WiFi-based Indoor Localization Data
with Densely Deployed APs
Many indoor positioning algorithms have been proposed in the last decade, most of which are based on WiFi RSS fingerprints. However, the environment has changed dramatically since the original algorithms using only a few Access Points (APs). A typical building with densely deployed APs might contain hundreds of APs. The explosive growth of the number of APs introduces new challenges to these WiFi-based localization algorithms.
This project aims to explore how the AP-intensive environment influences localization accuracy. It presents an empirical study of WiFi fingerprint-based indoor localization algorithms in a real-world environment with hundreds of APs.
- Publication and Data
Xin Chen, Junjun Kong, Yao Guo,
Xiangqun Chen, "An Empirical Study of Indoor Localization Algorithms with Densely Deployed APs", GlobeCom
- Data Release:
Data.rar (Detailed information and
data format can be found in Readme.txt.)
- Experiment Environment
- The experiment environment was the corridor of the fourth floor of an office building in Peking University. The map is shown above, which is roughly 50 m by 50 m.
- In each round of the experiments, we walked along the corridor and took samples every 4 meters.
- At each position, we collected 60 fingerprints every 0.5 seconds for training and testing.
- The sampling configuration is listed above. We conducted five rounds of samplings within 14 months. The second to fifth samplings are, respectively, 3 days, 20 days, 2 months and 14 months after the initial sampling.
- We collected the data from three different smartphones in sampling #5.
- Data Characteristics