Gaze estimation systems powered by deep neural networks are commonly used in sensitive applications such as driver assist or human-computer interaction. While backdoor attacks have been widely studied for classification tasks, vulnerability of regression networks like gaze estimators to these kind of attacks still remain underexplored. This research investigates the effectiveness of full-image backdoor attacks on appearance-based gaze estimation models. Specifically, the study explores dirty-label attacks with two types of global backdoor triggers: a spatial-domain sinusoidal pattern and a randomized frequency-domain perturbation. Experimental results on the MPIIFaceGaze dataset demonstrate that both triggers can reliably induce malicious outputs while preserving high accuracy on clean data, with the frequency-domain trigger offering superior stealth. These findings highlight a significant vulnerability in deep regression models, emphasizing the need for defensive mechanisms in real-world gaze estimation systems.

The global adoption of 5G technology is rapidly accelerating and 5G traffic is growing exponentially. This increase in demand compels network operators to evaluate whether their current and upcoming 5G infrastructure can effectively accommodate the growing data traffic. A key component within the 5G network is the User Plane Function (UPF), which connects the end-devices to the data networks. Therefore, it is vital for both equipment manufacturers and service providers to analyze the performance of existing UPF implementations. This paper presents an initial approach to assess the Open5GS UPF performance by conducting stress testing in virtualized environment. We focus on finding the optimal UPF configuration by measuring the resource consumption and latency of the UPF under varying intensity of generated traffic, and providing a simple queueing model for the UPF. Numerical results show that a CPU load of 70-80% balances between latency and throughput while ensuring that 99% of the packets are forwarded within 150 ms.

With the exponential growth of mobile traffic in 5G networks, accurate forecasting is essential for efficient resource management. This research provides a comparative analysis of time series forecasting models for predicting near-future network traffic. Using a public dataset from a 5G base station in Barcelona, this study evaluates the performance of a traditional statistical model, against deep learning models: a Recurrent Neural Network (RNN), a Long Short-Term Memory (LSTM) and a Google timesFM model. The results demonstrate that while the SARIMAX model struggles to capture near-future traffic demand, the deep learning approaches yield significantly higher predictive accuracy. Specifically, a simple LSTM architecture shows great results, outperforming even a more complex one. However, the timesFM model, in particular, shows the most robust generalization capabilities. Additionally, the models trained on data from one base station do not generalize well to others, highlighting significant differences in traffic characteristics even between geographically close locations. This suggests that while locally trained LSTMs are a powerful tool, future work should focus on developing more adaptive and transferable models, such as those using federated learning or graph neural networks.^p

Academic research in 5G networking faces a lack of accessible, realistic packet-level datasets, limiting innovation and reproducibility. This paper evaluates two state-of-the-art machine learning approaches, PAC-GAN and TabularARGN, for generating synthetic 5G TCP/IP packet headers. Using a real 5G packet-capture dataset, we adapt both models to include inter-packet timing and rigorously assess them on protocol validity, marginal distribution alignment, and joint distribution fidelity. Results show that PAC-GAN produces highly valid and statistically faithful synthetic packets, effectively modeling complex header dependencies and temporal patterns. While TabularARGN ensures strict protocol compliance, it struggles to capture higher-order correlations and traffic diversity. Our findings establish convolutional generative models like PAC-GAN as practical tools for producing realistic, protocol-compliant synthetic 5G traffic, broadening access to datasets for benchmarking and security testing.

Modern mobile networks must adapt to rapidly changing traffic patterns and increasing user demands. A key challenge is understanding where user traffic terminates and how these destinations vary over time. This thesis addresses this challenge by introducing an open-source, modular analysis framework that analyzes passive Internet traffic traces, enriches them with geolocation and organizational metadata, and infers latency stability and routing dynamics, in order to characterize the infrastructures that terminate user traffic and assess their performance and reliability over time. The results show a long-term shift towards content-centric traffic, highlight geographic and temporal variations in performance, and demonstrate that content networks typically offer greater stability than enterprise or research destinations. These findings support adaptive traffic management strategies in 5G and future 6G networks.

This paper investigates the latency and resilience of user-plane anchor reconfiguration in a fully virtualized 5G core environment using Open5GS and UERANSIM. The experiment spans five VirtualBox virtual machines, each hosting a key component of the 5G core or radio stack: 5G-core gNB, UPF1, UPF2, and a single UE. All nodes communicate over a shared internal network, ensuring controland user-plane traffic remains isolated from external variability. The UE is initially anchored to UPF 1 via DNN “internet.” After the initial tunnel is established and validated, a re-anchoring procedure is triggered by calling the SMF’s REST API. Although the endpoint is intended to perform a PFCP Session Modification, Open5GS tears down the session and creates a new one on UPF 2 instead. By analyzing timestamped UE logs—capturing tunnel setup, session release, and re-establishment—we measure the latency of user-plane reattachment. Our results reveal high variability in recovery times, ranging from sub-second to over 50 seconds. These inconsistencies are attributed to limitations in Open5GS’s session handling, the lack of true migration support, and hardware limitations of the used machine. Despite these challenges, the study offers insights into the practical behavior of PFCP-driven anchor reconfiguration and the operational gaps that remain in open-source 5G core implementations.

Optical flow estimation with event cameras encompasses two primary algorithm classes: model-based and learning-based methods. Model-based approaches, do not require any training data while learning-based approaches utilize datasets of events to train neural networks. To effectively apply these algorithms, it's essential to understand their respective strengths and weaknesses.
This study compares model-based and learning-based optical flow estimation methods using event cameras, aiming to provide guidance for real-world applications. We evaluated these methods on the MVSEC and DSEC datasets, focusing on their accuracy and runtime. Our findings indicate that model-based methods excel on the MVSEC dataset, characterized by small motions, while learning-based approaches perform better on the more dynamic DSEC dataset. To investigate potential overfitting of learning-based methods to DSEC, we retrained the IDNet and TMA models on the BlinkFlow dataset. The retrained models demonstrated competitive accuracy, surpassing model-based methods which indicates that learning-based models perform better on datasets like DSEC even when not able to overfit. Finally, our analysis on runtime showed that model-based methods achieve real-time performance on CPUs and learning-based methods require a GPU to run in real-time.

Event cameras are bio-inspired sensors with high dynamic range, high temporal resolution, and low power consumption. These features enable precise motion detection even in challenging lighting conditions and fast-changing scenes, rendering them well-suited for optical flow estimation. However, event camera output is sparse and unstructured, making it challenging to process. Transformer architectures have shown to be effective in capturing long-term temporal dependencies and processing sparse input, hence they might be better suited to processing this output by leveraging the fine time granularity inherent to event camera data.
We introduce E-GMFlow, an approach for event-based optical flow inspired by the recent success in terms of accuracy of transformer-based models for frame-based optical flow. We explore the effect of temporal details on the accuracy of this transformer architecture by changing the number of temporal bins in which events are discretized. We observe that the increase in the number of temporal bins generally causes higher accuracy and comment on the limitations of this study.

Computer vision tasks have shown to benefit greatly from both developments in deep learning networks, and the emergence of event cameras. Deep networks can require a large amount of training data, which is not readily available for event cameras, specifically for optical flow estimation. The need for simulating this data in a realistic, physics-driven manner is therefore crucial. This paper compares the state of the art event camera simulators on different criteria, including event timestamp modeling, performance under low illumination, bandwidth simulation, computation speed and various types of noise simulation. We also summarize the shortcomings of some commonly used optical flow event datasets. For generating high-quality, realistic events, The V2E and DVS-Voltmeter simulators have shown to produce the most accurate data.

Outlier detection is an essential part of modern systems. It is used to detect anomalies in behaviour or performance of systems or subjects, such as fall detection in smartwatches or voltage irregularity detection in batteries. This provides early indications of something of potential problems.

A part of outlier detection that is not often analysed is the performance of algorithms in environments with data from only one subject, versus environments with data from multiple subjects. This paper aims to answer the questions regarding the performance of Gaussian Mixture Models (GMM) and DBSCAN in these different environments. This paper focuses on time series data collected from consumer-grade wearables like smartwatches. In this paper, the outliers are defined manually, as the used data set did not contain predefined outliers. This research considers both outliers defined within the subject data, and the use of other subjects as outliers.

Results from this paper indicate that the amount of subjects in the environment is not the sole factor in the performance of these algorithms. Rather, it is a combination of the amount of subjects in the environment and the type of outlier to be detected. Results show that a GMM has difficulty distinguishing subjects that are similar when using another subject as outlier data. On average, DBSCAN outperforms a GMM in almost all cases, and DBSCAN is a lot more consistent in its performance than a GMM.

Advancements in the precision and accuracy of consumer-grade wearables, such as a Fitbit, have enabled the identification and therefore authentication of individuals based on their emitted heart frequencies using these wrist-worn devices. With this type of authentication, a password is essentially sent out every second. This makes it a perfect form of authentication in fields where constant authentication is crucial. However, not much is known about how different types of cardiac diagnosis (e.g. fit or obese) influence the accuracy of this type of authentication. In this paper, it will be shown how different types of cardiac diagnosis affect the accuracy of Deep Neural Networks to identify individuals by heart rate. This study is done with data obtained from 14 subjects, having different types of cardiac diagnosis. A deep neural network consisting of multiple convolutional layers is being used to conduct the experiments. It has been shown that subjects with a paroxysmal atrial fibrillation diagnosis improve the accuracy the most, compared to the reference (normal healthy) subjects. On the other hand, (very) fit subjects decrease the accuracy the most. Heart failure and obese subjects have a similar accuracy compared to reference subjects.

Heart rate data and other data collected by consumer-grade wearable devices can give away quite useful information about the user. It can for example be used by machine learning algorithms such as Deep Neural Networks (DNN) to learn patterns about cardiovascular disease and fitness, or be used for identification. Heart rate patterns can also change quickly within the span of several months, which could make older heart rate data less useful when training a DNN. This paper shows that the DNN did indeed perform significantly worse when trying to identify people on older data compared to recent data. The accuracy calculated from the test set was 63.64% when trained on the most recently available training data, in comparison to 33.88% when trained on the least recent data which was more than 200 days older. When changing the recentness of training data only for a single user, there was also always an improvement in the accuracy of the model to identify that particular person. The accuracy to identify all users however did not necessarily increase, and sometimes even decreased. Using more data for training still outperforms using a smaller amount of samples of more recent data by slight margins, showing the trade-off between the recentness of data and the amount of data used for training. However, if fast training times are required, taking the most recent data windows can still lead to a similar performance as when training on all available data.

The aim of this paper is to complete the gap in the knowledge and experiment using as little as only the heart rate of some subjects to manage to successfully authorise them in some supposed system. The focus will be on the Gaussian Mixture model and the One Class Support Vector Machine, both outlier detectors, because most of the past research was focused on supervised models. Using these two, this paper will experiment with recognising intruders with models trained to distinguish one authorised person and multiple authorised persons. In the first case, multiple data processing methods and hyperparameters will be tested together and compared. In the second case, the goal will be to use and modify the best-found parameters of the first case to train models that are able to detect multiple persons as authorised. This time, there are two methods that this paper is going to look into and compare their performance: training one single model to detect all the subjects and training multiple models, one per subject. The most notable results are with one authorised person, with a score of 0.936, with two authorised, 0.88, and with 12 authorised, 0.713, when using the area under the curve metric.

In recent years, with the rapid expansion of IoT (Internet of Things) devices, more and more research and commercial projects have focused on various application areas of IoT. Signify, as a leading player in the smart home industry, has been deeply involved in this field for many years, particularly focusing on smart lighting for smart homes and providing consumers with a whole-house smart lighting solution (Hue System). The Hue System consists of several IoT devices, such as the Hue Bridge, Hue Bulbs, and Hue Accessories. This research paper specifically targets the IoT field and aims to reduce the firmware update cycle of the Hue Bridge.

The Hue Bridge, which serves as a central device in the Philips Hue internet-connected lighting system, connects users with other Hue devices. However, the Hue Bridge faces the challenge of inefficient firmware updates, which require validation engineers to wait for 6-8 weeks to ensure firmware reliability. To address this issue, this paper proposes a virtual system solution that improves the virtualization procedure of the Hue Bridge devices and utilizes Kubernetes for large-scale deployment to accelerate the generation of diagnostic data. Furthermore, a Use Case Model is established based on users' daily data, and a model based on Frequent Pattern Mining is applied to simulate users' daily behaviors in the Kubernetes Deployment.

To validate our virtual system, we designed validation experiments from multiple perspectives, including validation of the use case model and automated feedback. Our validation results demonstrate that this system enables more efficient and convenient acquisition of automated feedback (issues/bugs), while significantly enhancing the generation of diagnostic data in the firmware update cycle. Moreover, it offers advantages such as high availability, convenience, and cost-effectiveness in deployment. This research provides valuable references and insights for firmware update-related studies in the IoT domain

Badnets are a type of backdoor attack that aims at manipulating the behavior of Convolutional Neural Networks. The training is modified such that when certain triggers appear in the inputs the CNN is going to behave accordingly. In this paper, we apply this type of backdoor attack to a regression task on gaze estimation. We examine different triggers to discover which of them lead to better performance and thus infer which trigger aspects one can take the most advantage from. It turns out that placing frames around and drawing multiple lines across the images are the most effective for the training of Badnets.

Recent years have seen an increasing interest in stablecoins from major corporate and governmental parties. The European Central Bank is investigating the possibility of introducing its own Central Bank Digital Currency. The desired features of such a currency are under discussion. One such feature is offline spending: the ability to use the currency without an internet connection, like cash. This thesis describes a token-based transaction prototype and its implementation on the Kotlin-IPv8 protocol stack. The prototype allows funds to be spent in an offline setting and provides retroactive fraud detection. The prototype is not intended for deployment but instead serves as a trial for building digital currencies on Kotlin-IPv8. The included performance analysis demonstrates that various facets of Kotlin-IPv8 perform suboptimally, of which most notably its UDP data throughput.

Model extraction attacks are attacks which generate a substitute model of a targeted victim neural network. It is possible to perform these attacks without a preexisting dataset, but doing so requires a very high number of queries to be sent to the victim model. This is otfen in the realm of several million queries. The more difficult the dataset, the more queries required to gain an accurate substitute model. Through each state-of-the-art model extraction algorithm, one thing that is not thoroughly optimised are the hyperparameters of the models, and optimizing them has been found to have a strong impact on accuracy of the substitute model. To attempt to reduce the number of queries required, research has been done to find the effects of optimizing hyperparameters for both MNIST and fashionMNIST datasets. This is done through grid search and random search. The results show that proper hyperparameter tuning can reduce the number of queries required to perform model stealing if they are not already optimized. Examples include requiring 125000 + queries to achieve 95% accuracy for the MNIST dataset with some hyperparameter combinations to only requiring 15000.

Adversarial training and its variants have become the standard defense against adversarial attacks - perturbed inputs designed to fool the model. Boosting techniques such as Adaboost have been successful for binary classification problems, however, there is limited research in the application of them for providing adversarial robustness. In this work, we explore the question: How can AdaBoost ensemble learning provide adversarial robustness to white-box attacks when the "weak" learners are neural networks that do adversarial training? We design an extension of AdaBoost to support adversarial training in a multiclass setting, and name it Adven. To answer the question, we systematically study the effect of six variables of Adven’s training procedure on adversarial robustness. From a theoretical standpoint, our experiments show that known characteristics from adversarial training and ensemble learning apply in the combined context. Empirically, we demonstrate that an Adven ensemble is more robust than a single learner in every scenario. Using the best found values of the six tested variables, we derive an Adven ensemble that can defend against 91.88% of PGD attacks and obtain 96.72% accuracy on the MNIST dataset.

Machine Learning (ML) models are vulnerable to adversarial samples — human imperceptible changes to regular input to elicit wrong output on a given model. Plenty of adversarial attacks assume an attacker has access to the underlying model or access to the data used to train the model. Instead, in this paper we focus on the effects the data distributions has on the transferability of adversarial samples under a ``black-box'' scenario. We assume an attacker has to train a separate model (the ``substitute model'') and generate adversaries using this independent model. The substitute models are trained with different data distributions: symmetric, cross-section or completely disjoint data to the one used to train the target model. The results demonstrate that an attacker only needs semantically similar data to execute an effective attack using a substitute model and well-known gradient based adversarial generation techniques. Under ideal attack scenarios, target model accuracies can drop below 50\%. Furthermore, our research shows that generating adversarial images from an ensemble increases average attack success.

A machine learning classifier can be tricked us- ing adversarial attacks, attacks that alter images slightly to make the target model misclassify the image. To create adversarial attacks on black-box classifiers, a substitute model can be created us- ing model stealing. The research question this re- port address is the topic of using model stealing while minimizing the amount of querying the sub- stitute model needs to train. The solution used in this report is a variant of the ActiveThief algo- rithm that makes use of active learning to deter- mine which data is being queried. The paper exper- iments with different subset selection strategies to find the most informative data points. Also, a seed- ing algorithm based on clustering is explored and finally, a stopping criterion for the ActiveThief al- gorithm is proposed. These variations are evaluated on their accuracy and the number of queries they take to achieve that accuracy. This paper shows cluster seeding is an alternative to random seeding in ActiveThief. This paper also presents different subset selection strategies that outperform the ran- dom sampling strategy. Finally, a stopping criterion based on entropy is introduced that halts the algo- rithm when an uncertainty threshold is reached.