Open Lab

OL17 Drone detection and classification from radar data

Background

Drones are becoming a major threat for both civilian (because of privacy issues) and industrial/defence facilities. Radar is a useful sensor to detect drones because of their day-night and all-weather functionalities.

Aim/goal/research question

Radar data analysis using various statistical methods like methods Visualisation of the data using PCA, kPCA, UMAP etc Design of classifier based on the micro-Doppler and/or PCA features. Design of a deep-learning classifier based on the aove features. (Optional) Use Kalman filter to track the drones

Method

In this project, the student would work on designing classification algorithms to classify drones. Mostly, we shall use the publicly released dataset: Drone identification and tracking | Kaggle

Recommended past experience/interest

Coding skills, DSP Skills

Related Work

Application of linear and nonlinear PCA to SAR ATR | IEEE Conference Publication | IEEE Xplore

Contact

amit.kumar-mishra@hv.se