Abstract | ||
---|---|---|
With the popularity of wireless communication technology, insights into WiFi communication can be obtained conveniently by means of WiFi probe technology. These WiFi signals enable abundant information in physical space, and effective data service based on space or time domain for public. Knowledge discovery on WiFi data has recently become a new research hotspot. In this paper, we use a novel approach to solve a common problem, that is looking for coupled devices of one owner using WiFi probe data. Understanding a defined relation between two WiFi devices can make a data transmission link more efficient and secure. Besides, it can also provide more abundant date for user profile refinement, enabling an enhanced customized recommendation. In our method, we extract 10 features, including Received Signal Strength (RSS) variation similarity, time similarity and event similarity, to pair one user's device to the other. Feature parameters are derived from statistical analysis of positive samples and classification accuracy. Sufficient experiments verified that our proposed device pairing algorithm is robust, accurate and efficient. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/BIGCOM.2019.00053 | 2019 5th International Conference on Big Data Computing and Communications (BIGCOM) |
Keywords | Field | DocType |
WiFi signal, device pairing, RSS, event similarity, time similarity | Passive monitoring,Computer science,Artificial intelligence,Recognition algorithm,Machine learning | Conference |
ISBN | Citations | PageRank |
978-1-7281-4025-4 | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mingxing Jiang | 1 | 22 | 3.30 |
Zhongwen Guo | 2 | 4 | 3.85 |
Jinxin Wang | 3 | 0 | 0.34 |
Xi Wang | 4 | 0 | 1.01 |