Classification of Flying Objects Using Data from UAV Mounted Radar

Abstract

Unmanned aerial vehicles (UAVs)/Drones have been broadly used in modern civilization over the past few years due to their low cost and ease of accessibility, which has raised concerns about privacy and security. It needs to classify flying objects, such as helicopters, birds, UAV/drone, etc., in order to maintain a watchful eye on the invader UAV/drone in the restricted area. In this paper classification of the flying object is done using Hybrid Convolutional Neural Network-Memetic (CNN-Memetic) Algorithm based on MicroDoppler Signature (MDS) for various arrangement of radar array in order to verify the significance of direction of signal received. The evaluation is done based on data acquired from the radar borne on the drone by varying different specifications.