Explore CTU's Digital Library

Access research publications, theses and academic works

Faculty of Civil Engineering Faculty of Mechanical Engineering Faculty of Electrical Engineering Faculty of Nuclear Sciences and Physical Engineering Faculty of Architecture Faculty of Transportation Sciences Faculty of Biomedical Engineering Faculty of Information Technology Klokner Institute Masaryk Institute of Advanced Studies Central Library Acta Polytechnica Acta Polytechnica CTU Proceedings CTU Publishing Activity

Recent Submissions

Encrypted Traffic Classification with Deep Learning
Klasifikace šifrovaného provozu pomocí hlubokého učení
(Czech Technical University in Prague) Luxemburk, Jan; Kubátová, Hana; Pescapè, Antonio; Crispo, Bruno; Drašar, Martin
Modern security protocols, such as TLS 1.3, QUIC, Encrypted Client Hello, and encrypted DNS, progressively reduce the amount of information exposed to on-path observers, rendering network monitoring approaches that rely on payload inspection, plaintext handshake fields, or domain-level identifiers increasingly ineffective. While these developments significantly improve user privacy and communication confidentiality, they also pose a growing challenge for network operators responsible for network management and security. This thesis addresses this challenge by developing traffic classification methods that rely exclusively on observable side-channel information, namely packet sizes, directions, and inter-packet times. The research was conducted using backbone network traffic data obtained in cooperation with CESNET, the operator of the Czech national research and education network. The thesis thus combines realistic data, rigorous evaluation practices, and modern deep learning methods to address the problem of encrypted traffic classification.
Threat Detection in Network Traffic using Time Series Analysis
Detekce hrozeb v síťovém provozu pomocí analýzy časových řad
(Czech Technical University in Prague) Koumar, Josef; Čejka, Tomáš; Magán Carrión, Roberto; Sadre, Ramin; Varga, Pál
This dissertation thesis deals with the advancement of network threat detection methodologies through the application of systematic time series analysis. As modern network security faces a critical visibility gap due to the near-universal adoption of encryption, traditional payload-based inspection has become largely ineffective. This research addresses this challenge by defining a taxonomy for network-derived time series to model behavioral and temporal dependencies inherent in encrypted traffic. By treating network metadata as temporal signals, the dissertation thesis develops novel detection mechanisms for identifying malicious activity, ranging from botnet Command and Control (C&C) channels to sophisticated data exfiltration and Distributed Denial of Service (DDoS) attacks, without requiring access to unencrypted payloads.
Investigation of Soil Improvement Techniques Applied on Flood Protection Systems (with Specification on Pakistan)
Výzkum technik zlepšování půdy uplatňovaných v systémech protipovodňové ochrany (se zaměřením na Pákistán)
(Czech Technical University in Prague) Gul, Hasan; Pruška, Jan; Beňo, Jaroslav; Vojtasík, Karel
This research responds to the growing threat of climate-induced inundation by focusing on the geotechnical integrity and long-term resilience of earthen dikes constructed primarily from locally sourced materials. It highlights the insufficiency of the prevailing risk-based approach, in a scenario where flood recurrence has shortened significantly, demanding a more robust and adaptive flood protection strategy. This research introduces a novel geotechnical evaluation of construction and demolition waste (CDW) as a sustainable material for dike construction in Pakistan, a topic not previously addressed in scientific literature. It uniquely integrates historical flood data, field investigations, laboratory testing, and numerical modeling to assess dike performance under climate-induced hydrological extremes, including rapid drawdown and steady-state seepage scenarios. The results showed that the local soil is predominantly composed of poorly graded sandy soil, silty sand, or clayey sand with silty sand. These soil types are inadequate for dike construction due to their poor gradation and compaction characteristics. The laboratory testing program included a range of tests to evaluate the geotechnical properties of the local soil. The CDW particles are typically coarse (sand to gravel size), which introduces larger particles into the matrix. This leads to a broader particle size range and improved uniformity, with coarser CDW particles increasing the sand and gravel fractions. The mineralogical composition of the soil and CDW was evaluated using X-ray diffraction (XRD) analysis. The results showed that the soil is primarily composed of quartz, while the CDW contains a mixture of minerals, including quartz, calcite, and feldspar. The microstructural analysis using scanning electron microscopy (SEM) showed that the addition of CDW leads to a more open particle arrangement, which improves drainage characteristics. This doctoral thesis concludes that the incorporation of CDW up to 30% can enhance the stability of dikes under various hydrological scenarios. The findings of this research contribute to the development of sustainable and resilient infrastructure planning for flood protection in Pakistan. By utilizing CDW as a sustainable alternative to conventional materials, the study promotes a balanced approach to sustainable dike construction.
Tuning magnitude and direction of lattice thermal conductivity in transition metal dichalcogenide heterobilayers
(IOP Publishing Ltd, 2026) Perviz E.; Cammarata A.
We investigate the nanoscale mechanisms determining in-plane lattice thermal conductivity (LTC) of pristine and W-doped MX2–M′X′2 transition metal dichalcogenide heterobilayers from first principles, using the exact solution of the linearised Boltzmann transport equation in both phonon and relaxon bases. For the W-doped systems, we evaluate the anharmonic force con stants using machine-learning potentials trained on first-principles data. Pristine heterobilay ers exhibit isotropic LTC with preserved ordering across temperature. Under nearest-neighbour chalcogen substitution the LTC follows the ordering Te < Se < S; relaxon analysis further links the LTC and viscosity to harmonic phonon descriptors including phonon mode layer localisation and cophonicity. A sufficiently large mass contrast is required to induce layer localisation of the transport-relevant phonon modes, leading to enhanced LTC. The viscosity, which correlates positively with LTC, is associated with the low-cost cophonicity descriptor. Doped heterobilayers exhibit reduced and anisotropic LTC across all temperatures, while the aforementioned descriptors no longer correlate with the LTC. We find that the direction of maximum LTC can in principle be modulated by composition and temperature, although further work is required to establish robustness of the LTC magnitude and anisotropy trends. Thanks to its general formulation the analysis protocol can be readily extended to other van der Waals heterostructures, and the descriptors may be implemented in high-throughput engines to identify promising layered materials with tailored thermal transport characteristics.
Environmental qualification of a high-temperature superconducting magnet and flux pump system for spaceflight
(IOP Publishing, 2026) Pavri B.; Pollock R.; Long N.; Goddard-Winchester M.; Malich M.; Filgas R.
The Heki technology demonstration payload was launched to the International Space Station (ISS) in September 2025. Heki is designed to advance the technology readiness level of two novel technologies: a high-temperature superconducting magnet and a 'flux pump'-an inductive superconducting rectifier which energizes the magnet to field. This paper describes the rigorous pre-launch test and analysis program developed to meet ISS safety requirements and to demonstrate that these new components will survive the environment of launch and operations in space. The tests included structural analyses and random vibration testing to emulate launch conditions, thermal vacuum testing to simulate operations in space, and electromagnetic compatibility testing to verify Heki can be operated without interfering with neighbouring experiments and electronics. Heki successfully completed this pre-launch test program without degradation in performance, reducing the risk of failure in flight and marking an important milestone toward commercialization of these technologies for space applications.