Engine knock detection based on wavelet transform

Engine knock is a major cause of pollution, engine parts damage and metallic noise in an engine. But somewhat knock in an engine can improve the power and economic performance of it. Therefore, Existence and intensity of the knock are preconditions for the automatic knock control system to work. The wavelet transform is an effective and promising signal processing method for shock feature detection, especially for light shock feature detection. In this work, the knock phenomenon is measured from a gasoline engine by a pressure sensor mounted in a spark plug and an accelerometer mounted on the engine block head. The discrete wavelet transform(DWT) is applied to detect the knock feature from the measured pressure signals. It is shown that there is a significant improvement on the signal to noise ratio (SNR) by using this approach, especially for light knock.