GNSS Spectrum & Low-Cost Controlled Indoor Dataset

Spectrum & Low-Cost (LC) Laboratory Dataset for Interference Classification in GNSS Data

Introduction

We publish the Indoor Spectrum & LC Laboratory Dataset that was recorded in two different controlled environments: a small laboratory and the large-scale industrial Fraunhofer IIS L.I.N.K. test and application center. The dataset contains a variety of interference classes in GNSS signals from FIOT and low-cost (LC) hardware. For the dataset recording, we developed a hardware setup that captures short, wideband snapshots in both E1 and E6 GNSS bands. The setup records 20ms raw IQ snapshots triggered from the energy with a sample rate of 62.5MHz, an analog bandwidth of 50MHz and an 8bit bit-width.

Dataset

The folder GNSS_dataset contains the Indoor Spectrum & LC Laboratory Dataset with the corresponding labels in .txt files for the train-test split. To use the dataset extract all .npy files from the .7z file and copy all files to the data_FIOT and data_LC folders in the GNSS_dataset folder.

Spectrum Dataset

One sample in the text file is defined as for example:

data_FIOT/00566_0.npy 0 -1 -1

This file has the ID 00566, is the crop 0 from the crops 0 to 5, has the jammer class 0, the sub jammer id -1 (-1 stands for "no sub jammer"), and has the power -1 (-1 stands for "no known power"). For training your methods, use the train_FIOT.txt and test_FIOT.txt files to train and evaluate your model. Manual labeling of these snapshots has resulted in 9 classes: Class 0 represents samples with no interferences, while classes 1 to 8 contain different interferences. We applied six different croppings to augment the dataset.

Other Datasets

For our spectrum highway dataset 1 recorded along a highway bridge, see the following website: https://gitlab.cc-asp.fraunhofer.de/darcy_gnss/FIOT_highway

For our spectrum highway dataset 2 recorded along a highway bridge, see the following website: https://gitlab.cc-asp.fraunhofer.de/darcy_gnss/fiot_highway2

For our dataset recorded in a controlled environment with a high-frequency antenna, see the following website: https://gitlab.cc-asp.fraunhofer.de/darcy_gnss/FIOT_LC_laboratory

For our dataset recorded in a controlled environment with a low-frequency antenna, see the following website: https://gitlab.cc-asp.fraunhofer.de/darcy_gnss/controlled_low_frequency

LC Dataset

The train-test split of the LC Laboratory dataset is defined in the train_LC.txt and test_LC.txt files. With the definition:

data_LC/00566.npy 0 -1 -1

The dataset is recorded simultaneously with the FIOT dataset and the datasets can be matched with the file ID's. However, in the FIOT dataset are some samples (~400) that have no corresponding LC sample. The train-test split is the same. The dataset contains 736 samples. The LC dataset has no cropping included as the FIOT dataset. On LC file contains 10 timesteps and the following features:

lat,lon,alt,mode,epx,epy,epv,sep,ecefpAcc,ecefvAcc,gdop,hdop,vdop,tdop,nSat,uSat,prn_1,prn_2,prn_3,prn_4,prn_5,prn_6,prn_7,prn_8,prn_9,prn_10,prn_11,prn_12,prn_13,prn_14,prn_15,prn_16,prn_17,prn_18,prn_19,prn_20,prn_21,prn_22,prn_23,prn_24,prn_25,prn_26,prn_27,prn_28,prn_29,prn_30,prn_31,prn_32,prn_33,prn_34,prn_35,prn_36,prn_37,prn_38,prn_39,prn_40,prn_41,ss_1,ss_2,ss_3,ss_4,ss_5,ss_6,ss_7,ss_8,ss_9,ss_10,ss_11,ss_12,ss_13,ss_14,ss_15,ss_16,ss_17,ss_18,ss_19,ss_20,ss_21,ss_22,ss_23,ss_24,ss_25,ss_26,ss_27,ss_28,ss_29,ss_30,ss_31,ss_32,ss_33,ss_34,ss_35,ss_36,ss_37,ss_38,ss_39,ss_40,ss_41,az_1,az_2,az_3,az_4,az_5,az_6,az_7,az_8,az_9,az_10,az_11,az_12,az_13,az_14,az_15,az_16,az_17,az_18,az_19,az_20,az_21,az_22,az_23,az_24,az_25,az_26,az_27,az_28,az_29,az_30,az_31,az_32,az_33,az_34,az_35,az_36,az_37,az_38,az_39,az_40,az_41,el_1,el_2,el_3,el_4,el_5,el_6,el_7,el_8,el_9,el_10,el_11,el_12,el_13,el_14,el_15,el_16,el_17,el_18,el_19,el_20,el_21,el_22,el_23,el_24,el_25,el_26,el_27,el_28,el_29,el_30,el_31,el_32,el_33,el_34,el_35,el_36,el_37,el_38,el_39,el_40,el_41,sdr_1,sdr_2,sdr_3,sdr_4,sdr_5,sdr_6,sdr_7,sdr_8,sdr_9,sdr_10,sdr_11,sdr_12,sdr_13,sdr_14,sdr_15,sdr_16,sdr_17,sdr_18,sdr_19,sdr_20,sdr_21,sdr_22,sdr_23,sdr_24,sdr_25,sdr_26,sdr_27,sdr_28,sdr_29,sdr_30,sdr_31,sdr_32,sdr_33,sdr_34,sdr_35,sdr_36,sdr_37,sdr_38,sdr_39,sdr_40,sdr_41,sdr_42,sdr_43,sdr_44,sdr_45,sdr_46,sdr_47,sdr_48,sdr_49,sdr_50,sdr_51,sdr_52,sdr_53,sdr_54,sdr_55,sdr_56,sdr_57,sdr_58,sdr_59,sdr_60,sdr_61,sdr_62,sdr_63,sdr_64,kur_1,kur_2,kur_3,kur_4,kur_5,kur_6,kur_7,kur_8,kur_9,kur_10,kur_11,kur_12,kur_13,kur_14,kur_15,kur_16,kur_17,kur_18,kur_19,kur_20,kur_21,kur_22,kur_23,kur_24,kur_25,kur_26,kur_27,kur_28,kur_29,kur_30,kur_31,kur_32,kur_33,kur_34,kur_35,kur_36,kur_37,kur_38,kur_39,kur_40,kur_41,kur_42,kur_43,kur_44,kur_45,kur_46,kur_47,kur_48,kur_49,kur_50,kur_51,kur_52,kur_53,kur_54,kur_55,kur_56,kur_57,kur_58,kur_59,kur_60,kur_61,kur_62,kur_63,kur_64

License

This work is licensed under a CC BY-NC-SA 4.0: Creative Commons Attribution-Noncommercial-ShareAlike, see https://creativecommons.org/licenses/by-nc-sa/4.0/

Further Datasets & Methods

For more information on the dataset, links to other datasets, and our methodologies, visit the following Gitlab page: https://gitlab.cc-asp.fraunhofer.de/darcy_gnss