Aiming at problem of quantity of atmospheric monitoring sites in our country is limited and positions are discrete

collecting data can not represent air quality of the entire area

interpolation model is established based on RBF netural network for air quality monitoring

longitude

latitude and pollutant concentrations of the vicinity stations are designed as the input of net

then build corresponding relation between interpolating point and geographic coordinate and neighborhood points. Experimental results show that the method has high interpolation precision and provide an effective processing method of predicting data for the place with unknown

therefore

treatment of air polution and control provide theory basic.