Blog
-
Forecasting for Network Anomaly Detection – Insights from NeCS Winter PhD School
A collaborative team of researchers from the Faculty of Information Technology at Czech Technical University in Prague (FIT ČVUT) and the Faculty of Information Technology at Brno University of Technology (FIT VUT) introduced a unique anomaly detection method for network traffic using forecasting at the interdisciplinary NeCS Winter PhD School. The team, consisting of Josef Koumar and Jaroslav Pešek from FIT CTU, along with Kamil Jeřábek, and Jiří Setinský from FIT BUT, conducted an innovation workshop focused on utilizing predictive models for detecting network anomalies in advance.
-
Detecting Changes in Encrypted Traffic with AI: Introducing MFWDD
In the rapidly evolving world of network traffic monitoring, machine learning (ML) has become a cornerstone for classification tasks. But as network protocols evolve and data patterns shift, maintaining the reliability of ML models is a constant challenge. Our research team at the Czech Technical University in Prague, together with CESNET, developed a solution to tackle these challenges head-on: Model-based Feature Weight Drift Detection (MFWDD).
-
Reflecting on CNSM 2024: A Milestone in Network and Service Management
The 20th International Conference on Network and Service Management (CNSM 2024), held from October 28 to 31, 2024, in Prague, Czech Republic, marked a significant milestone in the field of network and service management. As the Poster Session Chair and a member of the local organizing team, I had the privilege of contributing to this landmark event.
-
CESNET-TimeSeries24: A Novel Dataset for Network Traffic Analysis and Forecasting
The CESNET-TimeSeries24 dataset is a groundbreaking contribution to the field of network traffic analysis, anomaly detection, and forecasting. Created from the real-world CESNET3 ISP network, this dataset spans 40 weeks of data from over 275,000 active IP addresses, offering unparalleled diversity and depth for researchers and practitioners.
-
Cybersecurity Horizons 2024: Bridging Academia and Industry in Cybersecurity
On September 26, 2024, as a president of IEEE student branch for CTU I coorganize the Cybersecurity Horizons 2024 conference which took place aboard the steamboat Klára in Prague. The event aimed to connect academic experts with industrial partners in the field of cybersecurity. It was organized by the IEEE Student Section at CTU in collaboration with the Faculty of Information Technology (FIT CTU) and the Faculty of Electrical Engineering (FEL CTU), with support from the CTU Rectorate.
-
Cybersecurity Summer School: Advancing Cybersecurity Awareness and Skills
On June 15, 2024, as a president of IEEE student branch for CTU I coorganize the Cybersecurity Summer School event which took place at the Czech Technical University in Prague (CTU). This initiative aimed to enhance cybersecurity awareness and skills among students and professionals. The event was organized by the Center for Cybersecurity in collaboration with IEEE student branch for CTU.
-
AI for network traffic classification workshop NECS 2024
Author: Jaroslav Pešek
-
Revolutionizing Network Traffic Classification Through Single Flow Time Series Analysis
Network traffic analysis has always been critical for maintaining security and efficiency in modern networks. However, with the rise of encryption protocols and high-speed infrastructures, traditional approaches often fail to provide the necessary insights. My research introduces a novel methodology for classifying network traffic based on Single Flow Time Series (SFTS) analysis, achieving unprecedented accuracy and universality.
-
Harnessing Periodicity Detection for Network Traffic Classification
With the rise of encryption protocols and increasing reliance on secure communication, monitoring and classifying network traffic has become a daunting yet critical task. Despite these challenges, periodic behavior in network communications presents an exploitable feature for identifying and classifying traffic types, even in encrypted environments. Here, I discuss my novel method for detecting periodicity in network traffic and how it can be leveraged for accurate traffic classification.