77-GHz FMCW Radar Network ADC Data under Various Interference Scenarios at Parking Lot

The dataset contains 77-GHz frequency-modulated continuous-wave (FMCW) radar analog-to-digital converter (ADC) data from seven measurements of a loosely coupled two-node radar network under different interference and scene configurations. All measurements were performed outdoors at a university parking lot. The two radar-network stations operate in a master/slave configuration (Station1 triggers Station 2), are time-synchronized, and start their chirp sequences with a small time offset, which allows the system to capture both monostatic and bistatic returns. The interferer, on the other hand, is unsynchronized with respect to the radar network. For each measurement, one of four predefined chirp-sequence parameter sets was selected for the interferer, in order to generate different interference patterns at the stations of the radar network. For all measurements, the (static) scene contained several parked cars on both sides of the radar network. Moreover, except for the initial (reference) measurement, a corner reflector was positioned in front of Station 2. In addition, during the fourth measurement, a person was walking in front of Station 1 toward the interferer. The dataset can be used for developing and testing interference detection and mitigation algorithms, as well as for research on multistatic processing and robust signal processing for automotive radar. Each .mat file contains two 4-D arrays data_station_1 and data_station_2 corresponding to Station 1 and Station 2, respectively. Both arrays are of dimensions N x M x L x N_frames, where N = 512 (samples per chirp) M = 128 (chirps per frame) L = 16 (Rx channel) N_frames = 100 (frames per measurement) An exemplary processing file processing.m that loads the measurement data, and computes and visualizes for each frame the signal power, the range map, as well as the range-Doppler map, is provided. The provided .mp4 video files show the visualization output for each measurement in real-time, based on the processing steps in processing.m. Furthermore, Octave- and Python-compatible processing files processing_octave.m and processing.py, are provided. For additional information about the setup and/or the parameterizations of the chirp sequences, please see the documentation file documentation.pdf attached to this dataset.