In this article, a technique based on the acoustic emission (AE) signal fractal and wavelet analysis are proposed for tool condition monitoring. it is difficult to obtain an effective result by these raw acoustic emission data. The local characterize of frequency band, which contains the main energy of AE signals, is depicted by the wavelet multi-resolution analysis, fractal dimension can describe the complexity of time series. It is found that the fault signal can effectively be extracted by wavelet transform and fractal dimension. Experimental results prove that this method is effectively.