How to detect the weak signals buried in noises is a fundamental and important problem. Conventional wavelet denoising fails for signals with low signal-to-noise ratio (SNR). This paper discussed an approach which is based on the use of adaptive wavelet probabilistic neural network (AWPNN). The biorthogonal 9-7 wavelet is used to extract the features from original signal, and then the probabilistic neural network (PNN) is used to analyze the meaningful features and perform discrimination tasks. Simulations indicated that the AWPNN is suitable for increasing the SNR of weak signals which commonly have below 0 dB SNR, and our method can deal with the signals with fairly low (approximately -20 dB) SNR.