces significant security vulnerabilities, particularly in the domain of authentication, which is a foundational requirement for establishing trust in vehicular environments [5]. Traditional authentication schemes, such as those based on Public Key Infrastructure (PKI), offer strong cryptographic assurances but fall short in meeting the real-time responsiveness, dynamic adaptability, and privacy requirements of vehicular networks [6, 7]. Static credential-based systems are vulnerable to impersonation, identity spoofing, and certificate-based attacks, particularly in high-mobility scenarios with rapidly changing network topologies. Moreover, centralized authentication infrastructures present scalability bottlenecks and single points of failure that compromise system robustness [8, 9]. Recent advances in edge computing and artificial intelligence (AI) have introduced promising avenues for context-aware authentication. Edge AI allows real-time processing of behavioral and contextual data—such as speed, acceleration patterns, and location trajectories—enabling systems to assess trustworthiness based on dynamic patterns rather than static credentials. Concurrently, blockchain technology offers a decentralized and tamper-proof infrastructure for managing identity, revocation lists, and access control through smart contracts, thereby enhancing accountability and transparency. In this paper, we propose a novel authentication framework that combines Edge AI and blockchain to provide adaptive, privacy-preserving, and decentralized security for future vehicular networks. The framework continuously evaluates vehicle behavior at the edge and makes authentication decisions using lightweight machine learning models. These decisions are recorded and verified through a permissioned blockchain network, ensuring that the system remains auditable and resistant to manipulation.

EARLIER¶4-8

Keywords: Vehicular networks, federated learning, V2X security, Edge AI, blockchain authentication, context-aware systems, intelligent transportation systems, real-time authentication, privacy preservation, decentralized identity, future mobility, and 6G networks. STAP Journal of Security Risk Management https://www.jsrm.thestap.com/ How to cite the article Ali, A. (2024). Adaptive and Context-Aware Authentication Framework Using Edge AI and Blockchain in Future Vehicular Networks. STAP Journal of Security Risk Management, 2024(1), 45–56. https://doi.org/10.63180/jsrm.thestap.2024.1.3 STAP Journal of Security Risk Management Vol.2024, No.1 46 1. Introduction The evolution of intelligent transportation systems (ITS) and the proliferation of connected and autonomous vehicles (CAVs) have revolutionized modern vehicular communication [1, 2]. Vehicles now routinely exchange information with other vehicles (V2V), infrastructure (V2I), pedestrians (V2P), and broader networks (V2X), enabling real-time data sharing for enhanced road safety, traffic management, and passenger experience[3, 4]. However, this increasing level of connectivity introduces significant security vulnerabilities, particularly in the domain of authentication, which is a foundational requirement for establishing trust in vehicular environments [5]. Traditional authentication schemes, such as those based on Public Key Infrastructure (PKI), offer strong cryptographic assurances but fall short in meeting the real-time responsiveness, dynamic adaptability, and privacy requirements of vehicular networks [6, 7]. Static credential-based systems are vulnerable to impersonation, identity spoofing, and certificate-based attacks, particularly in high-mobility scenarios with rapidly changing network topologies. Moreover, centralized authentication infrastructures present scalability bottlenecks and single points of failure that compromise system robustness [8, 9]. Recent advances in edge computing and artificial intelligence (AI) have introduced promising avenues for context-aware authentication. Edge AI allows real-time processing of behavioral and contextual data—such as speed, acceleration patterns, and location trajectories—enabling systems to assess trustworthiness based on dynamic patterns rather than static credentials. Concurrently, blockchain technology offers a decentralized and tamper-proof infrastructure for managing identity, revocation lists, and access control through smart contracts, thereby enhancing accountability and transparency. In this paper, we propose a novel authentication framework that combines Edge AI and blockchain to provide adaptive, privacy-preserving, and decentralized security for future vehicular networks. The framework continuously evaluates vehicle behavior at the edge and makes authentication decisions using lightweight machine learning models. These decisions are recorded and verified through a permissioned blockchain network, ensuring that the system remains auditable and resistant to manipulation.

EARLIER¶15-19

STAP Journal of Security Risk Management Vol.2024, No.1 48 Figure 2. Proposed Framework 2.2 Authentication Process Flow The authentication process initiates when a vehicle captures real-time contextual data and securely transmits it to a nearby RSU. The edge AI engine at the RSU evaluates this data using a behavior model trained to distinguish between legitimate and anomalous patterns. The engine generates a dynamic trust score that indicates whether the current behavior aligns with the historical profile of the vehicle. Upon passing a predefined trust threshold, the system invokes a smart contract on the blockchain. This contract verifies the digital credentials, behavioral hash, and temporal consistency of the request. If validated, the contract authorizes the vehicle’s access to network resources or infrastructure components. The outcome of the authentication process—along with the associated trust score and relevant metadata—is immutably logged on the blockchain, contributing to a tamper-proof audit trail. Vehicles that consistently deviate from expected behaviors may be flagged and added to revocation lists or be subjected to more stringent verification steps. 2.3 Features and Advantages This architecture provides several key benefits for next-generation vehicular networks. First, it supports context-aware authentication by considering environmental, temporal, and behavioral factors. The system dynamically adjusts authentication strictness based on risk, such as inclement weather or high-speed driving conditions. Second, its adaptive capability is achieved through continual model updates using federated learning, allowing vehicles to refine their behavioral profiles without exposing raw data. Decentralized trust is ensured through the blockchain layer, which eliminates the reliance on a central authority and provides immutable logs of authentication events. This enhances both accountability and fault tolerance. Moreover, the use of edge computing ensures that latency is minimized, making the framework suitable for time-sensitive applications like collision avoidance and lane merging. Finally, privacy is preserved as raw behavioral data never leaves the vehicle or RSU; only hashed behavioral representations are shared and stored, preventing identity exposure and profiling. 3. Classification of Authentication Systems STAP Journal of Security Risk Management Vol.2024, No.1 49 Authentication in vehicular communication is essential to ensure the reliability and integrity of messages exchanged between vehicles (V2V), with infrastructure (V2I), and broader networks (V2X), as shown in Figure 3. This section classifies authenti- cation systems used in vehicular networks, outlining their principles, advantages, and limitations. 3.1 Group-Based Authentication Group-based authentication involves authenticating vehicles as members of a logical group using shared keys or group signatures. This approach reduces the need to authenticate each vehicle individually [10-14]. It offers significant improvements in efficiency and scalability, especially in dense vehicular environments, by minimizing cryptographic operations for individual vehicles within the same group [15-17]. Managing group keys and ensuring secure re-keying after member changes can be complex [18, 19]. Moreover, compromised vehicles within the group can pose internal threats that are harder to isolate [20-22]. 3.2 PKI-Based Authentication Systems Public Key Infrastructure (PKI) is a well-established method in vehicular networks. Each vehicle is issued a digital certificate by a trusted Certificate Authority (CA), enabling message signing and verification through public-private key pairs [23, 24].

EARLIER¶15-19

STAP Journal of Security Risk Management Vol.2024, No.1 48 Figure 2. Proposed Framework 2.2 Authentication Process Flow The authentication process initiates when a vehicle captures real-time contextual data and securely transmits it to a nearby RSU. The edge AI engine at the RSU evaluates this data using a behavior model trained to distinguish between legitimate and anomalous patterns. The engine generates a dynamic trust score that indicates whether the current behavior aligns with the historical profile of the vehicle. Upon passing a predefined trust threshold, the system invokes a smart contract on the blockchain. This contract verifies the digital credentials, behavioral hash, and temporal consistency of the request. If validated, the contract authorizes the vehicle’s access to network resources or infrastructure components. The outcome of the authentication process—along with the associated trust score and relevant metadata—is immutably logged on the blockchain, contributing to a tamper-proof audit trail. Vehicles that consistently deviate from expected behaviors may be flagged and added to revocation lists or be subjected to more stringent verification steps. 2.3 Features and Advantages This architecture provides several key benefits for next-generation vehicular networks. First, it supports context-aware authentication by considering environmental, temporal, and behavioral factors. The system dynamically adjusts authentication strictness based on risk, such as inclement weather or high-speed driving conditions. Second, its adaptive capability is achieved through continual model updates using federated learning, allowing vehicles to refine their behavioral profiles without exposing raw data. Decentralized trust is ensured through the blockchain layer, which eliminates the reliance on a central authority and provides immutable logs of authentication events. This enhances both accountability and fault tolerance. Moreover, the use of edge computing ensures that latency is minimized, making the framework suitable for time-sensitive applications like collision avoidance and lane merging. Finally, privacy is preserved as raw behavioral data never leaves the vehicle or RSU; only hashed behavioral representations are shared and stored, preventing identity exposure and profiling. 3. Classification of Authentication Systems STAP Journal of Security Risk Management Vol.2024, No.1 49 Authentication in vehicular communication is essential to ensure the reliability and integrity of messages exchanged between vehicles (V2V), with infrastructure (V2I), and broader networks (V2X), as shown in Figure 3. This section classifies authenti- cation systems used in vehicular networks, outlining their principles, advantages, and limitations. 3.1 Group-Based Authentication Group-based authentication involves authenticating vehicles as members of a logical group using shared keys or group signatures. This approach reduces the need to authenticate each vehicle individually [10-14]. It offers significant improvements in efficiency and scalability, especially in dense vehicular environments, by minimizing cryptographic operations for individual vehicles within the same group [15-17]. Managing group keys and ensuring secure re-keying after member changes can be complex [18, 19]. Moreover, compromised vehicles within the group can pose internal threats that are harder to isolate [20-22]. 3.2 PKI-Based Authentication Systems Public Key Infrastructure (PKI) is a well-established method in vehicular networks. Each vehicle is issued a digital certificate by a trusted Certificate Authority (CA), enabling message signing and verification through public-private key pairs [23, 24].

EARLIER¶17-21

3. Classification of Authentication Systems STAP Journal of Security Risk Management Vol.2024, No.1 49 Authentication in vehicular communication is essential to ensure the reliability and integrity of messages exchanged between vehicles (V2V), with infrastructure (V2I), and broader networks (V2X), as shown in Figure 3. This section classifies authenti- cation systems used in vehicular networks, outlining their principles, advantages, and limitations. 3.1 Group-Based Authentication Group-based authentication involves authenticating vehicles as members of a logical group using shared keys or group signatures. This approach reduces the need to authenticate each vehicle individually [10-14]. It offers significant improvements in efficiency and scalability, especially in dense vehicular environments, by minimizing cryptographic operations for individual vehicles within the same group [15-17]. Managing group keys and ensuring secure re-keying after member changes can be complex [18, 19]. Moreover, compromised vehicles within the group can pose internal threats that are harder to isolate [20-22]. 3.2 PKI-Based Authentication Systems Public Key Infrastructure (PKI) is a well-established method in vehicular networks. Each vehicle is issued a digital certificate by a trusted Certificate Authority (CA), enabling message signing and verification through public-private key pairs [23, 24]. Figure 3. Classification of Authentication Systems PKI-based systems provide strong cryptographic guarantees for authenticity and message integrity [25-27]. They are wellstandardized and have been adopted in frameworks like IEEE 1609.2, making them widely deployable and trustworthy [28-32]. These systems suffer from high overhead in certificate revocation and renewal, particularly in highly dynamic environments. They also offer limited privacy since vehicle identities can be traced via certificates, and a compromised CA could potentially disrupt the entire system [4, 33, and 34]. 3.3 Blockchain-Based Authentication STAP Journal of Security Risk Management Vol.2024, No.1 50 Blockchain technology enables decentralized identity verification and trust establishment through distributed ledgers and smart contracts. Blockchain eliminates the need for a centralized authority, providing a tamper-proof, transparent, and auditable authentication mechanism [35–39]. This is particularly beneficial for building long- term trust and secure identity management in vehicular networks. Despite its benefits, blockchain introduces latency and computational burdens that make it unsuitable for real-time vehicular communications. Additionally, scalability remains a concern as the number of vehicles and transactions increases.

EARLIER¶18-22

3.1 Group-Based Authentication Group-based authentication involves authenticating vehicles as members of a logical group using shared keys or group signatures. This approach reduces the need to authenticate each vehicle individually [10-14]. It offers significant improvements in efficiency and scalability, especially in dense vehicular environments, by minimizing cryptographic operations for individual vehicles within the same group [15-17]. Managing group keys and ensuring secure re-keying after member changes can be complex [18, 19]. Moreover, compromised vehicles within the group can pose internal threats that are harder to isolate [20-22]. 3.2 PKI-Based Authentication Systems Public Key Infrastructure (PKI) is a well-established method in vehicular networks. Each vehicle is issued a digital certificate by a trusted Certificate Authority (CA), enabling message signing and verification through public-private key pairs [23, 24]. Figure 3. Classification of Authentication Systems PKI-based systems provide strong cryptographic guarantees for authenticity and message integrity [25-27]. They are wellstandardized and have been adopted in frameworks like IEEE 1609.2, making them widely deployable and trustworthy [28-32]. These systems suffer from high overhead in certificate revocation and renewal, particularly in highly dynamic environments. They also offer limited privacy since vehicle identities can be traced via certificates, and a compromised CA could potentially disrupt the entire system [4, 33, and 34]. 3.3 Blockchain-Based Authentication STAP Journal of Security Risk Management Vol.2024, No.1 50 Blockchain technology enables decentralized identity verification and trust establishment through distributed ledgers and smart contracts. Blockchain eliminates the need for a centralized authority, providing a tamper-proof, transparent, and auditable authentication mechanism [35–39]. This is particularly beneficial for building long- term trust and secure identity management in vehicular networks. Despite its benefits, blockchain introduces latency and computational burdens that make it unsuitable for real-time vehicular communications. Additionally, scalability remains a concern as the number of vehicles and transactions increases. 3.4 Pseudonym-Based and Privacy-Preserving Authentication This category involves frequently changing pseudonyms and privacy techniques to prevent vehicle tracking and identity exposure [40–46]. By obfuscating vehicle identities through pseudonyms and mix-zones, these systems offer a high degree of privacy protection while maintaining authentication validity [47–49]. They are also compatible with most existing V2X infrastructures [50–54]. Pseudonym management requires secure synchronization and timely updates. If these mechanisms fail or are compromised, the vehicle’s privacy can be exposed or even exploited [48]. 3.5 AI/ML-Based Behavioral Authentication

EARLIER¶18-22

3.1 Group-Based Authentication Group-based authentication involves authenticating vehicles as members of a logical group using shared keys or group signatures. This approach reduces the need to authenticate each vehicle individually [10-14]. It offers significant improvements in efficiency and scalability, especially in dense vehicular environments, by minimizing cryptographic operations for individual vehicles within the same group [15-17]. Managing group keys and ensuring secure re-keying after member changes can be complex [18, 19]. Moreover, compromised vehicles within the group can pose internal threats that are harder to isolate [20-22]. 3.2 PKI-Based Authentication Systems Public Key Infrastructure (PKI) is a well-established method in vehicular networks. Each vehicle is issued a digital certificate by a trusted Certificate Authority (CA), enabling message signing and verification through public-private key pairs [23, 24]. Figure 3. Classification of Authentication Systems PKI-based systems provide strong cryptographic guarantees for authenticity and message integrity [25-27]. They are wellstandardized and have been adopted in frameworks like IEEE 1609.2, making them widely deployable and trustworthy [28-32]. These systems suffer from high overhead in certificate revocation and renewal, particularly in highly dynamic environments. They also offer limited privacy since vehicle identities can be traced via certificates, and a compromised CA could potentially disrupt the entire system [4, 33, and 34]. 3.3 Blockchain-Based Authentication STAP Journal of Security Risk Management Vol.2024, No.1 50 Blockchain technology enables decentralized identity verification and trust establishment through distributed ledgers and smart contracts. Blockchain eliminates the need for a centralized authority, providing a tamper-proof, transparent, and auditable authentication mechanism [35–39]. This is particularly beneficial for building long- term trust and secure identity management in vehicular networks. Despite its benefits, blockchain introduces latency and computational burdens that make it unsuitable for real-time vehicular communications. Additionally, scalability remains a concern as the number of vehicles and transactions increases. 3.4 Pseudonym-Based and Privacy-Preserving Authentication This category involves frequently changing pseudonyms and privacy techniques to prevent vehicle tracking and identity exposure [40–46]. By obfuscating vehicle identities through pseudonyms and mix-zones, these systems offer a high degree of privacy protection while maintaining authentication validity [47–49]. They are also compatible with most existing V2X infrastructures [50–54]. Pseudonym management requires secure synchronization and timely updates. If these mechanisms fail or are compromised, the vehicle’s privacy can be exposed or even exploited [48]. 3.5 AI/ML-Based Behavioral Authentication

EARLIER¶16-20

2.3 Features and Advantages This architecture provides several key benefits for next-generation vehicular networks. First, it supports context-aware authentication by considering environmental, temporal, and behavioral factors. The system dynamically adjusts authentication strictness based on risk, such as inclement weather or high-speed driving conditions. Second, its adaptive capability is achieved through continual model updates using federated learning, allowing vehicles to refine their behavioral profiles without exposing raw data. Decentralized trust is ensured through the blockchain layer, which eliminates the reliance on a central authority and provides immutable logs of authentication events. This enhances both accountability and fault tolerance. Moreover, the use of edge computing ensures that latency is minimized, making the framework suitable for time-sensitive applications like collision avoidance and lane merging. Finally, privacy is preserved as raw behavioral data never leaves the vehicle or RSU; only hashed behavioral representations are shared and stored, preventing identity exposure and profiling. 3. Classification of Authentication Systems STAP Journal of Security Risk Management Vol.2024, No.1 49 Authentication in vehicular communication is essential to ensure the reliability and integrity of messages exchanged between vehicles (V2V), with infrastructure (V2I), and broader networks (V2X), as shown in Figure 3. This section classifies authenti- cation systems used in vehicular networks, outlining their principles, advantages, and limitations. 3.1 Group-Based Authentication Group-based authentication involves authenticating vehicles as members of a logical group using shared keys or group signatures. This approach reduces the need to authenticate each vehicle individually [10-14]. It offers significant improvements in efficiency and scalability, especially in dense vehicular environments, by minimizing cryptographic operations for individual vehicles within the same group [15-17]. Managing group keys and ensuring secure re-keying after member changes can be complex [18, 19]. Moreover, compromised vehicles within the group can pose internal threats that are harder to isolate [20-22]. 3.2 PKI-Based Authentication Systems Public Key Infrastructure (PKI) is a well-established method in vehicular networks. Each vehicle is issued a digital certificate by a trusted Certificate Authority (CA), enabling message signing and verification through public-private key pairs [23, 24]. Figure 3. Classification of Authentication Systems PKI-based systems provide strong cryptographic guarantees for authenticity and message integrity [25-27]. They are wellstandardized and have been adopted in frameworks like IEEE 1609.2, making them widely deployable and trustworthy [28-32]. These systems suffer from high overhead in certificate revocation and renewal, particularly in highly dynamic environments. They also offer limited privacy since vehicle identities can be traced via certificates, and a compromised CA could potentially disrupt the entire system [4, 33, and 34].

EARLIER¶19-23

3.2 PKI-Based Authentication Systems Public Key Infrastructure (PKI) is a well-established method in vehicular networks. Each vehicle is issued a digital certificate by a trusted Certificate Authority (CA), enabling message signing and verification through public-private key pairs [23, 24]. Figure 3. Classification of Authentication Systems PKI-based systems provide strong cryptographic guarantees for authenticity and message integrity [25-27]. They are wellstandardized and have been adopted in frameworks like IEEE 1609.2, making them widely deployable and trustworthy [28-32]. These systems suffer from high overhead in certificate revocation and renewal, particularly in highly dynamic environments. They also offer limited privacy since vehicle identities can be traced via certificates, and a compromised CA could potentially disrupt the entire system [4, 33, and 34]. 3.3 Blockchain-Based Authentication STAP Journal of Security Risk Management Vol.2024, No.1 50 Blockchain technology enables decentralized identity verification and trust establishment through distributed ledgers and smart contracts. Blockchain eliminates the need for a centralized authority, providing a tamper-proof, transparent, and auditable authentication mechanism [35–39]. This is particularly beneficial for building long- term trust and secure identity management in vehicular networks. Despite its benefits, blockchain introduces latency and computational burdens that make it unsuitable for real-time vehicular communications. Additionally, scalability remains a concern as the number of vehicles and transactions increases. 3.4 Pseudonym-Based and Privacy-Preserving Authentication This category involves frequently changing pseudonyms and privacy techniques to prevent vehicle tracking and identity exposure [40–46]. By obfuscating vehicle identities through pseudonyms and mix-zones, these systems offer a high degree of privacy protection while maintaining authentication validity [47–49]. They are also compatible with most existing V2X infrastructures [50–54]. Pseudonym management requires secure synchronization and timely updates. If these mechanisms fail or are compromised, the vehicle’s privacy can be exposed or even exploited [48]. 3.5 AI/ML-Based Behavioral Authentication Machine learning approaches authenticate vehicles based on behavioral data such as driving style, speed, and trajectory, enabling more dynamic and adaptive verification. These systems enable continuous and context-aware authentication, improving security without requiring explicit user input [50]. They are effective in detecting anomalies or spoofed behaviors that traditional methods might miss. Their performance heavily depends on the availability of large, diverse training datasets. They can also be affected by legitimate changes in behavior (e.g., different drivers), leading to false positives or reduced accuracy.

EARLIER¶20-25

Figure 3. Classification of Authentication Systems PKI-based systems provide strong cryptographic guarantees for authenticity and message integrity [25-27]. They are wellstandardized and have been adopted in frameworks like IEEE 1609.2, making them widely deployable and trustworthy [28-32]. These systems suffer from high overhead in certificate revocation and renewal, particularly in highly dynamic environments. They also offer limited privacy since vehicle identities can be traced via certificates, and a compromised CA could potentially disrupt the entire system [4, 33, and 34]. 3.3 Blockchain-Based Authentication STAP Journal of Security Risk Management Vol.2024, No.1 50 Blockchain technology enables decentralized identity verification and trust establishment through distributed ledgers and smart contracts. Blockchain eliminates the need for a centralized authority, providing a tamper-proof, transparent, and auditable authentication mechanism [35–39]. This is particularly beneficial for building long- term trust and secure identity management in vehicular networks. Despite its benefits, blockchain introduces latency and computational burdens that make it unsuitable for real-time vehicular communications. Additionally, scalability remains a concern as the number of vehicles and transactions increases. 3.4 Pseudonym-Based and Privacy-Preserving Authentication This category involves frequently changing pseudonyms and privacy techniques to prevent vehicle tracking and identity exposure [40–46]. By obfuscating vehicle identities through pseudonyms and mix-zones, these systems offer a high degree of privacy protection while maintaining authentication validity [47–49]. They are also compatible with most existing V2X infrastructures [50–54]. Pseudonym management requires secure synchronization and timely updates. If these mechanisms fail or are compromised, the vehicle’s privacy can be exposed or even exploited [48]. 3.5 AI/ML-Based Behavioral Authentication Machine learning approaches authenticate vehicles based on behavioral data such as driving style, speed, and trajectory, enabling more dynamic and adaptive verification. These systems enable continuous and context-aware authentication, improving security without requiring explicit user input [50]. They are effective in detecting anomalies or spoofed behaviors that traditional methods might miss. Their performance heavily depends on the availability of large, diverse training datasets. They can also be affected by legitimate changes in behavior (e.g., different drivers), leading to false positives or reduced accuracy. 3.6 Post-Quantum Cryptography-Based Authentication With quantum computing on the horizon, researchers are exploring post-quantum cryptographic techniques for long-term security. These schemes promise resilience against quantum-enabled attacks, offering future-proof authentication methods crucial for the longevity of vehicular systems deployed today [49]. Post-quantum algorithms are currently computationally expensive and often unsuitable for real-time or resource-constrained environments. Their integration into vehicular systems is still in early experimental stages. 4. Results and Evaluation To evaluate the feasibility and effectiveness of the proposed authentication framework, we conducted a simulation-based performance analysis using a custom vehicular net- work testbed built with the Veins framework integrated with SUMO and OMNeT++. A combination of real-time vehicle mobility patterns and synthetic behavioral datasets was used to simulate various vehicular authentication scenarios under differing loads, speeds, and trust contexts.

EARLIER¶20-24

Figure 3. Classification of Authentication Systems PKI-based systems provide strong cryptographic guarantees for authenticity and message integrity [25-27]. They are wellstandardized and have been adopted in frameworks like IEEE 1609.2, making them widely deployable and trustworthy [28-32]. These systems suffer from high overhead in certificate revocation and renewal, particularly in highly dynamic environments. They also offer limited privacy since vehicle identities can be traced via certificates, and a compromised CA could potentially disrupt the entire system [4, 33, and 34]. 3.3 Blockchain-Based Authentication STAP Journal of Security Risk Management Vol.2024, No.1 50 Blockchain technology enables decentralized identity verification and trust establishment through distributed ledgers and smart contracts. Blockchain eliminates the need for a centralized authority, providing a tamper-proof, transparent, and auditable authentication mechanism [35–39]. This is particularly beneficial for building long- term trust and secure identity management in vehicular networks. Despite its benefits, blockchain introduces latency and computational burdens that make it unsuitable for real-time vehicular communications. Additionally, scalability remains a concern as the number of vehicles and transactions increases. 3.4 Pseudonym-Based and Privacy-Preserving Authentication This category involves frequently changing pseudonyms and privacy techniques to prevent vehicle tracking and identity exposure [40–46]. By obfuscating vehicle identities through pseudonyms and mix-zones, these systems offer a high degree of privacy protection while maintaining authentication validity [47–49]. They are also compatible with most existing V2X infrastructures [50–54]. Pseudonym management requires secure synchronization and timely updates. If these mechanisms fail or are compromised, the vehicle’s privacy can be exposed or even exploited [48]. 3.5 AI/ML-Based Behavioral Authentication Machine learning approaches authenticate vehicles based on behavioral data such as driving style, speed, and trajectory, enabling more dynamic and adaptive verification. These systems enable continuous and context-aware authentication, improving security without requiring explicit user input [50]. They are effective in detecting anomalies or spoofed behaviors that traditional methods might miss. Their performance heavily depends on the availability of large, diverse training datasets. They can also be affected by legitimate changes in behavior (e.g., different drivers), leading to false positives or reduced accuracy. 3.6 Post-Quantum Cryptography-Based Authentication With quantum computing on the horizon, researchers are exploring post-quantum cryptographic techniques for long-term security. These schemes promise resilience against quantum-enabled attacks, offering future-proof authentication methods crucial for the longevity of vehicular systems deployed today [49]. Post-quantum algorithms are currently computationally expensive and often unsuitable for real-time or resource-constrained environments. Their integration into vehicular systems is still in early experimental stages.

EARLIER¶22-26

3.4 Pseudonym-Based and Privacy-Preserving Authentication This category involves frequently changing pseudonyms and privacy techniques to prevent vehicle tracking and identity exposure [40–46]. By obfuscating vehicle identities through pseudonyms and mix-zones, these systems offer a high degree of privacy protection while maintaining authentication validity [47–49]. They are also compatible with most existing V2X infrastructures [50–54]. Pseudonym management requires secure synchronization and timely updates. If these mechanisms fail or are compromised, the vehicle’s privacy can be exposed or even exploited [48]. 3.5 AI/ML-Based Behavioral Authentication Machine learning approaches authenticate vehicles based on behavioral data such as driving style, speed, and trajectory, enabling more dynamic and adaptive verification. These systems enable continuous and context-aware authentication, improving security without requiring explicit user input [50]. They are effective in detecting anomalies or spoofed behaviors that traditional methods might miss. Their performance heavily depends on the availability of large, diverse training datasets. They can also be affected by legitimate changes in behavior (e.g., different drivers), leading to false positives or reduced accuracy. 3.6 Post-Quantum Cryptography-Based Authentication With quantum computing on the horizon, researchers are exploring post-quantum cryptographic techniques for long-term security. These schemes promise resilience against quantum-enabled attacks, offering future-proof authentication methods crucial for the longevity of vehicular systems deployed today [49]. Post-quantum algorithms are currently computationally expensive and often unsuitable for real-time or resource-constrained environments. Their integration into vehicular systems is still in early experimental stages. 4. Results and Evaluation To evaluate the feasibility and effectiveness of the proposed authentication framework, we conducted a simulation-based performance analysis using a custom vehicular net- work testbed built with the Veins framework integrated with SUMO and OMNeT++. A combination of real-time vehicle mobility patterns and synthetic behavioral datasets was used to simulate various vehicular authentication scenarios under differing loads, speeds, and trust contexts. 4.1 Experimental Setup The simulated network includes 50 to 200 vehicles operating under V2X conditions in an urban grid. Edge RSUs were deployed at intersections with access to GPU- accelerated inference engines, and the blockchain network was simulated using a Hyperledger Fabric testbed with five validator nodes. The Edge AI model was implemented using a lightweight random forest classifier trained on behavioral features such as braking frequency, acceleration-deceleration patterns, and trajectory deviation. The framework was benchmarked against three baseline authentication models: • Traditional PKI-based authentication • Pseudonym-based conditional privacy authentication • AI-only (non-blockchain) behavioral trust models

EARLIER¶22-26

3.4 Pseudonym-Based and Privacy-Preserving Authentication This category involves frequently changing pseudonyms and privacy techniques to prevent vehicle tracking and identity exposure [40–46]. By obfuscating vehicle identities through pseudonyms and mix-zones, these systems offer a high degree of privacy protection while maintaining authentication validity [47–49]. They are also compatible with most existing V2X infrastructures [50–54]. Pseudonym management requires secure synchronization and timely updates. If these mechanisms fail or are compromised, the vehicle’s privacy can be exposed or even exploited [48]. 3.5 AI/ML-Based Behavioral Authentication Machine learning approaches authenticate vehicles based on behavioral data such as driving style, speed, and trajectory, enabling more dynamic and adaptive verification. These systems enable continuous and context-aware authentication, improving security without requiring explicit user input [50]. They are effective in detecting anomalies or spoofed behaviors that traditional methods might miss. Their performance heavily depends on the availability of large, diverse training datasets. They can also be affected by legitimate changes in behavior (e.g., different drivers), leading to false positives or reduced accuracy. 3.6 Post-Quantum Cryptography-Based Authentication With quantum computing on the horizon, researchers are exploring post-quantum cryptographic techniques for long-term security. These schemes promise resilience against quantum-enabled attacks, offering future-proof authentication methods crucial for the longevity of vehicular systems deployed today [49]. Post-quantum algorithms are currently computationally expensive and often unsuitable for real-time or resource-constrained environments. Their integration into vehicular systems is still in early experimental stages. 4. Results and Evaluation To evaluate the feasibility and effectiveness of the proposed authentication framework, we conducted a simulation-based performance analysis using a custom vehicular net- work testbed built with the Veins framework integrated with SUMO and OMNeT++. A combination of real-time vehicle mobility patterns and synthetic behavioral datasets was used to simulate various vehicular authentication scenarios under differing loads, speeds, and trust contexts. 4.1 Experimental Setup The simulated network includes 50 to 200 vehicles operating under V2X conditions in an urban grid. Edge RSUs were deployed at intersections with access to GPU- accelerated inference engines, and the blockchain network was simulated using a Hyperledger Fabric testbed with five validator nodes. The Edge AI model was implemented using a lightweight random forest classifier trained on behavioral features such as braking frequency, acceleration-deceleration patterns, and trajectory deviation. The framework was benchmarked against three baseline authentication models: • Traditional PKI-based authentication • Pseudonym-based conditional privacy authentication • AI-only (non-blockchain) behavioral trust models

EARLIER¶37-40

Finally, large-scale testbeds and standardization efforts are needed to transition the framework from theoretical design to practical deployment. Future research should focus on building interoperable simulation environments and real-world trials involving heterogeneous vehicles, communication stacks, and infrastructure. Collaboration with industry stakeholders, regulatory bodies, and international standards organizations will be essential to validate performance, ensure legal compliance, and drive global adoption. 7. Conclusion This paper has presented an adaptive and context-aware authentication framework for future vehicular networks, leveraging Edge AI for behavioral analysis and blockchain technology for decentralized trust management. The framework is designed to address the core challenges of security, privacy, scalability, and latency that are critical in connected and autonomous vehicle (CAV) environments. By integrating edge-based machine learning models, the system enables real-time authentication decisions that are responsive to dynamic driving contexts. This adaptive capability ensures that trust is continuously assessed based on current behavioral patterns, reducing the risk of impersonation, spoofing, and misuse. The inclusion of a permissioned blockchain infrastructure supports immutable identity management and distributed revocation, thereby eliminating reliance on centralized authorities and enabling transparent, audit- ready trust operations. A comprehensive security and privacy analysis demonstrates the framework’s robustness against known vehicular attack vectors while preserving conditional anonymity through pseudonym-based cryptographic mechanisms and behavioral hashing. Furthermore, the architecture supports modular extensibility, allowing future integration of quantum-resilient algorithms, 6G edge connectivity, and personalized AI models. In conclusion, the proposed framework advances the state of vehicular authentication by combining intelligence, decentralization, and privacy preservation. It lays the groundwork for scalable, secure, and accountable communication in next-generation vehicular networks. Future efforts will focus on standardization, real-world validation, and the incorporation of multi-layered trust metrics to further enhance its resilience and practical adoption. Corresponding author Dr. Aitizaz Ali aitizaz.ali@apu.edu.my Acknowledgements NA. Funding No funding. Contributions A.A; Conceptualization, A.A; Investigation, A.A; Writing (Original Draft), A.A; Writing (Review and Editing) Supervision, A.A; Project Administration. Ethics declarations This article does not contain any studies with human participants or animals performed by any of the authors. Consent for publication Not applicable.