Physical Layer Jamming detection: a Machine Learning Approach

This paper aims to illustrate the laboratory experience carried out during March-July 2023 at Hochschule Darmstadt having as its goal the writing of a master’s thesis. The initial goal of the project was to use machine learning techniques to analyze the physical characteristics (i.e:ISO/OSI layer 1) of a wireless cellular channel in order to detect the presence of an attacker. Thus, the expected outcome of the project is to construct a binary classifier, which takes in input information from the wireless channel and outputs the state of the channel through a binary classification: that is, whether the channel is in a state recognized as normal or whether it has been corrupted by the presence of an attacker. Lab experiences were carried out using software to implement SDR, both user-side and attacker- side. Therefore, the methodologies used to conduct these experiments will be explained, speci- fying the theoretical background and commenting from a technical point of view on the results obtained.