arXiv subjects

Gabriela Nicolescu

Publications and source records attributed to Gabriela Nicolescu.

15 recordsLinked to original sources

arXiv2026

Quantum Enhanced Anomaly Detection for ADS-B Data using Hybrid Deep Learning

The emerging field of Quantum Machine Learning (QML) has shown promising advantages in accelerating processing speed and effectively handling the high dimensionality associated with complex datasets. Quantum Computing (QC) enables more efficient data manipulation through the quantum properties of superposition and entanglement. In this paper, we present a novel approach combining quantum and classical machine learning techniques to explore the impact of quantum properties for anomaly detection in Automatic Dependent Surveillance-Broadcast (ADS-B) data. We compare the performance of a Hybrid-Fully Connected Quantum Neural Network (H-FQNN) with different loss functions and use a publicly available ADS-B dataset to evaluate the performance. The results demonstrate competitive performance in detecting anomalies, with accuracies ranging from 90.17% to 94.05%, comparable to the performance of a traditional Fully Connected Neural Network (FNN) model, which achieved accuracies between 91.50% and 93.37%.

quant-ph↗

arXiv2025

DivQAT: Enhancing Robustness of Quantized Convolutional Neural Networks against Model Extraction Attacks

Convolutional Neural Networks (CNNs) and their quantized counterparts are vulnerable to extraction attacks, posing a significant threat of IP theft. Yet, the robustness of quantized models against these attacks is little studied compared to large models. Previous defenses propose to inject calculated noise into the prediction probabilities. However, these defenses are limited since they are not incorporated during the model design and are only added as an afterthought after training. Additionally, most defense techniques are computationally expensive and often have unrealistic assumptions about the victim model that are not feasible in edge device implementations and do not apply to quantized models. In this paper, we propose DivQAT, a novel algorithm to train quantized CNNs based on Quantization Aware Training (QAT) aiming to enhance their robustness against extraction attacks. To the best of our knowledge, our technique is the first to modify the quantization process to integrate a model extraction defense into the training process. Through empirical validation on benchmark vision datasets, we demonstrate the efficacy of our technique in defending against model extraction attacks without compromising model accuracy. Furthermore, combining our quantization technique with other defense mechanisms improves their effectiveness compared to traditional QAT.

cs.LG↗

arXiv2025

New Machine Learning Approaches for Intrusion Detection in ADS-B

With the growing reliance on the vulnerable Automatic Dependent Surveillance-Broadcast (ADS-B) protocol in air traffic management (ATM), ensuring security is critical. This study investigates emerging machine learning models and training strategies to improve AI-based intrusion detection systems (IDS) for ADS-B. Focusing on ground-based ATM systems, we evaluate two deep learning IDS implementations: one using a transformer encoder and the other an extended Long Short-Term Memory (xLSTM) network, marking the first xLSTM-based IDS for ADS-B. A transfer learning strategy was employed, involving pre-training on benign ADS-B messages and fine-tuning with labeled data containing instances of tampered messages. Results show this approach outperforms existing methods, particularly in identifying subtle attacks that progressively undermine situational awareness. The xLSTM-based IDS achieves an F1-score of 98.9%, surpassing the transformer-based model at 94.3%. Tests on unseen attacks validated the generalization ability of the xLSTM model. Inference latency analysis shows that the 7.26-second delay introduced by the xLSTM-based IDS fits within the Secondary Surveillance Radar (SSR) refresh interval (5-12 s), although it may be restrictive for time-critical operations. While the transformer-based IDS achieves a 2.1-second latency, it does so at the cost of lower detection performance.

cs.CR↗

arXiv2025

Implementation of the Collision Avoidance System for DO-178C Compliance

This technical report presents the detailed implementation of a Collision Avoidance System (CAS) for Unmanned Aerial Vehicles (UAVs), developed as a case study to demonstrate a rigorous methodology for achieving DO-178C compliance in safety-critical software. The CAS is based on functional requirements inspired by NASA's Access 5 project and is designed to autonomously detect, evaluate, and avoid potential collision threats in real-time, supporting the safe integration of UAVs into civil airspace. The implementation environment combines formal methods, model-based development, and automated verification tools, including Alloy, SPIN, Simulink Embedded Coder, and the LDRA tool suite. The report documents each phase of the software lifecycle: requirements specification and validation, architectural and detailed design, coding, verification, and traceability, with a strong focus on compliance with DO-178C Design Assurance Level B objectives. Results demonstrate that formal modelling and automated toolchains enabled early detection and correction of specification defects, robust traceability, and strong evidence of verification and validation across all development stages. Static and dynamic analyses confirmed code quality and coverage, while formal verification methods provided mathematical assurance of correctness for critical components. Although the integration phase was not fully implemented, the approach proved effective in addressing certification challenges for UAV safety-critical systems. \keywords Collision Avoidance System (CAS), Unmanned Aerial Vehicles (UAVs), DO-178C compliance, Safety-critical software, Formal methods, Model-based development, Alloy, SPIN model checker, Simulink Embedded Coder, LDRA tool suite, Software verification and validation, Traceability, Certification.

cs.SE↗

arXiv2024

CABBA: Compatible Authenticated Bandwidth-efficient Broadcast protocol for ADS-B

The Automatic Dependent Surveillance-Broadcast (ADS-B) is a surveillance technology that mandated in many airspaces. It improves safety, increases efficiency and reduces air traffic congestion by broadcasting aircraft navigation data. Yet, ADS-B is vulnerable to spoofing attacks as it lacks mechanisms to ensure the integrity and authenticity of the data being supplied. None of the existing cryptographic solutions fully meet the backward compatibility and bandwidth preservation requirements of the standard. Hence, we propose the Compatible Authenticated Bandwidth-efficient Broadcast protocol for ADS-B (CABBA), an improved approach that integrates TESLA, phase-overlay modulation techniques and certificate-based PKI. As a result, entity authentication, data origin authentication, and data integrity are the security services that CABBA offers. To assess compliance with the standard, we designed an SDR-based implementation of CABBA and performed backward compatibility tests on commercial and general aviation (GA) ADS-B in receivers. Besides, we calculated the 1090ES band's activity factor and analyzed the channel occupancy rate according to ITU-R SM.2256-1 recommendation. Also, we performed a bit error rate analysis of CABBA messages. The results suggest that CABBA is backward compatible, does not incur significant communication overhead, and has an error rate that is acceptable for Eb/No values above 14 dB.

cs.CR↗

arXiv2023

Efficient Defense Against Model Stealing Attacks on Convolutional Neural Networks

Model stealing attacks have become a serious concern for deep learning models, where an attacker can steal a trained model by querying its black-box API. This can lead to intellectual property theft and other security and privacy risks. The current state-of-the-art defenses against model stealing attacks suggest adding perturbations to the prediction probabilities. However, they suffer from heavy computations and make impracticable assumptions about the adversary. They often require the training of auxiliary models. This can be time-consuming and resource-intensive which hinders the deployment of these defenses in real-world applications. In this paper, we propose a simple yet effective and efficient defense alternative. We introduce a heuristic approach to perturb the output probabilities. The proposed defense can be easily integrated into models without additional training. We show that our defense is effective in defending against three state-of-the-art stealing attacks. We evaluate our approach on large and quantized (i.e., compressed) Convolutional Neural Networks (CNNs) trained on several vision datasets. Our technique outperforms the state-of-the-art defenses with a $\times37$ faster inference latency without requiring any additional model and with a low impact on the model's performance. We validate that our defense is also effective for quantized CNNs targeting edge devices.

cs.LG↗

arXiv2023

Integrated Photonic AI Accelerators under Hardware Security Attacks: Impacts and Countermeasures

Integrated photonics based on silicon photonics platform is driving several application domains, from enabling ultra-fast chip-scale communication in high-performance computing systems to energy-efficient optical computation in artificial intelligence (AI) hardware accelerators. Integrating silicon photonics into a system necessitates the adoption of interfaces between the photonic and the electronic subsystems, which are required for buffering data and optical-to-electrical and electrical-to-optical conversions. Consequently, this can lead to new and inevitable security breaches that cannot be fully addressed using hardware security solutions proposed for purely electronic systems. This paper explores different types of attacks profiting from such breaches in integrated photonic neural network accelerators. We show the impact of these attacks on the system performance (i.e., power and phase distributions, which impact accuracy) and possible solutions to counter such attacks.

cs.ET↗

arXiv2023

SerIOS: Enhancing Hardware Security in Integrated Optoelectronic Systems

Silicon photonics (SiPh) has different applications, from enabling fast and high-bandwidth communication for high-performance computing systems to realizing energy-efficient optical computation for AI hardware accelerators. However, integrating SiPh with electronic sub-systems can introduce new security vulnerabilities that cannot be adequately addressed using existing hardware security solutions for electronic systems. This paper introduces SerIOS, the first framework aimed at enhancing hardware security in optoelectronic systems by leveraging the unique properties of optical lithography. SerIOS employs cryptographic keys generated based on imperfections in the optical lithography process and an online detection mechanism to detect attacks. Simulation and synthesis results demonstrate SerIOS's effectiveness in detecting and preventing attacks, with a small area footprint of less than 15% and a 100% detection rate across various attack scenarios and optoelectronic architectures, including photonic AI accelerators.

cs.ET↗

arXiv2022

Careful What You Wish For: on the Extraction of Adversarially Trained Models

Recent attacks on Machine Learning (ML) models such as evasion attacks with adversarial examples and models stealing through extraction attacks pose several security and privacy threats. Prior work proposes to use adversarial training to secure models from adversarial examples that can evade the classification of a model and deteriorate its performance. However, this protection technique affects the model's decision boundary and its prediction probabilities, hence it might raise model privacy risks. In fact, a malicious user using only a query access to the prediction output of a model can extract it and obtain a high-accuracy and high-fidelity surrogate model. To have a greater extraction, these attacks leverage the prediction probabilities of the victim model. Indeed, all previous work on extraction attacks do not take into consideration the changes in the training process for security purposes. In this paper, we propose a framework to assess extraction attacks on adversarially trained models with vision datasets. To the best of our knowledge, our work is the first to perform such evaluation. Through an extensive empirical study, we demonstrate that adversarially trained models are more vulnerable to extraction attacks than models obtained under natural training circumstances. They can achieve up to $\times1.2$ higher accuracy and agreement with a fraction lower than $\times0.75$ of the queries. We additionally find that the adversarial robustness capability is transferable through extraction attacks, i.e., extracted Deep Neural Networks (DNNs) from robust models show an enhanced accuracy to adversarial examples compared to extracted DNNs from naturally trained (i.e. standard) models.

cs.LG↗

arXiv2021

Vehicle Detection and Tracking From Surveillance Cameras in Urban Scenes

Detecting and tracking vehicles in urban scenes is a crucial step in many traffic-related applications as it helps to improve road user safety among other benefits. Various challenges remain unresolved in multi-object tracking (MOT) including target information description, long-term occlusions and fast motion. We propose a multi-vehicle detection and tracking system following the tracking-by-detection paradigm that tackles the previously mentioned challenges. Our MOT method extends an Intersection-over-Union (IOU)-based tracker with vehicle re-identification features. This allows us to utilize appearance information to better match objects after long occlusion phases and/or when object location is significantly shifted due to fast motion. We outperform our baseline MOT method on the UA-DETRAC benchmark while maintaining a total processing speed suitable for online use cases.

cs.CV↗

arXiv2020

On the benchmarking of partitioned real-time systems

Avionic software is the subject of critical real time, determinism and safety constraints. Software designers face several challenges, one of them being the estimation of worst-case execution time (WCET) of applications, that dictates the execution time of the system. A pessimistic WCET estimation can lead to low execution performances of the system, while an over-optimistic estimation can lead to deadline misses, breaking one the basic constraints of critical real-time systems (RTS). Partitioned systems are one special category of real time systems, employed by the avionic community to deploy avionic software. The ARINC-653 standard is one common avionic standard that employs the concept of partitions. This standard defines partitioned architectures where one partition should never directly interfere with another one. Assessing WCET of general purpose RTSs is achievable by the usage of one of the many published benchmark or WCET estimation frameworks. Contrarily, partitioned RTSs are special cases, in which common benchmark tools may not capture all the metrics. In this document, we present SFPBench, a generic benchmark framework for the assessment of performance metrics on partitioned RTSs. The general organization of the framework and its applications are illustrated, as well as an use-case, employing SFPBench on an industrial partitioned operating system (OS) executing on a Commercial Off-The-shelf (COTS) processor.

cs.SE↗

arXiv2015

Optical Crossbars on Chip: a comparative study based on worst-case losses

The many cores design research community have shown high interest in optical crossbars on chip for more than a decade. Key properties of optical crossbars, namely a) contention free data routing b) low latency communication and c) potential for high bandwidth through the use of WDM, motivate several implementations of this type of interconnect. These implementations demonstrate very different scalability and power efficiency ability depending on three key design factors: a) the network topology, b) the considered layout and the c) the injection losses induced by the fabrication process. In this paper, the worst-case optical losses of crossbar implementations are compared according to the factors mentioned above. The comparison results has the potential to help many cores system designer to select the most appropriate crossbar implementation according, for instance, to the number of IP cores and the chip die size.

cs.ET↗

arXiv2015

Energy-efficient optical crossbars on chip with multi-layer deposited silicon

The many cores design research community have shown high interest in optical crossbars on chip for more than a decade. Key properties of optical crossbars, namely a) contention-free data routing b) low-latency communication and c) potential for high bandwidth through the use of WDM, motivate several implementations. These implementations demonstrate very different scalability and power efficiency ability depending on three key design factors: a) the network topology, b) the considered layout and c) the insertion losses induced by the fabrication process. The emerging design technique relying on multi-layer deposited silicon allows reducing optical losses, which may lead to significant reduction of the power consumption. In this paper, multi-layer deposited silicon based crossbars are proposed and compared. The results indicate that the proposed ring-based network exhibits, on average, 22% and 51.4% improvement for worst-case and average losses respectively compared to the most power-efficient related crossbars.

cs.ET↗

arXiv2015

Thermal Aware Design Method for VCSEL-Based On-Chip Optical Interconnect

Optical Network-on-Chip (ONoC) is an emerging technology considered as one of the key solutions for future generation on-chip interconnects. However, silicon photonic devices in ONoC are highly sensitive to temperature variation, which leads to a lower efficiency of Vertical-Cavity Surface-Emitting Lasers (VCSELs), a resonant wavelength shift of Microring Resonators (MR), and results in a lower Signal to Noise Ratio (SNR). In this paper, we propose a methodology enabling thermal-aware design for optical interconnects relying on CMOS-compatible VCSEL. Thermal simulations allow designing ONoC interfaces with low gradient temperature and analytical models allow evaluating the SNR.

cs.ET↗

arXiv2013

On the Verification of a WiMax Design Using Symbolic Simulation

In top-down multi-level design methodologies, design descriptions at higher levels of abstraction are incrementally refined to the final realizations. Simulation based techniques have traditionally been used to verify that such model refinements do not change the design functionality. Unfortunately, with computer simulations it is not possible to completely check that a design transformation is correct in a reasonable amount of time, as the number of test patterns required to do so increase exponentially with the number of system state variables. In this paper, we propose a methodology for the verification of conformance of models generated at higher levels of abstraction in the design process to the design specifications. We model the system behavior using sequence of recurrence equations. We then use symbolic simulation together with equivalence checking and property checking techniques for design verification. Using our proposed method, we have verified the equivalence of three WiMax system models at different levels of design abstraction, and the correctness of various system properties on those models. Our symbolic modeling and verification experiments show that the proposed verification methodology provides performance advantage over its numerical counterpart.

cs.LO↗