中文摘要：

提出了一种基于单个三维加速度传感器的人体行为的特征提取及识别方法，主要识别了站立、走、跑、上楼和下楼五种动作．该方法提取了多种统计特征包括标准差、偏度、峰度和相关系数实现多层分类．实验表明，本文采用的方法能够有效地识别这五种动作．

英文摘要：

In this paper, a method for activity recognition based on 3D acceleration sensor is introduced.We recognized five activity including standing, walking, rtmning,upstair and downstair with this method. Many statistical features are extracted such as standard deviation, skewness, kurtosis and correlation coefficient for classification. The result of experiments for testing the effectiveness of the proposed method is presented.