Adaptive chaotic map-based key extraction for efficient cross-layer authentication in VANETs

Vehicle-to-everything (V2X) communication is expected to offer users available and ultra-reliable transmission, particularly for critical applications related to safety and autonomy. In this context, establishing a secure and resilient authentication process with low latency and high functionality may not be achieved using conventional cryptographic methodologies due to their significant computation costs. Recent research has focused on employing the physical (PHY) characteristics of wireless channels to develop efficient discrimination techniques to overcome the shortcomings of crypto-based authentication. This paper presents a cross-layer authentication scheme for multicarrier communication, leveraging the spatially/temporally correlated wireless channel features to facilitate key verification without exposing its secrecy. By mapping the time-stamped hashed key and masking it with channel phase responses, we create a PHY-layer signature, allowing for verifying the sender's identity while employing the correlated channel responses between subcarriers to verify messages' integrity. Furthermore, we developed a Diffie-Hellman secret key extraction algorithm that employs the computationally intractable problems of the Chebyshev chaotic mapping for channel probing. Thus, terminals can extract high entropy shared keys that can be used to create dynamic PHY-layer signatures, supporting forward and backward secrecy. We evaluated the scheme's security strength against active/passive attacks. Besides theoretical analysis, we designed a 3-Dimensional (3D) scattering Doppler emulator to investigate the scheme's performance at different speeds of a moving vehicle and signal-to-noise ratios (SNRs) for a realistic vehicular channel. Theoretical and hardware implementation analyses proved the capability of the proposed scheme to support high detection probability at SNR >= 0 dB and speed <= 45 m/s.(c) 2022 The Author(s). Published by Elsevier Inc.