Artificial neural network for wireless field strength prediction under power lines in Taiwan railway

A new model based on artificial neural network for the wireless field strength prediction for the message communicated to trolleys and level-crossings under Power lines inside Taiwan Railway. The purpose used with wireless communication, linked with level-crossings, and uses the radio propagation to dispatch message from the wireless equipments to the trolleys along Taiwan Railway. In contrast to other prediction models, the new method shows a better prediction result and generalization, so the forecast results are very accurate in Taiwan Railway. We perform some practical measurements under power lines, and then use the neural network model to verify the measured data. This model is particularly suited for application in Taiwan Railway planning from Tainan to Fangliao separated by Western Line, Pingtung Line, and Kaohsiung city local line respectively. The consequences for static and dynamic measurement of this wireless system, concerned with the level-crossing and the trolley, is able to present an improved mission regulation in railway.