HZAUIndoorLoc is a self-built indoor localization dataset collected at Huazhong Agricultural University (HZAU). The dataset is designed to facilitate research on indoor positioning, especially focused on cross-environment generalization and environment-invariant feature extraction.
The dataset includes Wi-Fi Received Signal Strength Indicator (RSSI) data collected across three different buildings:
- Building 0 (Symmetrical structure)
- Building 1 (Symmetrical structure)
- Building 2 (Independent structure)
Buildings 0 and 1 share symmetrical architectural layouts, which aids the learning of environment-invariant features. Building 2 serves as an independent validation scenario, helping evaluate model generalization to different architectural conditions.
| Attribute | Description |
|---|---|
| Buildings | Building 0, Building 1, Building 2 |
| Floors per Building | 2 floors per building |
| Coordinate Points | 37 coordinate points per floor |
| Wi-Fi Access Points | 259 unique WAPs across buildings |
| Training Records | 10,269 RSSI records |
| Validation Records | 2,568 RSSI records |
The dataset is organized as follows:
HZAUIndoorLoc/
├── Training set (10,269 records)
└── Validation set (2,568 records)
Each record contains RSSI values corresponding to the Wi-Fi Access Points (WAPs) at predefined coordinate locations within buildings and floors.
Dataset can be downloaded from the following link:
Buildings (0, 1, and 2) used for dataset collection at Huazhong Agricultural University.