Automatically identifying apps in mobile traffic

• DOI: 10.1002/cpe.3703
• Publication Date: 2015-12-09T12:01:15Z
• Authors: Jianhua Sun; Lingjun She; Hao Chen; Wenyong Zhong; Cheng Chang; Zhiwen Chen; Wentao Li; Shuna Yao
• Keywords: Non-negative Matrix Factorization; Mobile apps; Visibility; Traffic Analysis; Traffic classification

Summary With the rapid development of smartphones in recent years, we have witnessed an exponential growth number mobile apps. Considering security and management issues, network operators need to a clear visibility into apps running network. To this end, paper presents novel approach generating fingerprints for from traffic. The that characterize unique behaviors specific can be used identify real In order handle large volume traffic efficiently, use non‐negative matrix factorization (NMF) perform analysis cluster similar groups. Then, access patterns individual are extracted each group as distinguishing others uniquely. experimental evaluations show proposed random mixed with high precision. Copyright © 2015 John Wiley & Sons, Ltd.