arXiv subjects

Gurcan Comert

Publications and source records attributed to Gurcan Comert.

At least 37 records · Page 2Linked to original sources

arXiv2025

A Multi-Scale Isolation Forest Approach for Real-Time Detection and Filtering of FGSM Adversarial Attacks in Video Streams of Autonomous Vehicles

Deep Neural Networks (DNNs) have demonstrated remarkable success across a wide range of tasks, particularly in fields such as image classification. However, DNNs are highly susceptible to adversarial attacks, where subtle perturbations are introduced to input images, leading to erroneous model outputs. In today's digital era, ensuring the security and integrity of images processed by DNNs is of critical importance. One of the most prominent adversarial attack methods is the Fast Gradient Sign Method (FGSM), which perturbs images in the direction of the loss gradient to deceive the model. This paper presents a novel approach for detecting and filtering FGSM adversarial attacks in image processing tasks. Our proposed method evaluates 10,000 images, each subjected to five different levels of perturbation, characterized by $ε$ values of 0.01, 0.02, 0.05, 0.1, and 0.2. These perturbations are applied in the direction of the loss gradient. We demonstrate that our approach effectively filters adversarially perturbed images, mitigating the impact of FGSM attacks. The method is implemented in Python, and the source code is publicly available on GitHub for reproducibility and further research.

cs.CV↗

arXiv2025

Analysis of Distracted Pedestrians Crossing Behavior: An Immersive Virtual Reality Application

Pedestrian safety is a critical public health priority, with pedestrian fatalities accounting for 18% of all U.S. traffic deaths in 2022. The rising prevalence of distracted walking, exacerbated by mobile device use, poses significant risks at signalized intersections. This study utilized an immersive virtual reality (VR) environment to simulate real-world traffic scenarios and assess pedestrian behavior under three conditions: undistracted crossing, crossing while using a mobile device, and crossing with Light-emitting diode (LED) safety interventions. Analysis using ANOVA models identified speed and mobile-focused eye-tracking as significant predictors of crossing duration, revealing how distractions impair situational awareness and response times. While LED measures reduced delays, their limited effectiveness highlights the need for integrated strategies addressing both behavioral and physical factors. This study showcases VRs potential to analyze complex pedestrian behaviors, offering actionable insights for urban planners and policymakers aiming to enhance pedestrian safety.

cs.HC↗

arXiv2025

A Systematic Evaluation of Generative Models on Tabular Transportation Data

The sharing of large-scale transportation data is beneficial for transportation planning and policymaking. However, it also raises significant security and privacy concerns, as the data may include identifiable personal information, such as individuals' home locations. To address these concerns, synthetic data generation based on real transportation data offers a promising solution that allows privacy protection while potentially preserving data utility. Although there are various synthetic data generation techniques, they are often not tailored to the unique characteristics of transportation data, such as the inherent structure of transportation networks formed by all trips in the datasets. In this paper, we use New York City taxi data as a case study to conduct a systematic evaluation of the performance of widely used tabular data generative models. In addition to traditional metrics such as distribution similarity, coverage, and privacy preservation, we propose a novel graph-based metric tailored specifically for transportation data. This metric evaluates the similarity between real and synthetic transportation networks, providing potentially deeper insights into their structural and functional alignment. We also introduced an improved privacy metric to address the limitations of the commonly-used one. Our experimental results reveal that existing tabular data generative models often fail to perform as consistently as claimed in the literature, particularly when applied to transportation data use cases. Furthermore, our novel graph metric reveals a significant gap between synthetic and real data. This work underscores the potential need to develop generative models specifically tailored to take advantage of the unique characteristics of emerging domains, such as transportation.

cs.LG↗

arXiv2024

Quantum Adversarial Machine Learning and Defense Strategies: Challenges and Opportunities

As quantum computing continues to advance, the development of quantum-secure neural networks is crucial to prevent adversarial attacks. This paper proposes three quantum-secure design principles: (1) using post-quantum cryptography, (2) employing quantum-resistant neural network architectures, and (3) ensuring transparent and accountable development and deployment. These principles are supported by various quantum strategies, including quantum data anonymization, quantum-resistant neural networks, and quantum encryption. The paper also identifies open issues in quantum security, privacy, and trust, and recommends exploring adaptive adversarial attacks and auto adversarial attacks as future directions. The proposed design principles and recommendations provide guidance for developing quantum-secure neural networks, ensuring the integrity and reliability of machine learning models in the quantum era.

quant-ph↗

arXiv2024

An AutoML-based approach for Network Intrusion Detection

In this paper, we present an automated machine learning (AutoML) approach for network intrusion detection, leveraging a stacked ensemble model developed using the MLJAR AutoML framework. Our methodology combines multiple machine learning algorithms, including LightGBM, CatBoost, and XGBoost, to enhance detection accuracy and robustness. By automating model selection, feature engineering, and hyperparameter tuning, our approach reduces the manual overhead typically associated with traditional machine learning methods. Extensive experimentation on the NSL-KDD dataset demonstrates that the stacked ensemble model outperforms individual models, achieving high accuracy and minimizing false positives. Our findings underscore the benefits of using AutoML for network intrusion detection, as the AutoML-driven stacked ensemble achieved the highest performance with 90\% accuracy and an 89\% F1 score, outperforming individual models like Random Forest (78\% accuracy, 78\% F1 score), XGBoost and CatBoost (both 80\% accuracy, 80\% F1 score), and LightGBM (78\% accuracy, 78\% F1 score), providing a more adaptable and efficient solution for network security applications.

cs.LG↗

arXiv2023

Development and Evaluation of Ensemble Learning-based Environmental Methane Detection and Intensity Prediction Models

The environmental impacts of global warming driven by methane (CH4) emissions have catalyzed significant research initiatives in developing novel technologies that enable proactive and rapid detection of CH4. Several data-driven machine learning (ML) models were tested to determine how well they identified fugitive CH4 and its related intensity in the affected areas. Various meteorological characteristics, including wind speed, temperature, pressure, relative humidity, water vapor, and heat flux, were included in the simulation. We used the ensemble learning method to determine the best-performing weighted ensemble ML models built upon several weaker lower-layer ML models to (i) detect the presence of CH4 as a classification problem and (ii) predict the intensity of CH4 as a regression problem.

cs.LG↗

arXiv2023

Autonomous Vehicles for All?

The traditional build-and-expand approach is not a viable solution to keep roadway traffic rolling safely, so technological solutions, such as Autonomous Vehicles (AVs), are favored. AVs have considerable potential to increase the carrying capacity of roads, ameliorate the chore of driving, improve safety, provide mobility for those who cannot drive, and help the environment. However, they also raise concerns over whether they are socially responsible, accounting for issues such as fairness, equity, and transparency. Regulatory bodies have focused on AV safety, cybersecurity, privacy, and legal liability issues, but have failed to adequately address social responsibility. Thus, existing AV developers do not have to embed social responsibility factors in their proprietary technology. Adverse bias may therefore occur in the development and deployment of AV technology. For instance, an artificial intelligence-based pedestrian detection application used in an AV may, in limited lighting conditions, be biased to detect pedestrians who belong to a particular racial demographic more efficiently compared to pedestrians from other racial demographics. Also, AV technologies tend to be costly, with a unique hardware and software setup which may be beyond the reach of lower-income people. In addition, data generated by AVs about their users may be misused by third parties such as corporations, criminals, or even foreign governments. AVs promise to dramatically impact labor markets, as many jobs that involve driving will be made redundant. We argue that the academic institutions, industry, and government agencies overseeing AV development and deployment must act proactively to ensure that AVs serve all and do not increase the digital divide in our society.

cs.CY↗

arXiv2021

A Combinatorial Approach for Nonparametric Short-Term Estimation of Queue Lengths using Probe Vehicles

Traffic state estimation plays an important role in facilitating effective traffic management. This study develops a combinatorial approach for nonparametric short-term queue length estimation in terms of cycle-by-cycle partially observed queues from probe vehicles. The method does not assume random arrivals and does not assume any primary parameters or estimation of any parameters but uses simple algebraic expressions that only depend on signal timing. For an approach lane at a traffic intersection, the conditional queue lengths given probe vehicle location, count, time, and analysis interval (e.g., at the end of red signal phase) are represented by a Negative Hypergeometric distribution. The estimators obtained are compared with parametric methods and simple highway capacity manual methods using field test data involving probe vehicles. The analysis indicates that the nonparametric methods presented in this paper match the accuracy of parametric methods used in the field test data for estimating queue lengths.

stat.ME↗

arXiv2021

The Effect of Dust and Sand on the 5G Terrestrial Links

Wireless connections are a communication channel used to support different applications in our life such as microwave connections, mobile cellular networks, and intelligent transportation systems. The wireless communication channels are affected by different weather factors such as rain, snow, fog, dust, and sand. This effect is more evident in the high frequencies of the millimeter-wave (mm-wave) band. Recently, the 5G opened the door to support different applications with high speed and good quality. A recent study investigates the effect of rain and snow on the 5G communication channel to reduce the challenge of using high millimeter-wave frequencies. This research investigates the impact of dust and sand on the communication channel of 5G mini links using Mie scattering model to estimate the propagating wave's attenuation by computing the free space loss of a dusty region. Also, the cross-polarization of the propagating wave with dust and sand is taken into account at different distances of the propagating length. Two kinds of mini links, ML-6363, and ML-6352, are considered to demonstrate the effect of dust and sand in these specific operating frequency bands. The 73.5 GHz (V-band) and (21.5GHz (K-band) are the ML-6352 and ML-6363 radio frequency, respectively. Also, signal depolarization is another important radio frequency transmission parameter that is considered heroin. The numerical and simulation results show that the 5G ML-6352 is more effect by dust and sand than ML6363. The 5G toolbox is used to build the communication system and simulate the effect of the dust and sand on the different frequency bands.

eess.SP↗

arXiv2021

Impact of Weather Conditions on 5G Communication Channel under Connected Vehicles Framework

Recent research focused on improving the vehicle-to-vehicle communication (V2V) based on the 5G technology. The V2V application is important because it will reduce the risk of accidents up to 70%-80%, improve traffic management, reduce congestion, and improve fuel consumption. Autonomous vehicles applications require a high bandwidth transmission channel where the 5G communication channel would be a reliable solution to support this disruptive technology. The dedicated short-range communications (DSRC), which is characterized with a frequency bandwidth of 5.9 gigahertz (GHz) (4G spectrum), was used as vehicular connectivity with a bandwidth of up to 200 megabytes per second (mb/s) and limited capacity. The 5G band can support connected multiple autonomous vehicles with high data rates and large bandwidth. In this study, the 5G communication channel is considered as vehicular connectivity with high bandwidth in the millimeter waves spectrum range. The quality of 5G wireless communication channels between connected vehicles possibly be affected by weather conditions such as rain, snow, fog, dust, and sand. In this paper, we estimate the effect of dust and sand on the propagation of millimeter waves. The Mie model is used to investigate the effect of dust and sand storms on the propagating mm-waves. The effect of dust and sand on the communication path loss of DSRC and 5G frequency band is investigated in the case of urban freeway and rural highway settings. Results show that the attenuation of dust and sand is changed when the particle size of sand, frequency of propagating wave, and concentration of dust are changed. Finally, the new model of link margin is created to estimate the effect of dust and sand on DSCR (5.9 GHz) and 5G (28 GHz) communication path loss.

cs.NI↗

arXiv2021

Bayesian Parameter Estimations for Grey System Models in Online Traffic Speed Predictions

This paper presents Bayesian parameter estimation for first order Grey system models' parameters (or sometimes referred to as hyperparameters). There are different forms of first-order Grey System Models. These include $GM(1,1)$, $GM(1,1| \cos(ωt)$, $GM(1,1| \sin(ωt)$, and $GM(1,1| \cos(ωt), \sin(ωt)$. The whitenization equation of these models is a first-order linear differential equation of the form \[ \frac{dx}{dt} + a x = f(t) \] where $a$ is a parameter and $f(t) = b$ in $GM(1,1|)$ , $f(t) = b_1\cos(ωt) + b_2$ in $GM(1,1| cos(ωt)$, $f(t) = b_1\sin(ωt)+b_2$ in $GM(1,1| \sin(ωt)$, $f(t) = b_1\sin(ωt) + b_2\cos(ωt) + b_3$ in $GM(1,1| \cos(ωt), \sin(ωt)$, $f(t) = b x^2$ in Grey Verhulst model (GVM), and where $b, b_1, b_2$, and $b_3$ are parameters. The results from Bayesian estimations are compared to the least square estimated models with fixed $ω$. We found that using rolling Bayesian estimations for GM parameters can allow us to estimate the parameters in all possible forms. Based on the data used for the comparison, the numerical results showed that models with Bayesian parameter estimations are up to 45\% more accurate in mean squared errors.

stat.ME↗

arXiv2021

Efficacy of Statistical and Artificial Intelligence-based False Information Cyberattack Detection Models for Connected Vehicles

Connected vehicles (CVs), because of the external connectivity with other CVs and connected infrastructure, are vulnerable to cyberattacks that can instantly compromise the safety of the vehicle itself and other connected vehicles and roadway infrastructure. One such cyberattack is the false information attack, where an external attacker injects inaccurate information into the connected vehicles and eventually can cause catastrophic consequences by compromising safety-critical applications like the forward collision warning. The occurrence and target of such attack events can be very dynamic, making real-time and near-real-time detection challenging. Change point models, can be used for real-time anomaly detection caused by the false information attack. In this paper, we have evaluated three change point-based statistical models; Expectation Maximization, Cumulative Summation, and Bayesian Online Change Point Algorithms for cyberattack detection in the CV data. Also, data-driven artificial intelligence (AI) models, which can be used to detect known and unknown underlying patterns in the dataset, have the potential of detecting a real-time anomaly in the CV data. We have used six AI models to detect false information attacks and compared the performance for detecting the attacks with our developed change point models. Our study shows that change points models performed better in real-time false information attack detection compared to the performance of the AI models. Change point models having the advantage of no training requirements can be a feasible and computationally efficient alternative to AI models for false information attack detection in connected vehicles.

cs.CR↗

arXiv2021

Hybrid Classical-Quantum Deep Learning Models for Autonomous Vehicle Traffic Image Classification Under Adversarial Attack

Image classification must work for autonomous vehicles (AV) operating on public roads, and actions performed based on image misclassification can have serious consequences. Traffic sign images can be misclassified by an adversarial attack on machine learning models used by AVs for traffic sign recognition. To make classification models resilient against adversarial attacks, we used a hybrid deep-learning model with both the quantum and classical layers. Our goal is to study the hybrid deep-learning architecture for classical-quantum transfer learning models to support the current era of intermediate-scale quantum technology. We have evaluated the impacts of various white box adversarial attacks on these hybrid models. The classical part of hybrid models includes a convolution network from the pre-trained Resnet18 model, which extracts informative features from a high dimensional LISA traffic sign image dataset. The output from the classical processor is processed further through the quantum layer, which is composed of various quantum gates and provides support to various quantum mechanical features like entanglement and superposition. We have tested multiple combinations of quantum circuits to provide better classification accuracy with decreasing training data and found better resiliency for our hybrid classical-quantum deep learning model during attacks compared to the classical-only machine learning models.

quant-ph↗

arXiv2021

Hybrid Quantum-Classical Neural Network for Incident Detection

The efficiency and reliability of real-time incident detection models directly impact the affected corridors' traffic safety and operational conditions. The recent emergence of cloud-based quantum computing infrastructure and innovations in noisy intermediate-scale quantum devices have revealed a new era of quantum-enhanced algorithms that can be leveraged to improve real-time incident detection accuracy. In this research, a hybrid machine learning model, which includes classical and quantum machine learning (ML) models, is developed to identify incidents using the connected vehicle (CV) data. The incident detection performance of the hybrid model is evaluated against baseline classical ML models. The framework is evaluated using data from a microsimulation tool for different incident scenarios. The results indicate that a hybrid neural network containing a 4-qubit quantum layer outperforms all other baseline models when there is a lack of training data. We have created three datasets; DS-1 with sufficient training data, and DS-2 and DS-3 with insufficient training data. The hybrid model achieves a recall of 98.9%, 98.3%, and 96.6% for DS-1, DS-2, and DS-3, respectively. For DS-2 and DS-3, the average improvement in F2-score (measures model's performance to correctly identify incidents) achieved by the hybrid model is 1.9% and 7.8%, respectively, compared to the classical models. It shows that with insufficient data, which may be common for CVs, the hybrid ML model will perform better than the classical models. With the continuing improvements of quantum computing infrastructure, the quantum ML models could be a promising alternative for CV-related applications when the available data is insufficient.

cs.LG↗

arXiv2020

Evaluation of Sustainable Green Materials: Pinecone in Permeable Adsorptive Barriers for Remediation of Groundwater Contaminated by $Pb^{2+}$ and Methylene Blue

We report herein, the potential of raw pinecone powder (PCP) and pinecone biochar (PCBC) as alternatives to activated carbon used in Permeable Adsorptive Barriers (PABs) for the in situ remediation of polluted groundwater. A constructed lab-scale unconfined aquifer ($38\times30\times17$ $cm$) fitted with PCP and PCBC PABs ($21\times3\times20$ $cm$), was evaluated for the removal of $Pb^{2+}$ ions in a continuous flow setup. Results indicate that after $3600$ minutes, PCP was able to reduce $Pb^{2+}$ ions from a Co=$50$ $mg/L$ to $7.94$ $mg/L$ for the first run and $19.4$ $mg/L$ for a second run, respectively. Comparatively, PCBC reached $6.5$ $mg/L$ for the first run and $8.94$ for the second run. It was confirmed that adsorption was best described by the first-order kinetic model with $R^2$ values above $0.95$. Maximum adsorption capacity values were found to be $1.00$, $0.63$, $1.08$, and $0.85$ mg/g for each scenario respectively. In addition, nonlinear regression models of exponential and Gaussian Processes are fit to explain remediation by time for $Pb^{2+}$ and Methylene Blue. Gaussian Processes are able to better explain the variation of pollution removal compared to simpler exponential models. When regressed against true removal percentages all models are able to provide $R^2>0.99$.

stat.AP↗

arXiv2020

Gaussian Processes for Traffic Speed Prediction at Different Aggregation Levels

Dynamic behavior of traffic adversely affect the performance of the prediction models in intelligent transportation applications. This study applies Gaussian processes (GPs) to traffic speed prediction. Such predictions can be used by various transportation applications, such as real-time route guidance, ramp metering, congestion pricing and special events traffic management. One-step predictions with various aggregation levels (1 to 60-minute) are tested for performance of the generated models. Univariate and multivariate GPs are compared with several other linear, nonlinear time series, and Grey system models using loop and Inrix probe vehicle datasets from California, Portland, and Virginia freeways respectively. Based on the test data samples, results are promising that GP models are able to consistently outperform compared models with similar computational times.

stat.AP↗

arXiv2020

Cycle-to-Cycle Queue Length Estimation from Connected Vehicles with Filtering on Primary Parameters

Estimation models from connected vehicles often assume low level parameters such as arrival rates and market penetration rates as known or estimate them in real-time. At low market penetration rates, such parameter estimators produce large errors making estimated queue lengths inefficient for control or operations applications. In order to improve accuracy of low level parameter estimations, this study investigates the impact of connected vehicles information filtering on queue length estimation models. Filters are used as multilevel real-time estimators. Accuracy is tested against known arrival rate and market penetration rate scenarios using microsimulations. To understand the effectiveness for short-term or for dynamic processes, arrival rates, and market penetration rates are changed every 15 minutes. The results show that with Kalman and Particle filters, parameter estimators are able to find the true values within 15 minutes and meet and surpass the accuracy of known parameter scenarios especially for low market penetration rates. In addition, using last known estimated queue lengths when no connected vehicle is present performs better than inputting average estimated values. Moreover, the study shows that both filtering algorithms are suitable for real-time applications that require less than 0.1 second computational time.

stat.AP↗

arXiv2020

Queue Length Estimation at Traffic Signals: Connected Vehicles with Range Measurement Sensors

Today vehicles are becoming a rich source of data as they are equipped with localization or tracking and with wireless communications technologies. With the increasing interest in automated- or self- driving technologies, vehicles are also being equipped with range measuring sensors (e.g., LIDAR, stereo cameras, and ultrasonic) to detect other vehicles and objects in the surrounding environment. It is possible to envision that such vehicles could share their data with the transportation infrastructure elements (e.g., a traffic signal controller) to enable different mobility and safety applications. Data from these connected vehicles could then be used to estimate the system state in real-time. This paper develops queue length estimators from connected vehicles equipped with range measurement sensors. Simple plug-and-play models are developed for queue length estimations without needing ground truth queue lengths by extending the previous formulations. The proposed method is simple to implement and can be adopted to cyclic queues at traffic signals with known phase lengths. The derived models are evaluated with data from microscopic traffic simulations. From numerical experiments, the QLE model with range sensors improves the errors as much as 25% in variance-to-mean ratio and 5\% in coefficient of variation at low less than 20% market penetration rates.

stat.AP↗