Data measurement was done using Google Pixel 2 smartphone and Google nest wifi router,
To classify the indoor scenario based on wifi RSS and RTT values which is useful for accuracy improvement of wifi indoor positioning
Measurements were taken on 4 different indoor locations scenarios:
- LOS- No obstacle between smartphone and router
- Glass - A glass door obstacle between smartphone and router
- Metal - A metal door obstacle between smartphone and router
- Wall - A wall obstacle between smartphone and router
In every scenario 150 samples were taken at each meter from 1 m to 14 m. L
ine of Sight Count : 3243 Glass Count : 2852 Metal Count : 3017 Wall Count : 3064
To make data set ready for building LOS and NLOS scenario classification model, samples are randomized to prevent overfitting of a model to particular places.
To import data set data into Python environment, dataset.py script can be used. Unneccessary columns are removed.
import dataset
# import raw data
data = dataset.import_from_files()
print(data)
Pytorch Version : 2.0.1+cpu Python Version : 3.10.10