The development of connected devices and their daily use are today at the origin of the omnipresence of Wi-Fi wireless networks. However, these Wi-Fi networks are often vulnerable, and can be used by malicious people to disturb services, intercept sensitive data or to gain access to system. In railways, trains are now equipped with wireless communication systems for operational purposes or for passenger services. In both cases, defense strategies have to be developed to prevent misuses of the networks. The first objective of this study is to propose a monitoring solution, which is independent of the communication networks, to detect the occurrence of attacks. The second objective is to develop a method able to classify attacks of different types: the intentional electromagnetic interference (IEMI), i.e., jamming attacks, and the protocol-based attacks. This study focuses on the IEEE 802.11n Wi-Fi protocol. To perform these analyses, we propose to monitor and to analyze electromagnetic (EM) signals received by a monitoring antenna and a receiver collecting EM spectra. After that, we build a classification protocol following two steps: the first consists in the construction of a Support Vector Machine (SVM) classification model using the collected spectra and the second step uses this SVM model to predict the class of the attack (if any). A time-based correction of this prediction using the nearest neighbors is also included in this second step.