Abstract: Vehicular Ad hoc Network (VANET) is a promising technology for future smart vehicle systems and an essential component of Intelligent Transportation Systems (ITSs). ITS includes a wide range of modern technologies such as Global Positioning System (GPS), digital maps, video cameras, sensing devices and the wireless communication devices. It provides necessary information about the condition of the roads in time for drivers and traffic management systems to improve traffic efficiency, reduce traffic congestion, waiting times and fuel consumptions. In this study, we design an Intelligent Emergency System ( IES) for the VANETs using self-organizing maps and Elman neural network. The IES automatically controls the traffic lights according to the priority associated with the Cooperative Awareness Messages (CAMs) exchanged between emergency vehicles and RSUs. The CAMs sent from the emergency vehicles to RSUs be encrypted using a hash function to distinguish them from normal messages issued from other vehicles. The IESs use the features extracted from the trace file that describes the normal and urgent behavior in the VANETs. The type and the number of features have an important role in increasing the classification rate and decreasing false alarms in the proposed emergency system.