BioTag RFID Dataset

Description:

BioTag RFID Dataset for user verification. This dataset contains the physiological characteristics captured by RFID of 41 participants. The experiments are conducted in four types of indoor spaces, a twin-size bedroom (4.2m×4.3m), a typical lab with office furniture and (3.0m×5.0m), a corridor with no obstacle (2.8m×2.8m), and a home office with office furniture (7.0m×4.0m). The dataset can be used by fellow researchers to reproduce the original work or to further explore other machine-learning problems in the domain of RFID Signals.

Format: .h5py format

Section 1: Device Configuration

• A commodity RFID reader Impinj. The detailed information of it can be found at https://www.dipolerfid.com/product/Impinj-Speedway-revolution-R420
• A directional antenna Laird S9028PCL. The detailed information can be found at https://connectivity-staging.s3.us-east-2.amazonaws.com/s3fs-public/2018-10/ANT-DS-S9028PCL%20S9028PCR-0515.pdf
• Low-level reader protocol - LLRP.
• Adaptive sampling rate between 100Hz to 200Hz.

Section 2: Data Format

We provide raw data received by the RFID reader. The data files all are in the h5py format. The details are shown in the following:

1. 41 participants are included in this experiment.
2. The dataset file "Lab_P1_D1_T1.h5py", "Corridor_P2_D2_T2.h5py" represents:
   • Lab, Corridor: The scene to conduct this experiment.
   • P1, P2: Participant number.
   • D1, D2: The day to collect the data.
   • T1, T2: Trial number.

Section 3: Experimental Setups

• Environments:
  • There are two tags attached to the participants, one is attached to the chest and another is attached to the abdomen.
  • The experiments are conducted in four types of indoor spaces including a bedroom with a twin-size bed (4.2m×4.3m), a typical lab with office furniture and (3.0m×5.0m), a corridor with no obstacle (2.8m×2.8m), and a home office with office furniture (7.0m×4.0m).
  • We implement a user interface model on a Thinkpad laptop, which collects data (i.e, phase, and RSSI) from the RFID reader through the low-level reader protocol - LLRP. We only use Port 1 to receive the signal.
• Participants:
  • We conduct experiments with 41 participants (i.e.,33 males and 8 females, aged from 12 to 70) for 3 days. Participants are asked to sit on a chair only wearing a sleeve or a T-shirt and breathe normally during the experiments. The height of participants is in the range of (160cm, 185cm) and the weight is in the range of (50kg, 90kg).
• Distances:
  • A participant sits on a chair that is 1m in front of the antenna. The antenna is at the same height as the participant’s chest.

Section 4: Data Description

We separate our data into different folders based on different participants. Most participants have 30 data files that are collected in 3 days (10 files each day) and some participants have less data files that are collected in one day or two days. All data files are in .h5py format and have 20000 samples which contain 5 columns. The meaning of the 5 columns is as followed:

• Column 1: Tag

  Type: Int

  Unit: Num

  Description: There are two tags involved in the experiments. Tag 1 is attached to the chest, and Tag 2 is attached to the abdomen.

• Column 2: RSSI

  Type: Int

  Unit: dBm

  Description: This is the received signal strength indicator (RSSI) of received RFID signals obtained from Tag1 and Tag2. It means the power level of the tag’s backscattered response signal. It can be used to measure the distance between an RFID Tag and an RFID Reader, and identify the movement direction of an RFID tag.

• Column 3: Phase

  Type: Double

  Unit: Radians

  Range: 0~2π

  Description: Phase is a measure in degrees of how much the received signal leads the sent signal. It can be used to measure the distance between an RFID tag and an RFID reader with RSSI.

• Column 4: Time Stamp

  Type: Int

  Unit: seconds

  Description: This is the Unix time when receiving the signal from Tag1 and Tag2. It measures time by the number of seconds that have elapsed since 00:00:00 UTC on 1 January 1970. It can be used to determine when the reader receives the signal.

• Column 5: Frequency

  Type: Double

  Unit: MHz

  Range: 902.5 ~ 927.5 MHz

  Description: This is the frequency of the received RFID signal. The RFID system operates in ultra-high frequency (UHF) bands which hop among 50 channels within a spectrum from 902.5MHz to 927.5MHz. UHF RFID has a faster data transfer rate and the read range can be as long as 12m.

Section 5: Codes

• load_data.py: load a h5py file and extract all samples by the column name (i.e, Tag, and RSSI).

Section 6: Citations

If your paper is related to our works, please cite our papers as follows.

https://dl.acm.org/doi/abs/10.1145/3492866.3549718?casa_token=aw0ei4vR-xIAAAAA:rfIbhb-X0kw37Z4KtphfflG_PnFaqtdpcZ9bcT8lEWs-9hlEmDAMcTmtJshJNBnNWizu15qIjQVAJA

Bin Hu, Tianming Zhao, Yan Wang, Jerry Cheng, Richard Howard, Yingying Chen, and Hao Wan. 2022. BioTag: robust RFID-based continuous user verification using physiological features from respiration. In Proceedings of the Twenty-Third International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc '22). Association for Computing Machinery, New York, NY, USA, 191–200.

Bibtex:

@inproceedings{hu2022biotag,
  title={BioTag: Robust RFID-based continuous user verification using physiological features from respiration},
  author={Hu, Bin and Zhao, Tianming and Wang, Yan and Cheng, Jerry and Howard, Richard and Chen, Yingying and Wan, Hao},
  booktitle={Proceedings of the Twenty-Third International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing},
  pages={191--200},
  year={2022}
}

Contact us by email: zijietang0316@gmail.com if you have any questions.