Research on HTTPS Service Identification Method, Device, Storage Medium and Electronic Equipment

• Published in: 2024 4th Asia-Pacific Conference on Communications Technology and Computer Science (ACCTCS) pp. 369 - 374
• Main Authors: Meng, Fei, Chen, Jiake
• Format: Conference Proceeding
• Language: English
• Published: IEEE 24.02.2024
• Subjects: Accuracy; Artificial neural networks; Deep neural network; HTTP; HTTPS service; Inspection; Object recognition; Service identification; Symbols; Three-party inspection; Training
• DOI: 10.1109/ACCTCS61748.2024.00071

With the rapid development of mobile Internet, HTTPS service plays an increasingly important role in the network. Due to the encrypted nature of HTTPS services, traditional service identification methods based on port numbers and load characteristics make it challenging to identify HTTPS services effectively. To solve this problem, this paper proposes an HTTPS service identification method based on three-party inspection. It utilizes the consistency of location information and service types between user equipment, servers, and mobile network operators to build a data table to identify HTTPS services accurately. This paper designs an HTTPS service identification system based on three-party inspection, including data acquisition, data preprocessing, and service identification modules. In addition, the service recognition module applies a deep neural network model to achieve efficient recognition of HTTPS services by training and optimizing the parameters. We analyzed the results of the proposed method based on the experimental data. We compared the performance of many other ways using Precision, Recall, and F1 values to verify the validity and superiority of the proposed method in this paper. In conclusion, the HTTPS service identification method based on the three-party inspection can improve the accuracy of service identification and provide strong support for the management and optimization of mobile networks.