LSS-DAUR-1.0: Digital Array Ubiquitous Radar Low, Slow, and Small Target Detection Dataset

The Low, Slow, and Small Target Detection Dataset for Digital Array Surveillance Radar (LSS-DAUR-1.0) includes a total of 154 items of Range-Doppler (RD) complex data and Track (TR) point data collected from 6 types of targets (passenger ships, speedboats, helicopters, rotary-wing UAVs, birds, fixed-wing UAVs). It can support research on detection, classification and recognition of typical maritime targets by digital array radar. 1. Data Collection Process The data collection process mainly includes: Set radar parameters → Detect targets → Collect echo signal data → Record target information → Determine the range bin where the target is located → Extract target Doppler data → Extract target track data. 2. Target Situation The collected typical sea-air targets include 6 categories: passenger ships, speedboats, helicopters, rotary-wing UAVs, birds and fixed-wing UAVs. 3. Range-Doppler (RD) Complex Data By calculating the target range, the echo data of the range bin where the target is located is intercepted. Based on the collected measured data, the Low, Slow, and Small Target RD Dataset for Digital Array Surveillance Radar is constructed, which includes 10 groups of passenger ship (passenger ship) data, 11 groups of speedboat (speedboat) data, 10 groups of helicopter (helicopter) data, 18 groups of rotary-wing UAV (rotary drone) data, 17 groups of bird (bird) data, and 11 groups of fixed-wing UAV (fixed-wing drone) data, totaling 77 groups. Each group of data includes the target's Doppler, GPS time, frame count, etc. The naming method of target RD data is: Start Collection Time_DAUR_RD_Target Type_Serial Number_Target Batch Number.Mat. For example, the file name "20231207093748_DAUR_RD_Passenger Ship_01_2619.mat", where "20231207" represents the date of data collection, "093748" represents the start time of collection which is 09:37:48, "DAUR" represents Digital Array Surveillance Radar, "RD" represents Range-Doppler spectrum complex data, "Passenger Ship_01" represents the target type is passenger ship with serial number 01, and "2619" represents the target track batch number. 4. Track (TR) Data Extract the track data within the time period of the echo data, and construct the Low, Slow, and Small Target TR Dataset for Digital Array Surveillance Radar, which includes 10 groups of passenger ship (passenger ship) data, 11 groups of speedboat (speedboat) data, 10 groups of helicopter (helicopter) data, 18 groups of rotary-wing UAV (rotary drone) data, 17 groups of bird (bird) data, and 11 groups of fixed-wing UAV (fixed-wing drone) data, totaling 77 groups. Each group of data includes target range, target azimuth, elevation angle, target speed, GPS time, signal-to-noise ratio (SNR), etc. The TR data and RD data have the same time and batch number, and they are data of different dimensions for the same target in the same time period. The naming method of target TR data is: Start Collection Time_DAUR_TR_Target Type_Serial Number_Target Batch Number.Mat. For example, the file name "20231207093748_DAUR_TR_Passenger Ship_01_2619.mat", where "20231207" represents the date of data collection, "093748" represents the start time of collection which is 09:37:48, "DAUR" represents Digital Array Surveillance Radar, "TR" represents Range-Doppler spectrum complex data, "Passenger Ship_01" represents the target type is passenger ship with serial number 01, and "2619" represents the target track batch number.