The present work deals in general with questions concerning the localisation of persons in underground structures, such as tunnels. Current projects of the so-called NIKE research group, which deal with similar topics to increase safety in underground and complex facilities, as well as other relevant investigations were accompanied during underground tests at the ZaB (Zentrum am Berg). In addition, current products for the detection and positioning of persons in underground facilities from various companies are being investigated. From the observed limitations of existing research projects, the idea emerged to develop a system that can detect and position persons without having to equip them with a specific device beforehand. Since the worldwide percentage of smartphone users is constantly increasing and in some countries such as Austria and Germany is close to 90%, a system that is able to detect these devices could therefore provide a good estimate of the number of people actually present in a certain environment. Since WLAN capable devices, such as smartphones, continuously send out so-called probe requests to maintain a stable network, on average about once a minute, various possibilities were investigated for basing a system for detecting people on them. The first implementations were then carried out using a small computer, a so-called Raspberry Pi, as a signal receiver and open-source software installed on it. This was followed by tests on the range and signal attenuation due to different environmental situations in the facility of the Tunnel Research Centre at ZaB. A major experiment by the NIKE research group then took place in the same centre, in which the project of this master thesis was able to participate along with other research undertakings. Various scenarios were carried out over several days in which units of the mine rescue and fire brigades, as well as military troops, were involved in possible training and fire simulations. The knowledge gained, especially from the joint execution of the numerous research projects for the first time, was also presented at a conference afterwards. The participation was particularly important and instructive for the project of this work, since the framework conditions of the operational capability could be determined as a result of the successful executions. Further ideas and suggestions for improvement, as well as alternative applications, were discussed in conclusion, always with the aim of supporting emergency services with additional information in the event of an incident.