基于信道状态特征的手势动作识别技术研究

手势识别作为人和机器之间重要的交互手段,在日常生活中具有广泛的应用场景。基于无线信号特别是WiFi的手势识别由于其无接触、成本低等优点成为当前热门的方式。为解决传统基于无线信号手势识别算法没有充分利用信号相位特征的缺点,本文提出利用WiFi信道状态信息幅值和相位结合的方式进行手势识别。通过子载波降维和动作曲线提取对接收的WiFi信号进行处理,并将信号的幅值和相位结合,利用机器学习算法对数据进行训练和分类,实现了单手向前、单手向后、单手向左、单手向右、单手向上和单手向下六种手势的识别。实验结果显示,本文算法在近距离和远距离下的精确度分别为96%和92%。

Gesture Recognition Based on Channel State Features

Gesture recognition is an important means of interaction between humans and machines,which has a wide range of application scenarios in daily life.Gesture recognition based on wireless signals has become a popular method due to its advantages such as contactless and low cost.In order to solve the shortcomings of the traditional wireless signal gesture recognition algorithm that does not make full use of the phase characteristics of the signal,this paper proposes to use the combination of WiFi channel state information amplitude and phase for gesture recognition.The method first processes the received WiFi signal by subcarrier dimensionality reduction and action extraction,and combines the signal amplitude and phase Combined as features.Machine learning algorithms are used to train and classify the data.The method achieves the recognition of six gestures:one hand forward,one hand backward,one hand left,one hand right,one hand up and one hand down.Experimental results show that the accuracy of the algorithm is 96%and 92%at short and long distances,respectively.