CSIDA

The dataset is for Wi-Fi-based human activity recognition. The dataset is comprised of six experiments performed by 5 different subjects in five different locations in two different indoor environments. Each subject performed 20 trials for each of the experiments which makes the overall number of recorded trials in the dataset equals to 3000 trials (6 experiments × 5 subjects × 5 locations × 20 times). To record the data, we used the channel state information (CSI) tool to capture the exchanged Wi-Fi packets between a Wi-Fi transmitter and receiver. The utilized transmitter and receiver are retrofitted with the Intel 5300 network interface card which enabled us to capture the CSI values that are contained in the recorded transmissions. The related articles is "WiFi-based Cross-Domain Gesture Recognition via Modified Prototypical Networks" , whose url is https://github.com/Zhang-xie/WiGr. For storage space reasons, the data is uploaded in two parts. The first part includes the final data and its label, and the second part includes the raw data.

The dataset folder contained seven sub-directories, which were the original data, the data with random phase offset removed, the data with noise removed, the action gesture label, the environment label, the location label, and the person label. This data was stored in zarr data format and needed to be read using python's zarr library. Zarr saved data in blocks. Frankly speaking, it was to divide a block of data into sub-blocks of the same size. Each sub-block was saved into a file named *.. The advantage of such processing was that it was very friendly to large scale data.