DYNAMAP, a European Life project, aims at developing a dynamical acoustic map of large urban areassuch as the city of Milan based on a limited number of noise monitoring stations. Generally, traffic noise, recorded as time series by field monitoring stations, is affected by the presence of Anomalous Noise Events (ANEs), which are extraneous to the regularvehicle noise. A dedicated algorithm integrated in the DYNAMAPsensors helps the identification and removal of ANEs from the time seriesin real time. Traffic noise data from the distributed monitoring stations, each one representative of a group of roads sharing similar characteristics (e.g. traffic flow), are used to build-up a dynamic noise map. An evaluation of the reliability of this procedure is performed by comparing the map prediction and in-field measurements. Preliminary results of this activity are presented.