AI Application in Road-noise Active Noise Control
RANC (Road-noise Active Noise Control)
As ICE engine vehicles are converted to electric motor vehicles, the proportion of road-noise is increased. Road-noise is a noise generated when a tire comes into contact with a road surface, mainly a low-frequency noise of 20 to 500 Hz. Road-noise Active Noise Control (RANC) is a representative way to reduce this. RANC is a method of reducing indoor noise by generating antiphase sound waves of road noise. For RANC to be effective, it is necessary to know in advance the sound transfer function between the speaker generating the antiphase sound wave and the passenger's ear. However, the sound transfer function varies depending on various conditions such as the location of the passenger in the vehicle and the presence or absence of a passenger, which lowers the performance of the RANC. Therefore, it is important to have a technology that can detect the variable environment in the vehicle in real-time and predict the sound transfer function that changes through this.
Prediction of Acoustic Transfer Function in Variable Environment for RANC using AI
AI technology is used to predict different acoustic transfer functions for variable conditions. A prediction model can be made by measuring the sound transfer function according to various variable conditions and learning through a neural network. Through this, the acoustic transfer function can be predicted with high accuracy and the performance of the RANC can be improved.