System overview

The intelligent radar perception system realizes the detection and intelligent recognition of “low, slow and small” targets such as unmanned aerial vehicles (rotors, fixed wings), helicopters, airborne balloons, birds, etc. in complex ground clutter and sea clutter environments, providing location, velocity and category information of targets, and providing information guidance for disposal methods such as laser damage and electronic interference.

The most advantage technical feature of the system is the use of deep modeling technology, which can achieve intelligent recognition of targets such as drones, helicopters, birds, etc. With a recognition accuracy of 90%, it can greatly reduce the workload of people in the loop. The functions include:

1) With azimuthal circumferential sweep/sector+pitch electric sweep mode for airspace target detection and tracking;

2) Deep learning capability of target images, using deep learning algorithms for recognition of target micro-Doppler feature maps to achieve high recognition rates of targets;

3) It can report 3D position and 3D velocity information of the target in real time;

4) It can predict the target flight path;

5) Software background, with the function of information interaction with the overall control software;

6) Support device startup self-test, status monitoring, automatic fault diagnosis display and report functions.

7) With automatic, all-weather working ability, the radar system can be unattended, reduce the difficulty of work, and avoid electromagnetic radiation interference to the operator.

System specification

Table 1 Intelligent Radar Detection System Specifications

System composition

The intelligent radar detection system consists of two parts: the radar (including tripod) and the intelligent processing device, as shown in the diagram below.

Figure 1 Intelligent Radar Detection System Composition

The intelligent radar detection system extracts the micro-Doppler motion features of typical targets such as drones and birds from the medium frequency signals, and then realizes the rapid intelligent recognition of targets such as drones, birds, floating balloons and helicopters through the customized deep neural network recognition network, with the recognition accuracy of better than 90%. And the system can quickly form the target situation, has the advantages of high intelligence, accuracy and efficiency, and able to solve the problem of high false alarm rate of target radar detection in complex scenes.