Vehicular Communication Security: Multi-Channel and Multi-Factor Authentication

arXiv:2505.00340 · cs.CR · Submitted 2025-05-01 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Security Radio. Generated commentary on the latest security and cryptography papers.

Nadia: We're looking at the paper: "Vehicular Communication Security".

Elias: Here's the gist of the paper. Secure and reliable communications are crucial for Intelligent Transportation Systems (ITSs), where Vehicle-to-Infrastructure (V2I) communication plays a key role in enabling mobility-enhancing and safety-critical services Current V2I authentication relies on credential-based methods over wireless Non-Line-of-Sight (NLOS) channels, leaving them exposed to remote impersonation and proximity attacks To mitigate these risks, we propose a unified Multi-Channel, Multi-Factor Authentication (MFA) scheme that combines NLOS cryptographic credentials with a Line-of-Sight (LOS) visual channel Our approach leverages a challenge–response security paradigm: the infrastructure issues “challenges” and the vehicle’s headlights respond by flashing a structured sequence containing encoded security data Deep learning models on the infrastructure side then decode the embedded information to authenticate the vehicle Real-world experimental evaluations demonstrate high test accuracy, reaching an average of 95% and 96.6%, respectively, under various lighting, weather, speed, and distance conditions Scheme Design and Mechanism The proposed scheme introduces an authentication process with a fully implemented LOS communication and response mechanism This scheme integrates multiple factors across NLOS and LOS channels as shown in Figure 2 and consists of the following three phases 1.

Nadia: NLOS Phase: the vehicle uses the NLOS channel to transmit its unique security credential to a Registration Authority (RA) over a Transport Layer Security (TLS)-protected channel, as defined in IEEE 1609.2.1 This serves as the first authentication factor (something you know) 2.

Elias: LOS Phase: After verifying the first authentication factor, the infrastructure requests a second one (e.g., something you are or something you know) The RSU then sends the vehicle a randomized challenge, which the vehicle answers through the optical channel 3.

Paper summary: Nadia: So we're looking at this paper today called Vehicular Communication Security: Multi-Channel and Multi-Factor Authentication. Basically, they’re trying to make vehicle identification in those smart road systems really tough against people who are either far away or right there next to the car.

Elias: Exactly. The core idea here is that current ways of authenticating vehicles over wireless channels, especially non-line-of-sight ones, are pretty weak and can get you fooled by someone pretending to be your car or just being physically close.

Priya: So what they're proposing is this unified system that mixes a secure way of sending data over the air with something visual, using the vehicle’s headlights as a second layer of proof.

Nadia: Right, they combine those non-line-of-sight cryptographic credentials with a line-of-sight visual channel. It’s like having two different keys to unlock the car's identity instead of just one.

Elias: The mechanism they use is a challenge response paradigm, which means the infrastructure sends out some kind of challenge, and then the vehicle answers it by flashing its headlights with a specific sequence that has security data hidden in it.

Priya: That sounds like something you could actually see, which is interesting because relying on visual cues introduces a whole new set of challenges compared to just listening to radio signals.

Nadia: It does. Because the infrastructure side then uses deep learning models to decode that embedded information from the flashing sequence and verify the vehicle's identity. That’s how they claim it works for authentication.

Elias: They did some testing, and those real-world experimental evaluations showed high test accuracy, reaching an average of ninety-five percent and ninety-six point six point six percent under various conditions like different lighting, weather, speed, and distance.

Priya: Ninety-five to ninety-six percent accuracy sounds high for something operating in the real world where things change constantly; what kind of lighting conditions were they testing?

Nadia: They tested under different lighting conditions and environmental variables because they wanted to show that this wasn't just a lab trick. It showed feasibility in varied scenarios, which is important for practical deployment.

Elias: Thinking about the design, the scheme starts with an NLOS phase where the vehicle uses a non-line-of-sight channel to send its unique security credential to a Registration Authority over something protected by transport layer security, which is like tls as defined in ieee one thousand six hundred nine point two one.

Priya: So that first part, the NLOS part, is about proving who you are using a standard secure connection for that initial credential transmission. That's the "something you know" factor they mention.

Nadia: Right. After verifying that first factor, the infrastructure then requests a second one from the vehicle through what they call the LOS phase, which is where the actual visual challenge happens.

Paper summary: Elias: The RSU, which is like a roadside unit on the infrastructure side, sends this randomized challenge to the vehicle, and then you answer it using those optical channels through your headlights. That’s that second factor they are looking for.

Priya: So if someone tries to impersonate you remotely, they can't just guess the visual response because that response is tied to a challenge sent specifically to them in real time.

Nadia: It adds a physical presence check to the digital credential check, which addresses those remote impersonation and proximity attacks they were worried about initially.

Elias: And they frame this whole approach using threat modeling frameworks like stride for classifying risks like spoofing or tampering, and they use the doolev-yao model to define what an attacker can actually do in a network scenario.

Priya: That’s useful because it shows they didn't just build a system; they tried to anticipate specific attack types and mapped them against established security models.

Nadia: It’s about showing that you have to consider both the digital messages traveling through the air and what is physically observable in your immediate environment to be secure.

Elias: So, we've covered how they set up this multi-channel authentication scheme, combining cryptographic credentials with visual challenges. Where do we go from here?

Priya: I wonder if this high accuracy holds up when you consider the limitations they mention; what does the paper say is not covered by this specific MFA scheme?

Nadia: The paper points out that it's focused on authenticated communication within certain scenarios, like restricted areas or reserved lanes in intelligent transportation systems. It doesn't necessarily cover every single possible security threat across all types of vehicular communication setups.

Elias: That’s a key limitation; they are designing this for specific critical ITS use cases rather than claiming universal security across every possible vehicle-to-infrastructure link.

Priya: So, it’s a very targeted solution for high-stakes environments where the threats are well understood and defined by those threat models.

Nadia: Right. It’s not a blanket fix for all wireless communication security; it’s highly tailored to the V2I context they are focusing on.

Elias: So, to wrap up this discussion on Vehicular Communication Security: Multi-Channel and Multi-Factor Authentication, the authors are presenting a specific combination of NLOS credentials and LOS visual response as a way to bolster vehicle identity in V2I communication.

Priya: They’ve shown that combining these factors can yield high classification accuracy when tested across various real-world environmental changes.

Nadia: The implication is that for safety-critical services, you might need to look beyond just digital credentials and incorporate physical verification methods too.

Elias: It suggests that future work could involve looking at how this scheme scales with more complex traffic scenarios or integrating other types of visual confirmation methods.

Conclusion: Nadia: So we're wrapping up on "Vehicular Communication Security: Multi-Channel and Multi-Factor Authentication." This paper is about adding a second layer of security to how cars talk to road infrastructure.

Elias: Right, it’s taking those standard digital credentials used wirelessly and mixing them with something physical, using the actual lights on the car.

Nadia: The authors are looking at how this works against remote impersonation and proximity attacks in those smart transportation systems.

Priya: I want to talk about who wrote this. It's a solid team of researchers, clearly focused on both the security side and how we actually measure things in real-world conditions.

Elias: Yeah, they bring different angles to it, which is what makes the research interesting for us here.

Nadia: So what does this mean for the people driving or just walking around? It means that safety services relying on vehicle communication can get a bit more robust because it’s not just one piece of code being used.

Priya: From my side, I’m looking at those test results—they claim high accuracy, up to ninety-six point six percent—but I'm curious if that holds up when the weather gets bad or the lighting is really poor.

Elias: That's a fair question because those environmental factors are exactly where you want to break a security scheme. The paper does mention testing under various conditions, but we need to see how sensitive this authentication method is to those real-world variables.

Nadia: It seems the authors are focusing on creating a unified way to authenticate vehicles across different communication channels, which is pretty smart for building reliable ITS infrastructure.

Priya: The core idea here is that you use something you know—like a secret key sent over a secure digital channel—and then you add something you are by flashing a specific sequence of light signals.

Elias: Exactly, it’s the challenge response part where the infrastructure sends a randomized signal, and the vehicle has to flash back with encoded security data for them to decode.

Nadia: So what's the big picture takeaway? It suggests that for safety-critical systems like traffic management or emergency response, we should stop relying only on digital signals floating in the air.

Priya: I see it as adding a physical verification step, which is a necessary thing when you’re dealing with something that needs to be truly trustworthy.

Elias: It’s about making sure the infrastructure isn't just talking to *something* that claims to be a car, but actually verifying the physical presence and the correct response pattern.

Nadia: So, this paper sets up a new framework for V2I authentication that merges cryptographic security with observable physical behavior.

Priya: It opens up a discussion about how we should measure the real-world performance of these multi-factor systems outside of ideal laboratory settings.

Elias: Next up, we're going to look at how this specific scheme handles those tricky edge cases when it comes to the challenge response mechanism itself.

Massachusetts Institute of Technology (MIT) · Arizona State University (ASU) · University of Oxford

cs.CR

Submitted: 2025-05-01

Updated: 2025-05-08

Journal ref: IEEE Transactions on Vehicular Technology 75(2), 1779-1792 (2026)

DOI: 10.1109/TVT.2025.3598113

License: http://creativecommons.org/licenses/by-nc-nd/4.0/

Importance score: 98/100

The gist: Secure and reliable communications are crucial for Intelligent Transportation Systems (ITSs), where Vehicle-to-Infrastructure (V2I) communication plays a key role in enabling mobility-enhancing and…

Key concepts

V2I Communication
This refers to the communication link between a vehicle and roadside infrastructure, like traffic lights or sensors. It is vital for Intelligent Transportation Systems (ITS) because it enables safety-critical services and mobility enhancements.
Multi-Channel MFA
This security approach combines two different communication pathways—Non-Line-of-Sight (NLOS) wireless and Line-of-Sight (LOS) visual—to verify a vehicle's identity. This layered defense makes it much harder for attackers to successfully impersonate a vehicle.
Challenge–Response Security
This is an authentication method where the infrastructure sends a random question or 'challenge,' and the vehicle provides an answer. In this system, the challenge is issued by the infrastructure, and the response is encoded in a structured sequence flashed by the vehicle's headlights.
Deep Learning Decoding
Artificial intelligence models are used on the infrastructure side to analyze and interpret the visual data from a vehicle's headlights. These models decode security information embedded within the flashing light patterns to accurately authenticate whether the vehicle is legitimate.

Terminology

Summary

Secure and reliable communications are crucial for Intelligent Transportation Systems (ITSs), where Vehicle-to-Infrastructure (V2I) communication plays a key role in enabling mobility-enhancing and safety-critical services Current V2I authentication relies on credential-based methods over wireless Non-Line-of-Sight (NLOS) channels, leaving them exposed to remote impersonation and proximity attacks To mitigate these risks, we propose a unified Multi-Channel, Multi-Factor Authentication (MFA) scheme that combines NLOS cryptographic credentials with a Line-of-Sight (LOS) visual channel Our approach leverages a challenge–response security paradigm: the infrastructure issues “challenges” and the vehicle’s headlights respond by flashing a structured sequence containing encoded security data Deep learning models on the infrastructure side then decode the embedded information to authenticate the vehicle Real-world experimental evaluations demonstrate high test accuracy, reaching an average of 95% and 96. p. 0

Where the paper says it

Sarma, and Dajiang Suo Abstract—Secure and reliable communications are crucial for Intelligent Transportation Systems (ITSs), where Vehicle-toInfrastructure (V2I) communication plays a key role in enabling mobility-enhancing and safety-critical services. Current V2I authentication relies on credential-based methods over wireless Non-Line-of-Sight (NLOS) channels, leaving them exposed to remote impersonation and proximity attacks.

6%, respectively, under various lighting, weather, speed, and distance conditions

Scheme Design and Mechanism

The proposed scheme introduces an authentication process with a fully implemented LOS communication and response mechanism This scheme integrates multiple factors across NLOS and LOS channels as shown in Figure 2 and consists of the following three phases

  1. NLOS Phase: the vehicle uses the NLOS channel to transmit its unique security credential to a Registration Authority (RA) over a Transport Layer Security (TLS)-protected channel, as defined in IEEE 1609. p. 2

    Where the paper says it

    The scheme integrates multiple factors across NLOS and LOS channels as shown in Figure 2 and consists of the following three phases. • NLOS Phase: the vehicle uses the NLOS channel to transmit its unique security credential to a Registration Authority (RA) over a Transport Layer Security (TLS)- protected channel, as defined in IEEE 1609.

    2.1 This serves as the first authentication factor (something you know)

  2. LOS Phase: After verifying the first authentication factor, the infrastructure requests a second one (e. p. 2

    Where the paper says it

    For example, the exchange of security credentials can be carried out with the support of Public Key Infrastructure (PKI) [28, 29]. • LOS Phase: After verifying the first authentication factor, the infrastructure requests a second one (e.

    g., something you are or something you know) The RSU then sends the vehicle a randomized challenge, which the vehicle answers through the optical channel

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved authentication by combining NLOS cryptographic credentials with a Line-of-Sight visual channel, which ensures that the communicating entity is a real vehicle, as visual confirmation provides an additional layer of authentication based on physical presence.

  2. Enhanced security against remote impersonation and proximity attacks by implementing a challenge-response security paradigm where the infrastructure issues challenges and the vehicle responds by flashing headlights containing encoded security data.

  3. Robust decoding of flashing sequences using a dual-channel design that enables simultaneous decoding of the flashing sequence and extraction of vehicle spatial and locational features for robust authentication.

  4. Optimized deep learning model performance by employing a SlowFast network architecture, which is specifically designed to concurrently analyze slow (spatial) and fast (temporal) streams of information for superior pattern recognition.

  5. Increased real-time processing efficiency by utilizing a two-stage pipeline that first uses an object detector with MultiObject Tracking to detect and identify the ROI sequences for each flashing vehicle, and then applies a video classification network.

Abstract

Secure and reliable communications are crucial for Intelligent Transportation Systems (ITSs), where Vehicle-to-Infrastructure (V2I) communication plays a key role in enabling mobility-enhancing and safety-critical services. Current V2I authentication relies on credential-based methods over wireless Non-Line-of-Sight (NLOS) channels, leaving them exposed to remote impersonation and proximity attacks. To mitigate these risks, we propose a unified Multi-Channel, Multi-Factor Authentication (MFA) scheme that combines NLOS cryptographic credentials with a Line-of-Sight (LOS) visual channel. Our approach leverages a challenge-response security paradigm: the infrastructure issues challenges and the vehicle's headlights respond by flashing a structured sequence containing encoded security data. Deep learning models on the infrastructure side then decode the embedded information to authenticate the vehicle. Real-world experimental evaluations demonstrate high test accuracy, reaching an average of 95% and 96.6%, respectively, under various lighting, weather, speed, and distance conditions. Additionally, we conducted extensive experiments on three state-of-the-art deep learning models, including detailed ablation studies for decoding the flashing sequence. Our results indicate that the optimal architecture employs a dual-channel design, enabling simultaneous decoding of the flashing sequence and extraction of vehicle spatial and locational features for robust authentication.

Sources

Related papers