Panos Papadimitratos

Seba Anna Varghese, Alireza Dehlaghi Ghadim, Ali Balador et al.

Digital twins have recently gained significant interest in simulation, optimization, and predictive maintenance of Industrial Control Systems (ICS). Recent studies discuss the possibility of using digital twins for intrusion detection in industrial systems. Accordingly, this study contributes to a digital twin-based security framework for industrial control systems, extending its capabilities for simulation of attacks and defense mechanisms. Four types of process-aware attack scenarios are implemented on a standalone open-source digital twin of an industrial filling plant: command injection, network Denial of Service (DoS), calculated measurement modification, and naive measurement modification. A stacked ensemble classifier is proposed as the real-time intrusion detection, based on the offline evaluation of eight supervised machine learning algorithms. The designed stacked model outperforms previous methods in terms of F1-Score and accuracy, by combining the predictions of various algorithms, while it can detect and classify intrusions in near real-time (0.1 seconds). This study also discusses the practicality and benefits of the proposed digital twin-based security framework.

Moritz Wiese, Karl Henrik Johansson, Tobias J. Oechtering et al.

Uncertain wiretap channels are introduced. Their zero-error secrecy capacity is defined. If the sensor-estimator channel is perfect, it is also calculated. Further properties are discussed. The problem of estimating a dynamical system with nonstochastic disturbances is studied where the sensor is connected to the estimator and an eavesdropper via an uncertain wiretap channel. The estimator should obtain a uniformly bounded estimation error whereas the eavesdropper's error should tend to infinity. It is proved that the system can be estimated securely if the zero-error capacity of the sensor-estimator channel is strictly larger than the logarithm of the system's unstable pole and the zero-error secrecy capacity of the uncertain wiretap channel is positive.

Rodothea Myrsini Tsoupidi, Roberto Castañeda Lozano, Elena Troubitsyna et al.

Side-channel attacks impose a serious threat to cryptographic algorithms, including widely employed ones, such as AES and RSA. These attacks take advantage of the algorithm implementation in hardware or software to extract secret information via side channels. Software masking is a mitigation approach against power side-channel attacks aiming at hiding the secret-revealing dependencies from the power footprint of a vulnerable implementation. However, this type of software mitigation often depends on general-purpose compilers, which do not preserve non-functional properties. Moreover, microarchitectural features, such as the memory bus and register reuse, may also leak secret information. These abstractions are not visible at the high-level implementation of the program. Instead, they are decided at compile time. To remedy these problems, security engineers often sacrifice code efficiency by turning off compiler optimization and/or performing local, post-compilation transformations. This paper proposes Secure by Construction Code Generation (SecCG), a constraint-based compiler approach that generates optimized yet secure against power side channels code. SecCG controls the quality of the mitigated program by efficiently searching the best possible low-level implementation according to a processor cost model. In our experiments with twelve masked cryptographic functions up to 100 lines of code on Mips32 and ARM Thumb, SecCG speeds up the generated code from 75% to 8 times compared to non-optimized secure code with an overhead of up to 7% compared to non-secure optimized code at the expense of a high compilation cost. In summary, this paper proposes a formal model to generate power side channel free low-level code.

Musa Furkan Keskin, Muralikrishnan Srinivasan, Onur Gunlu et al.

Integrated sensing and communication (ISAC) will be central to 6G-enabled transportation, providing both seamless connectivity and high-precision sensing. However, this tight integration exposes attack points not encountered in pure sensing and communication systems. In this article, we identify unique ISAC-induced security challenges and opportunities in three interrelated domains: cyber-physical (where manipulation of sensors and actuators can mislead perception and control), physical-layer (where over-the-air signals are vulnerable to spoofing and jamming) and protocol (where complex cryptographic protocols cannot detect lower-layer attacks). Building on these insights, we put forward a multi-domain security vision for 6G transportation and propose an integrated security framework that unifies protection across domains by leveraging existing ISAC measurements for lightweight cross-checks.

Saeif Al-Hazbi, Ahmed Hussain, Savio Sciancalepore et al.

Radio Frequency Fingerprinting (RFF) techniques promise to authenticate wireless devices at the physical layer based on inherent hardware imperfections introduced during manufacturing. Such RF transmitter imperfections are reflected into over-the-air signals, allowing receivers to accurately identify the RF transmitting source. Recent advances in Machine Learning, particularly in Deep Learning (DL), have improved the ability of RFF systems to extract and learn complex features that make up the device-specific fingerprint. However, integrating DL techniques with RFF and operating the system in real-world scenarios presents numerous challenges, originating from the embedded systems and the DL research domains. This paper systematically identifies and analyzes the essential considerations and challenges encountered in the creation of DL-based RFF systems across their typical development life-cycle, which include (i) data collection and preprocessing, (ii) training, and finally, (iii) deployment. Our investigation provides a comprehensive overview of the current open problems that prevent real deployment of DL-based RFF systems while also discussing promising research opportunities to enhance the overall accuracy, robustness, and privacy of these systems.

Sheng Liu, Panos Papadimitratos

FL has emerged as a transformative paradigm for ITS, notably camera-based Road Condition Classification (RCC). However, by enabling collaboration, FL-based RCC exposes the system to adversarial participants launching Targeted Label-Flipping Attacks (TLFAs). Malicious clients (vehicles) can relabel their local training data (e.g., from an actual uneven road to a wrong smooth road), consequently compromising global model predictions and jeopardizing transportation safety. Existing countermeasures against such poisoning attacks fail to maintain resilient model performance near the necessary attack-free levels in various attack scenarios due to: 1) not tailoring poisoned local model detection to TLFAs, 2) not excluding malicious vehicular clients based on historical behavior, and 3) not remedying the already-corrupted global model after exclusion. To close this research gap, we propose FedTrident, which introduces: 1) neuron-wise analysis for local model misbehavior detection (notably including attack goal identification, critical feature extraction, and GMM-based model clustering and filtering); 2) adaptive client rating for client exclusion according to the local model detection results in each FL round; and 3) machine unlearning for corrupted global model remediation once malicious clients are excluded during FL. Extensive evaluation across diverse FL-RCC models, tasks, and configurations demonstrates that FedTrident can effectively thwart TLFAs, achieving performance comparable to that in attack-free scenarios and outperforming eight baseline countermeasures by 9.49% and 4.47% for the two most critical metrics. Moreover, FedTrident is resilient to various malicious client rates, data heterogeneity levels, complicated multi-task, and dynamic attacks.

Saeif Alhazbi, Ahmed Mohamed Hussain, Gabriele Oligeri et al.

As Large Language Models (LLMs) become increasingly integrated into many technological ecosystems across various domains and industries, identifying which model is deployed or being interacted with is critical for the security and trustworthiness of the systems. Current verification methods typically rely on analyzing the generated output to determine the source model. However, these techniques are susceptible to adversarial attacks, operate in a post-hoc manner, and may require access to the model weights to inject a verifiable fingerprint. In this paper, we propose a novel passive and non-invasive fingerprinting technique that operates in real-time and remains effective even under encrypted network traffic conditions. Our method leverages the intrinsic autoregressive generation nature of language models, which generate text one token at a time based on all previously generated tokens, creating a unique temporal pattern like a rhythm or heartbeat that persists even when the output is streamed over a network. We find that measuring the Inter-Token Times (ITTs)-time intervals between consecutive tokens-can identify different language models with high accuracy. We develop a Deep Learning (DL) pipeline to capture these timing patterns using network traffic analysis and evaluate it on 16 Small Language Models (SLMs) and 10 proprietary LLMs across different deployment scenarios, including local host machine (GPU/CPU), Local Area Network (LAN), Remote Network, and Virtual Private Network (VPN). The experimental results confirm that our proposed technique is effective and maintains high accuracy even when tested in different network conditions. This work opens a new avenue for model identification in real-world scenarios and contributes to more secure and trustworthy language model deployment.

Johan Wahréus, Ahmed Mohamed Hussain, Panos Papadimitratos

Numerous studies have investigated methods for jailbreaking Large Language Models (LLMs) to generate harmful content. Typically, these methods are evaluated using datasets of malicious prompts designed to bypass security policies established by LLM providers. However, the generally broad scope and open-ended nature of existing datasets can complicate the assessment of jailbreaking effectiveness, particularly in specific domains, notably cybersecurity. To address this issue, we present and publicly release CySecBench, a comprehensive dataset containing 12662 prompts specifically designed to evaluate jailbreaking techniques in the cybersecurity domain. The dataset is organized into 10 distinct attack-type categories, featuring close-ended prompts to enable a more consistent and accurate assessment of jailbreaking attempts. Furthermore, we detail our methodology for dataset generation and filtration, which can be adapted to create similar datasets in other domains. To demonstrate the utility of CySecBench, we propose and evaluate a jailbreaking approach based on prompt obfuscation. Our experimental results show that this method successfully elicits harmful content from commercial black-box LLMs, achieving Success Rates (SRs) of 65% with ChatGPT and 88% with Gemini; in contrast, Claude demonstrated greater resilience with a jailbreaking SR of 17%. Compared to existing benchmark approaches, our method shows superior performance, highlighting the value of domain-specific evaluation datasets for assessing LLM security measures. Moreover, when evaluated using prompts from a widely used dataset (i.e., AdvBench), it achieved an SR of 78.5%, higher than the state-of-the-art methods.

Ahmed Mohamed Hussain, Nada Abughanam, Panos Papadimitratos

The deployment of the Internet of Things (IoT) in smart cities and critical infrastructure has enhanced connectivity and real-time data exchange but introduced significant security challenges. While effective, cryptography can often be resource-intensive for small-footprint resource-constrained (i.e., IoT) devices. Radio Frequency Fingerprinting (RFF) offers a promising authentication alternative by using unique RF signal characteristics for device identification at the Physical (PHY)-layer, without resorting to cryptographic solutions. The challenge is two-fold: how to deploy such RFF in a large scale and for resource-constrained environments. Edge computing, processing data closer to its source, i.e., the wireless device, enables faster decision-making, reducing reliance on centralized cloud servers. Considering a modest edge device, we introduce two truly lightweight Edge AI-based RFF schemes tailored for resource-constrained devices. We implement two Deep Learning models, namely a Convolution Neural Network and a Transformer-Encoder, to extract complex features from the IQ samples, forming device-specific RF fingerprints. We convert the models to TensorFlow Lite and evaluate them on a Raspberry Pi, demonstrating the practicality of Edge deployment. Evaluations demonstrate the Transformer-Encoder outperforms the CNN in identifying unique transmitter features, achieving high accuracy (> 0.95) and ROC-AUC scores (> 0.90) while maintaining a compact model size of 73KB, appropriate for resource-constrained devices.

Salahuddin Salahuddin, Ahmed Hussain, Jussi Löppönen et al.

While Large Language Models (LLMs) demonstrate exceptional natural language capabilities, general-purpose models lack specialized domain knowledge for effective cybersecurity analysis. In this work, we investigate Domain-Adaptive Continuous Pretraining (DAP) as a methodology for enhancing cybersecurity understanding in pretrained LLMs while preserving general language capabilities. We systematically adapted three decoder-based architectures -- Llama-3.1-8B, DeepSeek-R1-Distill-Qwen-14B, and Llama-3.3-70B-Instruct -- using a curated 126-million-word cybersecurity corpus from standards, academic literature, and various other sources. Our approach employed constrained training parameters and distributed FSDP training to balance domain specialization with knowledge preservation. Evaluation across three cybersecurity benchmarks, namely, CTI-MCQ, CyberMetric, and SecEval, demonstrates consistent improvements post-adaptation. The Llama-3.3-70B-Ins-DAP model achieved state-of-the-art accuracies of 0.718, 0.933, and 0.864, respectively, outperforming specialized models, including Llama-Primus-Base. Notably, competitive performance was achieved using substantially smaller datasets (118.8 million versus 2.77 billion tokens), demonstrating efficient domain specialization viability. We establish that targeted continuous pretraining enables effective cybersecurity domain adaptation with computational feasibility, providing foundations for specialized AI assistants in threat analysis, vulnerability assessment, and security documentation while challenging prevailing assumptions about data requirements for LLM specialization.

Johan Wahréus, Ahmed Hussain, Panos Papadimitratos

Large Language Models (LLMs) have transformed task automation and content generation across various domains while incorporating safety filters to prevent misuse. We introduce a novel jailbreaking framework that employs distributed prompt processing combined with iterative refinements to bypass these safety measures, particularly in generating malicious code. Our architecture consists of four key modules: prompt segmentation, parallel processing, response aggregation, and LLM-based jury evaluation. Tested on 500 malicious prompts across 10 cybersecurity categories, the framework achieves a 73.2% Success Rate (SR) in generating malicious code. Notably, our comparative analysis reveals that traditional single-LLM judge evaluation overestimates SRs (93.8%) compared to our LLM jury system (73.2%), with manual verification confirming that single-judge assessments often accept incomplete implementations. Moreover, we demonstrate that our distributed architecture improves SRs by 12% over the non-distributed approach in an ablation study, highlighting both the effectiveness of distributed prompt processing and the importance of robust evaluation methodologies in assessing jailbreak attempts.

Hexu Li, Konstantinos Kalogiannis, Ahmed Mohamed Hussain et al.

Vehicle platooning, with vehicles traveling in close formation coordinated through Vehicle-to-Everything (V2X) communications, offers significant benefits in fuel efficiency and road utilization. However, it is vulnerable to sophisticated falsification attacks by authenticated insiders that can destabilize the formation and potentially cause catastrophic collisions. This paper addresses this challenge: misbehavior detection in vehicle platooning systems. We present AttentionGuard, a transformer-based framework for misbehavior detection that leverages the self-attention mechanism to identify anomalous patterns in mobility data. Our proposal employs a multi-head transformer-encoder to process sequential kinematic information, enabling effective differentiation between normal mobility patterns and falsification attacks across diverse platooning scenarios, including steady-state (no-maneuver) operation, join, and exit maneuvers. Our evaluation uses an extensive simulation dataset featuring various attack vectors (constant, gradual, and combined falsifications) and operational parameters (controller types, vehicle speeds, and attacker positions). Experimental results demonstrate that AttentionGuard achieves up to 0.95 F1-score in attack detection, with robust performance maintained during complex maneuvers. Notably, our system performs effectively with minimal latency (100ms decision intervals), making it suitable for real-time transportation safety applications. Comparative analysis reveals superior detection capabilities and establishes the transformer-encoder as a promising approach for securing Cooperative Intelligent Transport Systems (C-ITS) against sophisticated insider threats.

Johan Wahréus, Ahmed Hussain, Panos Papadimitratos

Large Language Models (LLMs) are increasingly deployed for task automation and content generation, yet their safety mechanisms remain vulnerable to circumvention through different jailbreaking techniques. In this paper, we introduce \textit{Content Concretization} (CC), a novel jailbreaking technique that iteratively transforms abstract malicious requests into concrete, executable implementations. CC is a two-stage process: first, generating initial LLM responses using lower-tier, less constrained safety filters models, then refining them through higher-tier models that process both the preliminary output and original prompt. We evaluate our technique using 350 cybersecurity-specific prompts, demonstrating substantial improvements in jailbreak Success Rates (SRs), increasing from 7\% (no refinements) to 62\% after three refinement iterations, while maintaining a cost of 7.5\textcent~per prompt. Comparative A/B testing across nine different LLM evaluators confirms that outputs from additional refinement steps are consistently rated as more malicious and technically superior. Moreover, manual code analysis reveals that generated outputs execute with minimal modification, although optimal deployment typically requires target-specific fine-tuning. With eventual improved harmful code generation, these results highlight critical vulnerabilities in current LLM safety frameworks.

Sheng Liu, Panos Papadimitratos

Federated Learning (FL) has emerged as a promising solution for privacy-preserving autonomous driving, specifically camera-based Road Condition Classification (RCC) systems, harnessing distributed sensing, computing, and communication resources on board vehicles without sharing sensitive image data. However, the collaborative nature of FL-RCC frameworks introduces new vulnerabilities: Targeted Label Flipping Attacks (TLFAs), in which malicious clients (vehicles) deliberately alter their training data labels to compromise the learned model inference performance. Such attacks can, e.g., cause a vehicle to mis-classify slippery, dangerous road conditions as pristine and exceed recommended speed. However, TLFAs for FL-based RCC systems are largely missing. We address this challenge with a threefold contribution: 1) we disclose the vulnerability of existing FL-RCC systems to TLFAs; 2) we introduce a novel label-distance-based metric to precisely quantify the safety risks posed by TLFAs; and 3) we propose FLARE, a defensive mechanism leveraging neuron-wise analysis of the output layer to mitigate TLFA effects. Extensive experiments across three RCC tasks, four evaluation metrics, six baselines, and three deep learning models demonstrate both the severity of TLFAs on FL-RCC systems and the effectiveness of FLARE in mitigating the attack impact.

M. Spanghero, P. Papadimitratos

A wide gamut of important applications rely on global navigation satellite systems (GNSS) for precise time and positioning. Attackers dictating the GNSS receiver position and time solution are a significant risk, especially due to the inherent vulnerability of GNSS systems. A first line of defense, for a large number of receivers, is to rely on additional information obtained through the rich connectivity of GNSS enabled platforms. Network time can be used for direct validation of the GNSS receiver time; but this depends on network availability. To allow attack detection even when there are prolonged network disconnections, we present a method based on on-board ensemble of reference clocks. This allows the receiver to detect sophisticated attacks affecting the GNSS time solution, independently of the specific attack methodology. Results obtained with Chip-Scale Oven Compensated Oscillators (CS-OCXO) are promising and demonstrate the potential of embedded ensembles of reference clocks, detecting attacks causing modifications of the receiver time offset as low as 0.3us, with half the detection latency compared to related literature.

M. Lenhart, M. Spanghero, P. Papadimitratos

With the introduction of Navigation Message Authentication (NMA), future Global Navigation Satellite Systems (GNSSs) prevent spoofing by simulation, i.e., the generation of forged satellite signals based on public information. However, authentication does not prevent record-and-replay attacks, commonly termed as meaconing. These attacks are less powerful in terms of adversarial control over the victim receiver location and time, but by acting at the signal level, they are not thwarted by NMA. This makes replaying/relaying attacks a significant threat for GNSS. While there are numerous investigations on meaconing, the majority does not rely on actual implementation and experimental evaluation in real-world settings. In this work, we contribute to the improvement of the experimental understanding of meaconing attacks. We design and implement a system capable of real-time, distributed, and mobile meaconing, built with off-the-shelf hardware. We extend from basic distributed attacks, with signals from different locations relayed over the Internet and replayed within range of the victim receiver(s): this has high bandwidth requirements and thus depends on the quality of service of the available network to work. To overcome this limitation, we propose to replay on message level, including the authentication part of the payload. The resultant reduced bandwidth enables the attacker to operate in mobile scenarios, as well as to replay signals from multiple GNSS constellations and/or bands simultaneously. Additionally, the attacker can delay individually selected satellite signals to potentially influence the victim position and time solution in a more fine-grained manner. Our versatile test-bench, enabling different types of replaying/relaying attacks, facilitates testing realistic scenarios towards new and improved replaying/relaying-focused countermeasures in GNSS receivers.

M. Spanghero, P. Papadimitratos

To mitigate spoofing attacks targeting global navigation satellite systems (GNSS) receivers, one promising method is to rely on alternative time sources, such as network-based synchronization, in order to detect clock offset discrepancies caused by GNSS attacks. However, in case of no network connectivity, such validation references would not be available. A viable option is to rely on a local time reference; in particular, precision hardware clock ensembles of chip-scale thermally stable oscillators with extended holdover capabilities. We present a preliminary design and results towards a custom device capable of providing a stable reference, with smaller footprint and cost compared to traditional precision clocks. The system is fully compatible with existing receiver architecture, making this solution feasible for most industrial scenarios. Further integration with network-based synchronization can provide a complete time assurance system, with high short- and long-term stability.

M. Lenhart, M. Spanghero, P. Papadimitratos

Global Navigation Satellite Systems (GNSS) are ubiquitously relied upon for positioning and timing. Detection and prevention of attacks against GNSS have been researched over the last decades, but many of these attacks and countermeasures were evaluated based on simulation. This work contributes to the experimental investigation of GNSS vulnerabilities, implementing a relay/replay attack with off-the-shelf hardware. Operating at the signal level, this attack type is not hindered by cryptographically protected transmissions, such as Galileo's Open Signals Navigation Message Authentication (OS-NMA). The attack we investigate involves two colluding adversaries, relaying signals over large distances, to effectively spoof a GNSS receiver. We demonstrate the attack using off-the-shelf hardware, we investigate the requirements for such successful colluding attacks, and how they can be enhanced, e.g., allowing for finer adversarial control over the victim receiver.

K. Zhang, M. Spanghero, P. Papadimitratos

Global navigation satellite systems (GNSS) provide pervasive accurate positioning and timing services for a large gamut of applications, from Time based One-Time Passwords (TOPT), to power grid and cellular systems. However, there can be security concerns for the applications due to the vulnerability of GNSS. It is important to observe that GNSS receivers are components of platforms, in principle having rich connectivity to different network infrastructures. Of particular interest is the access to a variety of timing sources, as those can be used to validate GNSS-provided location and time. Therefore, we consider off-the-shelf platforms and how to detect if the GNSS receiver is attacked or not, by cross-checking the GNSS time and time from other available sources. First, we survey different technologies to analyze their availability, accuracy, and trustworthiness for time synchronization. Then, we propose a validation approach for absolute and relative time. Moreover, we design a framework and experimental setup for the evaluation of the results. Attacks can be detected based on WiFi supplied time when the adversary shifts the GNSS provided time, more than 23.942us; with Network Time Protocol (NTP) supplied time when the adversary-induced shift is more than 2.046ms. Consequently, the proposal significantly limits the capability of an adversary to manipulate the victim GNSS receiver.

M. Spanghero, K. Zhang, P. Papadimitratos

Information cross-validation can be a powerful tool to detect manipulated, dubious GNSS data. A promising approach is to leverage time obtained over networks a mobile device can connect to, and detect discrepancies between the GNSS-provided time and the network time. The challenge lies in having reliably both accurate and trustworthy network time as the basis for the GNSS attack detection. Here, we provide a concrete proposal that leverages, together with the network time servers, the nearly ubiquitous IEEE 802.11 (Wi-Fi) infrastructure. Our framework supports application-layer, secure and robust real time broadcasting by Wi-Fi Access Points (APs), based on hash chains and infrequent digital signatures verification to minimize computational and communication overhead, allowing mobile nodes to efficiently obtain authenticated and rich time information as they roam. We pair this method with Network Time Security (NTS), for enhanced resilience through multiple sources, available, ideally, simultaneously. We analyze the performance of our scheme in a dedicated setup, gauging the overhead for authenticated time data (Wi-Fi timestamped beacons and NTS). The results show that it is possible to provide security for the external to GNSS time sources, with minimal overhead for authentication and integrity, even when the GNSS-equipped nodes are mobile, and thus have short interactions with the Wi-Fi infrastructure and possibly intermittent Internet connectivity, as well as limited resources.

Marziyeh Soltani, Mahtab Mirmohseni, Panos Papadimitratos

Non-orthogonal multiple access (NOMA) and massive multiple-input multiple-output (MIMO) systems are highly efficient. Massive MIMO systems are inherently resistant to passive attackers (eavesdroppers), thanks to transmissions directed to the desired users. However, active attackers can transmit a combination of legitimate user pilot signals during the channel estimation phase. This way they can mislead the base station (BS) to rotate the transmission in their direction, and allow them to eavesdrop during the downlink data transmission phase. In this paper, we analyse this vulnerability in an improved system model and stronger adversary assumptions, and investigate how physical layer security can mitigate such attacks and ensure secure (confidential) communication. We derive the secrecy outage probability (SOP) and a lower bound on the ergodic secrecy capacity, using stochastic geometry tools when the number of antennas in the BSs tends to infinity. We adapt the result to evaluate the secrecy performance in massive orthogonal multiple access (OMA). We find that appropriate power allocation allows NOMA to outperform OMA in terms of ergodic secrecy rate and SOP.

Parisa Babaheidarian, Somayeh Salimi, Panos Papadimitratos

In this paper, we present an achievable security scheme for an interference channel with arbitrary number of users. In this model, each receiver should be able to decode its intended message while it cannot decode any meaningful information regarding messages intended for other receivers. Our scheme achieves individual secure rates which scale linearly with log(SNR) and achieves sum secure rates which is within constant gap of sum secure capacity. To design the encoders at the transmitters side, we combine nested lattice coding, random i.i.d. codes, and cooperative jamming techniques. Asymmetric compute-and-forward framework is used to perform the decoding operation at the receivers. The novelty of our scheme is that it is the first asymptotically optimal achievable scheme for this security scenario which scales to arbitrary number of users and works for any finite-valued SNR. Also, our scheme achieves the upper bound sum secure degrees of freedom of $1$ without using external helpers.

Hongyu Jin, Mohammad Khodaei, Panos Papadimitratos

We surveyed and presented the state-of-the-art VC systems, security and privacy architectures and technologies, emphasizing on security and privacy challenges and their solutions for P2P interactions in VSNs towards standardization and deployment. We note that beyond safety applications that have drawn a lot of attention in VC systems, there is significant and rising interest in vehicle-to-vehicle interaction for a range of transportation efficiency and infotainment applications, notably LBS as well as a gamut of services by mobile providers. While this enriches the VC systems and the user experience, security and privacy concerns are also intensified. This is especially so, considering (i) the privacy risk from the exposure of the users to the service providers, and (ii) the security risk from the interaction with malicious or selfish and thus misbehaving users or infrastructure. We showed existing solutions can in fact evolve and address the VSN-specific challenges, and improve or even accelerate the adoption of VSN applications.

Hongyu Jin, Panos Papadimitratos

Location-based Services (LBSs) provide valuable services, with convenient features for users. However, the information disclosed through each request harms user privacy. This is a concern particularly with honest-but-curious LBS servers, which could, by collecting requests, track users and infer additional sensitive user data. This is the motivation of both centralized and decentralized location privacy protection schemes for LBSs: anonymizing and obfuscating LBS queries to not disclose exact information, while still getting useful responses. Decentralized schemes overcome the disadvantages of centralized schemes, eliminating anonymizers and enhancing users' control over sensitive information. However, an insecure decentralized system could pose even more serious security threats than privacy leakage. We address exactly this problem, by proposing security enhancements for mobile data sharing systems. We protect user privacy while preserving accountability of user activities, leveraging pseudonymous authentication with mainstream cryptography. Our design leverages architectures proposed for large scale mobile systems, while it incurs minimal changes to LBS servers as it can be deployed in parallel to the LBS servers. This further motivates the adoption of our design, in order to cater to the needs of privacy-sensitive users. We provide an analysis of security and privacy concerns and countermeasures, as well as a performance evaluation of basic protocol operations showing the practicality of our design.

Hongyu Jin, Panos Papadimitratos

Security and privacy in Vehicular Ad-hoc Networks (VANETs) mandates use of short-lived credentials (pseudonyms) and cryptographic key pairs. This implies significant computational overhead for vehicles, needing to validate often numerous such pseudonyms within a short period. To alleviate such a bottleneck that could even place vehicle safety at risk, we propose a proactive pseudonym validation approach based on Bloom Filters (BFs). We show that our scheme could liberate computational resources for other (safety- and time-critical) operations with reasonable communication overhead without compromising security and privacy.

Nikolaos Alexiou, Marcello Laganà, Stylianos Gisdakis et al.

Standardization and harmonization efforts have reached a consensus towards using a special-purpose Vehicular Public-Key Infrastructure (VPKI) in upcoming Vehicular Communication (VC) systems. However, there are still several technical challenges with no conclusive answers; one such an important yet open challenge is the acquisition of short-term credentials, pseudonym: how should each vehicle interact with the VPKI, e.g., how frequently and for how long? Should each vehicle itself determine the pseudonym lifetime? Answering these questions is far from trivial. Each choice can affect both the user privacy and the system performance and possibly, as a result, its security. In this paper, we make a novel systematic effort to address this multifaceted question. We craft three generally applicable policies and experimentally evaluate the VPKI system performance, leveraging two large-scale mobility datasets. We consider the most promising, in terms of efficiency, pseudonym acquisition policies; we find that within this class of policies, the most promising policy in terms of privacy protection can be supported with moderate overhead. Moreover, in all cases, this work is the first to provide tangible evidence that the state-of-the-art VPKI can serve sizable areas or domain with modest computing resources.

Hongyu Jin, Panos Papadimitratos

Authenticated safety beacons in Vehicular Communication (VC) systems ensure awareness among neighboring vehicles. However, the verification of beacon signatures introduces significant processing overhead for resource-constrained vehicular On-Board Units (OBUs). Even worse in dense neighborhood or when a clogging Denial of Service (DoS) attack is mounted. The OBU would fail to verify for all received (authentic or fictitious) beacons. This could significantly delay the verifications of authentic beacons or even affect the awareness of neighboring vehicle status. In this paper, we propose an efficient cooperative beacon verification scheme leveraging efficient symmetric key based authentication on top of pseudonymous authentication (based on traditional public key cryptography), providing efficient discovery of authentic beacons among a pool of received authentic and fictitious beacons, and can significantly decrease waiting times of beacons in queue before their validations. We show with simulation results that our scheme can guarantee low waiting times for received beacons even in high neighbor density situations and under DoS attacks, under which a traditional scheme would not be workable.

Hongyu Jin, Panos Papadimitratos

VANET security introduces significant processing overhead for resource-constrained On-Board Units (OBUs). Here, we propose a novel scheme that allows secure Vehicular Communication (VC) systems to scale well beyond network densities for which existing optimization approaches could be workable, without compromising security (and privacy).

Mohammad Khodaei, Panos Papadimitratos

In spite of progress in securing Vehicular Communication (VC) systems, there is no consensus on how to distribute Certificate Revocation Lists (CRLs). The main challenges lie exactly in (i) crafting an efficient and timely distribution of CRLs for numerous anonymous credentials, pseudonyms, (ii) maintaining strong privacy for vehicles prior to revocation events, even with honest-but-curious system entities, (iii) and catering to computation and communication constraints of on-board units with intermittent connectivity to the infrastructure. Relying on peers to distribute the CRLs is a double-edged sword: abusive peers could "pollute" the process, thus degrading the timely CRLs distribution. In this paper, we propose a vehicle-centric solution that addresses all these challenges and thus closes a gap in the literature. Our scheme radically reduces CRL distribution overhead: each vehicle receives CRLs corresponding only to its region of operation and its actual trip duration. Moreover, a "fingerprint" of CRL 'pieces' is attached to a subset of (verifiable) pseudonyms for fast CRL 'piece' validation (while mitigating resource depletion attacks abusing the CRL distribution). Our experimental evaluation shows that our scheme is efficient, scalable, dependable, and practical: with no more than 25 KB/s of traffic load, the latest CRL can be delivered to 95% of the vehicles in a region (50x50 KM) within 15s, i.e., more than 40 times faster than the state-of-the-art. Overall, our scheme is a comprehensive solution that complements standards and can catalyze the deployment of secure and privacy-protecting VC systems.

Hamid Noroozi, Mohammad Khodaei, Panos Papadimitratos

The central building block of secure and privacy-preserving Vehicular Communication (VC) systems is a Vehicular Public-Key Infrastructure (VPKI), which provides vehicles with multiple anonymized credentials, termed pseudonyms. These pseudonyms are used to ensure message authenticity and integrity while preserving vehicle (and thus passenger) privacy. In the light of emerging large-scale multi-domain VC environments, the efficiency of the VPKI and, more broadly, its scalability are paramount. In this extended abstract, we leverage the state-of-the-art VPKI system and enhance its functionality towards a highly-available and dynamically-scalable design, this ensures that the system remains operational in the presence of benign failures or any resource depletion attack, and that it dynamically scales out, or possibly scales in, according to the requests' arrival rate. Our full-blown implementation on the Google Cloud Platform shows that deploying a VPKI for a large-scale scenario can be cost-effective, while efficiently issuing pseudonyms for the requesters.

Mohammad Khodaei, Andreas Messing, Panos Papadimitratos

Any on-demand pseudonym acquisition strategy is problematic should the connectivity to the credential management infrastructure be intermittent. If a vehicle runs out of pseudonyms with no connectivity to refill its pseudonym pool, one solution is the on-the-fly generation of pseudonyms, e.g., leveraging anonymous authentication. However, such a vehicle would stand out in the crowd: one can simply distinguish pseudonyms, thus signed messages, based on the pseudonym issuer signature, link them and track the vehicle. To address this challenge, we propose a randomized hybrid scheme, RHyTHM, to enable vehicles to remain operational when disconnected without compromising privacy: vehicles with valid pseudonyms help others to enhance their privacy by randomly joining them in using on-the-fly self-certified pseudonyms along with aligned lifetimes. This way, the privacy of disconnected users is enhanced with a reasonable computational overhead.

Satyam Dwivedi, John Olof Nilsson, Panos Papadimitratos et al.

A novel method and protocol establishing common secrecy based on physical parameters between two users is proposed. The four physical parameters of users are their clock frequencies, their relative clock phases and the distance between them. The protocol proposed between two users is backed by theoretical model for the measurements. Further, estimators are proposed to estimate secret physical parameters. Physically exchanged parameters are shown to be secure by virtue of their non-observability to adversaries. Under a simplified analysis based on a testbed settings, it is shown that 38 bits of common secrecy can be derived for one run of the proposed protocol among users. The method proposed is also robust against various kinds of active timing attacks and active impersonating adversaries.

Mohammad Khodaei, Hongyu Jin, Panos Papadimitratos

Several years of academic and industrial research efforts have converged to a common understanding on fundamental security building blocks for the upcoming Vehicular Communication (VC) systems. There is a growing consensus towards deploying a special-purpose identity and credential management infrastructure, i.e., a Vehicular Public-Key Infrastructure (VPKI), enabling pseudonymous authentication, with standardization efforts towards that direction. In spite of the progress made by standardization bodies (IEEE 1609.2 and ETSI) and harmonization efforts (Car2Car Communication Consortium (C2C-CC)), significant questions remain unanswered towards deploying a VPKI. Deep understanding of the VPKI, a central building block of secure and privacy-preserving VC systems, is still lacking. This paper contributes to the closing of this gap. We present SECMACE, a VPKI system, which is compatible with the IEEE 1609.2 and ETSI standards specifications. We provide a detailed description of our state-of-the-art VPKI that improves upon existing proposals in terms of security and privacy protection, and efficiency. SECMACE facilitates multi-domain operations in the VC systems and enhances user privacy, notably preventing linking pseudonyms based on timing information and offering increased protection even against honest-but-curious VPKI entities. We propose multiple policies for the vehicle-VPKI interactions, based on which and two large-scale mobility trace datasets, we evaluate the full-blown implementation of SECMACE. With very little attention on the VPKI performance thus far, our results reveal that modest computing resources can support a large area of vehicles with very low delays and the most promising policy in terms of privacy protection can be supported with moderate overhead.

Moritz Wiese, Tobias J. Oechtering, Karl Henrik Johansson et al.

We study the problem of securely estimating the states of an unstable dynamical system subject to nonstochastic disturbances. The estimator obtains all its information through an uncertain channel which is subject to nonstochastic disturbances as well, and an eavesdropper obtains a disturbed version of the channel inputs through a second uncertain channel. An encoder observes and block-encodes the states in such a way that, upon sending the generated codeword, the estimator's error is bounded and such that a security criterion is satisfied ensuring that the eavesdropper obtains as little state information as possible. Two security criteria are considered and discussed with the help of a numerical example. A sufficient condition on the uncertain wiretap channel, i.e., the pair formed by the uncertain channel from encoder to estimator and the uncertain channel from encoder to eavesdropper, is derived which ensures that a bounded estimation error and security are achieved. This condition is also shown to be necessary for a subclass of uncertain wiretap channels. To formulate the condition, the zero-error secrecy capacity of uncertain wiretap channels is introduced, i.e., the maximal rate at which data can be transmitted from the encoder to the estimator in such a way that the eavesdropper is unable to reconstruct the transmitted data. Lastly, the zero-error secrecy capacity of uncertain wiretap channels is studied.

Mohammad Khodaei, Panos Papadimitratos

Standardization and harmonization efforts have reached a consensus towards using a special-purpose Vehicular Public-Key Infrastructure (VPKI) in upcoming Vehicular Communication (VC) systems. However, there are still several technical challenges with no conclusive answers; one such an important yet open challenge is the acquisition of shortterm credentials, pseudonym: how should each vehicle interact with the VPKI, e.g., how frequently and for how long? Should each vehicle itself determine the pseudonym lifetime? Answering these questions is far from trivial. Each choice can affect both the user privacy and the system performance and possibly, as a result, its security. In this paper, we make a novel systematic effort to address this multifaceted question. We craft three generally applicable policies and experimentally evaluate the VPKI system performance, leveraging two large-scale mobility datasets. We consider the most promising, in terms of efficiency, pseudonym acquisition policies; we find that within this class of policies, the most promising policy in terms of privacy protection can be supported with moderate overhead. Moreover, in all cases, this work is the first to provide tangible evidence that the state-of-the-art VPKI can serve sizable areas or domain with modest computing resources.

Mohammad Khodaei, Hongyu Jin, Panos Papadimitratos

Several years of academic and industrial research efforts have converged to a common understanding on fundamental security building blocks for the upcoming Vehicular Communication (VC) systems. There is a growing consensus towards deploying a Vehicular Public-Key Infrastructure (VPKI) enables pseudonymous authentication, with standardization efforts in that direction. However, there are still significant technical issues that remain unresolved. Existing proposals for instantiating the VPKI either need additional detailed specifications or enhanced security and privacy features. Equally important, there is limited experimental work that establishes the VPKI efficiency and scalability. In this paper, we are concerned with exactly these issues. We leverage the common VPKI approach and contribute an enhanced system with precisely defined, novel features that improve its resilience and the user privacy protection. In particular, we depart from the common assumption that the VPKI entities are fully trusted and we improve user privacy in the face of an honest-but-curious security infrastructure. Moreover, we fully implement our VPKI, in a standard-compliant manner, and we perform an extensive evaluation. Along with stronger protection and richer functionality, our system achieves very significant performance improvement over prior systems - contributing the most advanced VPKI towards deployment.

Mahtab Mirmohseni, Panagiotis Papadimitratos

We investigate the problem of reliable communication in the presence of active adversaries that can tamper with the transmitted data. We consider a legitimate transmitter-receiver pair connected over multiple communication paths (routes). We propose two new models of adversary, a "memoryless" and a "foreseer" adversary. For both models, the adversaries are placing themselves arbitrarily on the routes, keeping their placement fixed throughout the transmission block. This placement may or may not be known to the transmitter. The adversaries can choose their best modification strategy to increase the error at the legitimate receiver, subject to a maximum distortion constraint. We investigate the communication rates that can be achieved in the presence of the two types of adversaries and the channel (benign) stochastic behavior. For memoryless adversaries, the capacity is derived. Our method is to use the typical set of the anticipated received signal for all possible adversarial strategies (including their best one) in a compound channel that also captures adversarial placement. For the foreseer adversaries, which have enhanced observation capabilities compared to the memoryless ones, we propose a new coding scheme to guarantee resilience, i.e., recovery of the codeword independently of the adversarial (best) choice. We derive an achievable rate and we propose an upper bound on the capacity. We evaluate our general results for specific cases (e.g., binary symbol replacement or erasing attacks), to gain insights.

Mahtab Mirmohseni, Panagiotis Papadimitratos

We study the secrecy capacity in the vicinity of colluding eavesdroppers. Contrary to the perfect collusion assumption in previous works, our new information-theoretic model considers constraints in collusion. We derive the achievable secure rates (lower bounds on the perfect secrecy capacity), both for the discrete memoryless and Gaussian channels. We also compare the proposed rates to the non-colluding and perfect colluding cases.

Mahtab Mirmohseni, Panagiotis Papadimitratos

We investigate large wireless networks subject to security constraints. In contrast to point-to-point, interference-limited communications considered in prior works, we propose active cooperative relaying based schemes. We consider a network with $n_l$ legitimate nodes, $n_e$ eavesdroppers, and path loss exponent $α\geq 2$. As long as $n_e^2(\log(n_e))^γ=o(n_l)$, for some positive $γ$, we show one can obtain unbounded secure aggregate rate. This means zero-cost secure communication, given fixed total power constraint for the entire network. We achieve this result through (i) the source using Wyner randomized encoder and a serial (multi-stage) block Markov scheme, to cooperate with the relays and (ii) the relays acting as a virtual multi-antenna to apply beamforming against the eavesdroppers. Our simpler parallel (two-stage) relaying scheme can achieve the same unbounded secure aggregate rate when $n_e^{\fracα{2}+1}(\log(n_e))^{γ+δ(\fracα{2}+1)}=o(n_l)$ holds, for some positive $γ,δ$. Finally, we study the improvement (to the detriment of legitimate nodes) the eavesdroppers achieve in terms of the information leakage rate in a large cooperative network in case of collusion. We show that again the zero-cost secure communication is possible, if $n_e^{(2+\frac{2}α)}(\log n_e)^γ=o(n_l)$ holds, for some positive $γ$; i.e., in case of collusion slightly fewer eavesdroppers can be tolerated compared to the non-colluding case.