Development of Helicopters Sound Detection Systems

The presence of a background noise in the helicopter´s sound signals can cause un-acceptance degradation in the detection of the performance method. The recent detection method uses artificial neural networks (ANNs), in combination with parametric spectral representation techniques to detect these targets. The feature encoding techniques based on linear prediction coefficients (LPC) have been applied to obtain spectral estimate of the acoustic signals. This paper investigates preprocessing of signal enhancement prior to feature extraction using different wavelet transform functions