|
|
Application-based Anomalous Communication Detection
Dostál, Michal ; Homoliak, Ivan (referee) ; Očenášek, Pavel (advisor)
This bachelor thesis deals with the analysis, design and implementation of a system for detecting anomalous network communication activities using high-level characteristics. The thesis contains a theoretical basis for the detection of anomalies using countries, autonomous systems and applications that are used to communicate. It also contains information about the techniques and methods of machine learning needed for implementation. The practical part describes the design, use and implementation of individual technologies. The result of this work is detection based on multiple machine learning methods, mostly classification.
|
|
|
Machine Learning from Intrusion Detection Systems
Dostál, Michal ; Očenášek, Pavel (referee) ; Hranický, Radek (advisor)
The current state of intrusion detection tools is insufficient because they often operate based on static rules and fail to leverage the potential of artificial intelligence. The aim of this work is to enhance the open-source tool Snort with the capability to detect malicious network traffic using machine learning. To achieve a robust classifier, useful features of network traffic were choosed, extracted from the output data of the Snort application. Subsequently, these traffic features were enriched and labeled with corresponding events. Experiments demonstrate excellent results not only in classification accuracy on test data but also in processing speed. The proposed approach and the conducted experiments indicate that this new method could exhibit promising performance even when dealing with real-world data.
|
|
|
Application-based Anomalous Communication Detection
Dostál, Michal ; Homoliak, Ivan (referee) ; Očenášek, Pavel (advisor)
This bachelor thesis deals with the analysis, design and implementation of a system for detecting anomalous network communication activities using high-level characteristics. The thesis contains a theoretical basis for the detection of anomalies using countries, autonomous systems and applications that are used to communicate. It also contains information about the techniques and methods of machine learning needed for implementation. The practical part describes the design, use and implementation of individual technologies. The result of this work is detection based on multiple machine learning methods, mostly classification.
|
|
|
Material properties of lithium fluoride for predicting XUV laser ablation rate and threshold fluence
Blejchař, T. ; Nevrlý, V. ; Vašinek, M. ; Dostál, Michal ; Pečínka, L. ; Dlabka, J. ; Stachoň, M. ; Juha, Libor ; Bitala, P. ; Zelinger, Zdeněk ; Pira, P. ; Wild, J.
This paper deals with prediction of extreme ultraviolet (XUV) laser ablation of lithium fluoride at nanosecond timescales. Material properties of lithium fluoride were determined based on bibliographic survey. These data are necessary for theoretical estimation of surface removal rate in relevance to XUV laser desorption/ablation process. Parameters of XUV radiation pulses generated by the Prague capillary-discharge laser (CDL) desktop system were assumed in this context. Prediction of ablation curve and threshold laser fluence for lithium fluoride was performed employing XUV-ABLATOR code. Quasi-random sampling approach was used for evaluating its predictive capabilities in the means of variance and stability of model outputs in expected range of uncertainties. These results were compared to experimental data observed previously.
|
|
|
Hydroxyl radical measurement in atmospheric pressure dimethyl ether-air laminar premixed flat flame using tunable diode laser absorption spectroscopy
Nevrlý, V. ; Dostál, Michal ; Bitala, P. ; Zelinger, Zdeněk ; Suchánek, Jan ; Válek, V. ; Klečka, V. ; Kubát, Pavel ; Engst, Pavel ; Vašinek, M. ; Wild, J.
Spectroscopic detection of hydroxyl (OH) radical and determination of its concentration in flames have an elusive history and considerable influence on combustion research. Electronic transitions in ultraviolet spectral region were extensively studied in this context and until recent time chemiluminescence or laser induced fluorescence of excited hydroxyl (OH*) radical is broadly used for absolute concentration and temperature measurement in flames.\nHowever, number densities of molecular species and population of relevant quantum levels in ground electronic state can be directly estimated from intensities of absorption lines observed by probing rovibrational transitions in infrared spectral region. Application of near-infrared tunable diode laser absorption spectroscopy (NIR-TDLAS) for the given purpose was demonstrated in an earlier work of Aizawa et al. [1]. Following his pioneering studies summarized in [2], we further explored feasibility of NIR-TDLAS (especially 2f-WMS technique) for monitoring minor species within combustion experiments particularly when dealing with dimethyl ether (DME) flames. Here we report our first results of NIR-TDLAS measurements focused on hydroxyl radical detection in laminar premixed flame burning DME-air mixture under fuel-lean conditions.
|
|
|
High resolution infrared spectroscopy as diagnostic tool for combustion and plasma chemistry
Zelinger, Zdeněk ; Nevrlý, Václav ; Grigorová, Eva ; Bitala, P. ; Dostál, Michal ; Suchánek, Jan ; Kubát, Pavel ; Engst, Pavel ; Ferus, Martin ; Kubelík, Petr ; Civiš, Svatopluk
Monitoring of transient species within combustion experiments (laminar flames, shock-tubes, flow reactors, etc.) is still relatively challenging task especially if application of non-invasive, i.e. optical detection methods is required. High resolution infrared spectroscopy is based on observation of the fine rotation structure that accompanies vibration transitions and thus provides direct information essential to characterization of both molecular structure and reaction dynamics. Thanks to its outstanding advantage enabling unambiguous assignment of specific molecular system according to its spectral feature, it can serve as a helpful tool for exploring complex reaction mechanisms as well as chemical reactivity of individual species present in laboratory flames or plasmas.\nPrevious studies gaining new insights into combustion and plasma chemistry as well as our recent advances targeted towards application of high resolution infrared spectroscopy for species concentration measurement in laminar flames are summarized here below.
|
|
|
Wavelength modulation spectroscopy for multicomponent ana-lytics of biomass burning tracers
Dostál, Michal ; Válek, V. ; Suchánek, Jan ; Kristlová, E. ; Roupcová, P. ; Zelinger, Zdeněk ; Nevrlý, Václav ; Bitala, P. ; Vašinek, M. ; Kubát, Pavel ; Ferus, Martin ; Civiš, Svatopluk
The potential of Tunable Diode Laser Absorption Spectroscopy (TDLAS) for monitoring of several species produced by biomass burning is in focus of this work. The infrared spectra of selected molecules (HCOOH, C2H2, CH3CN, N2O, CH3OH, CH3COCH3) are measured in laboratory conditions and the selectivity and of this method is demonstrated.
|
|
| |
guest ::
RSS feed.