This paper presents a noise annoyance identification tool that was developed to assess the interior noise quality of passenger cars on an objective basis. The tool automatically extracts resonances, masking effects, order non-linearities, booming phenomena and amplitude modulations from a vehicle run-up sound and visualizes these features on the rpm-frequency spectrogram of the sound.
A unique order-based approach is used. In a first stage, an automatic order detection algorithm is employed to detect the significant engine orders in the run-up sound. An advanced order tracking technique is then used to accurately track these order components in amplitude and phase. Once this is achieved, the noise annoyance features are automatically extracted from the order data and displayed in the rpm-frequency colormap of the sound.
The noise annoyance identification tool is interesting and useful for two reasons: (1) it can largely reduce the vehicle evaluation time and efforts when compared with traditional subjective jury test methods; and (2) it informs the sound engineer which parts of the sound need to be modified in order to improve the interior sound quality.
