In this paper, a model for SD rat focal ischemic cerebral injury is presented. Based on this experiment model, we recorded the EEG of ischemic region and normal region from before ischemia to after ischemia 30 minutes. We analyzed the complexity of these EEG signals, and found that the EEG complexity was very sensitive to focal ischemic cerebral injury. The EEG complexity on ischemic region was less than that on normal region. The difference in EEG complexity between two regions increased with the decreasing velocity. When insult took place, the EEG complexity was less than that before ischemai. These resluts agree to those we obtained by Discrete Gabor Spectrogram analysis and Heat Shock Protein (HSP) expression. The ischemic and normal regions can be clearly distinguished by analyzing their EEG complexity, and the extent can also be detected in this way.