Research progress of monitoring engineering ceramics grinding by acoustic emission (AE) is firstly described. It is found that the study of grinding diamond wheel wear monitoring is normally based on the analysis of the root mean square of acoustic emission signal, and the accuracy of acoustic emission monitoring is not high. In order to promote the accuracy of acoustic emission monitoring for wear states of grinding diamond wheel, support vector machine is applied to establish the classification model of grinding diamond wheel wear states. Through the analysis of acoustic emission signals, it is found that the strongest spectral energy of alumina ceramic acoustic emission signal under precision grinding is in 30~40 kHz band. The AE signal spectrum of the diamond grinding wheel wear states including mild wear, serious wear and after dressing are significantly different. The wavelet decomposition coefficient variances of the grinding acoustic emission signal can well reflect the wear states of the grinding diamond wheel. Therefore, the wavelet decomposition coefficient variances of grinding acoustic emission signal is applied as the input characteristics of support vector machine to identify grinding diamond wheel wear states, and the accuracy of classification test is 100%.