Experiment details
Authors : Quentin Vey, Réjane Dalcé, Adrien van den Bossche, Thierry Val.
This document presents the dataset 20220611-uwb-twr.json used in the experiments presented in the proposal for demonstration submitted to LCN 2022: Quentin Vey, Réjane Dalcé, Adrien van den Bossche, Thierry Val. Indoor UWB localisation: LocURa4IoT testbed and dataset presentation.
The experiment has taken place on June, 11th 2022 on the LocURa4IoT testbed. Dataset is under ODbL licence. Please cite the following paper on any publication made with this dataset : Quentin Vey, Réjane Dalcé, Adrien van Den Bossche., Thierry Val. Indoor UWB localisation: LocURa4IoT testbed and dataset presentation. 47th IEEE Conference on Local Computer Networks (LCN 2022), IEEE, Sep 2022, Edmonton, Canada. hal-03702352
Scenario description
Please refer to the paper for the full experiment description.
Dataset description
Each line of the dataset is the output of a ranging as a JSON dictionary. These JSON dictionaries are detailed in the following table:
| key | description | example |
|---|---|---|
| initiator | TWR client (mobile) address | “183” |
| target | TWR server (anchor) address | “182” |
| protocol | Ranging protocol used for the experiment | “TWR” |
| t1 | First message transmission timestamp using client clock | 827869768332 |
| t2 | First message reception timestamp using server clock | 350550466754 |
| t3 | Second message transmission timestamp using server clock | 350610701964 |
| t4 | Second message reception timestamp using client clock | 827930006159 |
| skew | Client-server clock skew (=skewAck) | -5.968155 |
| nlosIndicator | Non Line of Sight Indicator (dbm) * | 5.014733 |
| tof | Time of Flight, expressed as 64GHz ticks (15,65ps) | 1356 |
| tofSkew | Time of Flight, expressed as 64GHz ticks (15,65ps), after clock skew correction on second message (ACK) | 1536.961803 |
| range | Final distance estimation (=rangeSkew) | 7.169614 |
| rangeSkew | Ranging after clock skew correction on second message | 7.208881 |
| rangeNoSkew | Raw ranging (no skew correction) | 6.360108 |
| ranging unit | Distance unit for ranging values | “m” |
| seqnum | Ranging Sequence Number | 1924 |
| rssiRequest | Received signal strength indicator of first message (dbm) * | -77.901009 |
| rssiAck | Received signal strength indicator of second message (dbm) * | -77.939378 |
| rssiData | Received signal strength indicator of third message (dbm) * | -78.213673 |
| temperature | Client-side Temperature (°C) | 34.400002 |
| distantTemperature | Server-side Temperature (°C) | 40.099998 |
| registerDump_CIR | Data from Channel Impulse response (all fields from this table refer to the second message) | |
| registerDump_CIR.ACC_mem | Raw data from Channel Impulse Response accumulator registry ACC_mem as a base64-encoded string (excluding some data at the beginning, see ACC_MEM_first_index) * | |
| registerDump_CIR.ACC_MEM_first_index | Index of the first sample of registerDump_CIR.ACC_mem in register ACC_mem | 675 |
| registerDump_CIR.RXPACC | Preamble Accumulation Count | 118 |
| registerDump_CIR.RXPACC_NOSAT | Unsaturated Preamble Accumulation Count | 118 |
| registerDump_CIR.SFD_LENGTH | length of the SFD sequence | 16 |
| registerDump_CIR.RXM110K | Receiver Mode 110 kbps data rate (1: true / 0: false) | 0 |
| registerDump_CIR.DWSFD | Use of non-standard SFD sequence (1: true / 0: false) | 0 |
| registerDump_CIR.RNSSFD | Use of non-standard SFD sequence (1: true / 0: false) | 0 |
| registerDump_CIR.FP_INDEX | First path index. 16-bit value reporting the position within the CIR data determined by the hardware to be the first path. The 10 most significant bits the integer portion of the position and the 6 least significant bits represent the fractional part. | 47914 |
| timestamp | unix timestamp | -0.157, 6.896, 2.65 |
| localisation.initiator | Client position x, y, z (m) | -0.157, 6.896, 2.65 |
| localisation.target | Server position x, y, z (m) | 4.358, 1.331, 2.623 |
| distance | Client-server true distance (m) | 7.166 |
| rangingError | Ranging error (=range-distance, m) | 0.003 |
*: see DecaWave « DW1000 User Manual ».
Example use
Refer to the file draw.py as an example to use the dataset. This script will generate plots of some key metrics, including the figures used in the proposal for demonstration.
usage :
python3 draw.py 20220611-uwb-twr.json
Python module dependencies: matplotlib, numpy