An international standard stipulates the allowable limit of radio noise radiated from running trains to wayside areas and a standard method of measuring such noise. However, the intensity of radiated radio noise, the position of the radiation source and its characteristics could not be obtained until now unless actually measured. The RTRI therefore developed a basic analysis model to regard running trains as a radio noise source (oscillator) and the combination of wiring on vehicles, trolley wires, rails and substations as a radiation source (antenna) (see Fig. 1). Simulation of a running train with this analysis model reproduced a state where radio noise was received. In the frequency zone of 10 MHz or lower, periodically fluctuating radio noise was received, which was emitted from a train running at a distance. In the frequency zone of 10 MHz or higher, on the other hand, radio noise was received only right in front of the train (Fig. 2). Based on these results, the RTRI clarified that railway radio noise can be thought of as being emitted from a loop antenna whose size changes significantly in the direction of train movement. Application of this technique makes it possible to judge whether the addition of capacitors or reactors and changes in wiring contribute to the reduction of radio noise.