Bezpečnost v bezdrátových komunikačních sítích

Abstract

Eavesdropping on computer networks poses a number of risks to both the individual and the larger group. Risks include the theft of personal data, access data, tracking of users, etc. In particular, passive eavesdropping on computer networks poses a greater risk. This type of eavesdropping is very difficult to detect and is usually only demonstrated and detected when the attacker has access to user data, which he then misuses. The actual detection and proof of eavesdropping occurs only after damage has been done to someone. Motivated by the above facts, this dissertation aims to propose a novel method for passive eavesdropping detection in computer networks, focusing on Wi-Fi wireless networks, which are highly vulnerable to passive eavesdropping. The proposed method for passive eavesdropping detection can be applied to other types of networks. The proposed passive eavesdropping detection method is based on real user behavior in wireless networks using generated traffic that is inserted into the user's standard traffic and on an automatically generated server platform. During the work on this dissertation, a model of user behavior was developed to match as closely as possible the real user behavior in wireless networks. In order to detect passive eavesdropping correctly and uniquely, the server platform must be incorporated into the detection system. As part of the dissertation solution, an auto-generating server infrastructure was designed to best detect passive eavesdropping in wireless networks. All the proposed methods were experimentally verified.