论文部分内容阅读
针对目前行为识别通用模型对步行、上楼、下楼等易混淆行为识别准确率较低的情况,提出了一种基于小波分解的移动用户行为识别方法,从小波分解后不同频率子信号的低频近似系数中提取小波能量、小波峰个数和平均波峰幅值等特征,基于决策树分类器建立与用户无关的行为识别通用模型.分别用典型时域特征数据集和小波特征数据集对该通用模型进行验证.实验结果表明,采用新方法后,3种易混淆行为的平均识别准确率提高了14.82%,减少了误判.
Aiming at the low recognition accuracy of common confusion behaviors such as walking, going upstairs and going downstairs, a common behavioral recognition method based on wavelet decomposition is proposed to identify the behavior of mobile users based on wavelet decomposition. After wavelet decomposition, the low frequency Approximate coefficients were extracted from the wavelet energy, the number of wavelet peaks and the average peak amplitude characteristics, and based on the decision tree classifier to establish a universal model of behavior-independent identification with the user.Take the typical time domain feature dataset and wavelet feature dataset respectively, Model.Experimental results show that the average recognition accuracy rate of the three kinds of confusable behavior increases by 14.82% with the new method, reducing the misjudgment.