Nowadays, air quality monitoring is identified as one of the key impacts in assessing the quality of life in urban areas. Traditional measuring procedures include expensive equipment in the fixed monitoring stations which is not suitable for urban areas because of the low spatio-temporal density of measurements. On the other hand, the technological development of small wearable sensor devices has created new opportunities for air pollution monitoring. Therefore, in this paper we discuss statistical approaches to fuse the data from fixed and mobile sensors for air quality monitoring.