A Smart Automated Signature Extraction Scheme for Mobile Phone Number in Human-Centered Smart Home Systems

Human-centered smart devices profiling with Wi-Fi networks has received much attention from both research and industry, especially those network operators and security agencies who aim to enhance user experience and security of home network as well as free Wi-Fi services. One type of such profiling is the extraction of mobile phone numbers. In traditional cellular networks, such as 3G and 4G, mobile phone number extraction can be achieved from the analysis of the authentication signaling. However, this method cannot be used in the broadband network environment, e.g., Wi-Fi. Operators and security agencies of Wi-Fi networks often apply manual statistics, telephone inquiries or user input for information. Unfortunately, those traditional methods are inefficient in practice. Moreover, authenticity cannot be guaranteed with the traditional methods. In this paper, we propose a smart method for mobile phone number extraction in smart home networks and systems. In particular, the proposed method is based on deep packet inspection of home broadband traffic. To improve the efficiency and accuracy of detection, we further propose a smart automated signature extraction method of mobile phone numbers from home network traffic. Our proposed method can achieve 86.2% accuracy in the real-life human-centered smart home network test.