The knowledge of dependencies between services of an organization is vital for network management and security. However, manually keeping track of the dependencies is becoming too complicated in modern networks, and automatic methods based on network traffic monitoring are emerging. The goal of this thesis is to define service dependencies as complex events of network flows and propose a novel method of their discovery based on that definition.
The tasks of the thesis are the following:

• Familiarize with the concepts of passive network monitoring (IPFIX technology) and complex event processing (CEP).
• Study relevant literature and identify the types of dependencies suitable for CEP detection. These should include at least Local-Remote and Remote-Remote dependencies.
• Define selected types of dependencies as complex events.
• Implement a prototype of the detection method as a module of the CRUSOE situational awareness framework [1].

[1] Research of Tools for Cyber Situational Awareness and Decision Support of CSIRT Teams in Protection of Critical Infrastructure supported by the Security Research Programme of the Czech Republic 2015 - 2020 (BV III / 1 VS) granted by the Ministry of the Interior of the Czech Republic under No. VI20172020070.