Computer Engineering and Applications ›› 2023, Vol. 59 ›› Issue (15): 274-280.DOI: 10.3778/j.issn.1002-8331.2205-0574

• Engineering and Applications • Previous Articles     Next Articles

Smart Home Recognition Based on WiFi Data Frame Features

LIU Dandan, HAN Yi, LIU Xiangyu, XIE Rongrong, WANG Jingxiang, DU Yanhui   

  1. 1.College of Information and Cyber Security, People’s Public Security University of China, Beijing 100038, China
    2.The First Research Institute of the Ministry of Public Security, Beijing 100048, China
  • Online:2023-08-01 Published:2023-08-01



  1. 1.中国人民公安大学 信息网络安全学院,北京 100038
    2.公安部第一研究所,北京 100048

Abstract: The identification of smart home device under WiFi environment is one of the current research hotspots of IoT security. Due to the encryption characteristics of the 802.11 protocol, the characteristics of traditional unencrypted traffic to identify IoT devices cannot be applied. Based on the analysis of 802.11 protocol, it is found that the data frames of diverse smart home devices have different characteristics. In this paper, protocol field such as frame length, frame arrival time, duration, and frame sequence number are selected as traffic characteristics. Through the improved decision tree CART algorithm realize the identification of smart home devices in the WiFi environment. The experimental results show that this method can effectively identify the device model, and the  accuracy of recognition is 91.3%.

Key words: smart home, WiFi data frame, traffic characteristics, device identification, machine learning

摘要: WiFi环境下的智能家居设备识别是当前物联网安全的研究热点之一。由于WiFi环境下802.11协议具有加密的特性,使得传统明文流量识别物联网设备的特征无法适用。分析802.11协议发现不同智能家居设备的数据帧存在差异性的特点,基于此开展研究工作,选取帧长度、帧到达时间、持续时间、帧序列号等字段作为流量特征,通过改进的决策树CART算法,实现WiFi环境下智能家居设备识别。实验结果表明该方法能有效识别设备型号,识别准确率为91.3%。

关键词: 智能家居, WiFi数据帧, 流量特征, 设备识别, 机器学习