Electric Vehicle Testing and Verification of Electromagnetic Noise Suppression by Signal Processing in AM Radio

It has been reported that electromagnetic noise emitted from electric vehicles (EVs) cause interference with AM radio reception. As a countermeasure, a method has been proposed in which signal processing is performed on the radio to remove noise superimposed on the signal. Compared to conventional noise removal methods that use sensors, this approach has the advantage of reducing the number of components, but it requires understanding the noise characteristics and considering methods to separate it from the audio signal. In this study, with the aim to understand the characteristics of noise, noise emitted when driving an EV was measured and observations were made on how the spectrum changed over time. The causes of the time-varying spectrum were analyzed and its behavior was explained using a theoretical formula for inverter noise. Based on this analysis, a signal processing method capable of handling noise whose spectrum changes over time, such as that superimposed from an EV while driving, was proposed. It was demonstrated that noise could be reduced by about 30 dB.