Radar-Based Multiple Target Classification in Complex Environments Using 1D-CNN Models

In this paper, we propose a robust multiple target classification algorithm for real-world complex cluttered environments that can be mapped into low-cost millimeter-wave (mmWave) sensors considering limited memory and processing power budget. A novel approach is developed to create both μ-Doppler a...

Full description

Saved in:
Bibliographic Details
Published in2023 IEEE Radar Conference (RadarConf23) pp. 1 - 6
Main Authors Yanik, Muhammet Emin, Rao, Sandeep
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.05.2023
Subjects
Convolutional neural networks
Feature extraction
micro-Doppler
micro-range
Millimeter wave radar
multiple target classification
Neural networks
Radar tracking
Sensors
Target tracking
Online AccessGet full text

Cover

More Information
Summary:In this paper, we propose a robust multiple target classification algorithm for real-world complex cluttered environments that can be mapped into low-cost millimeter-wave (mmWave) sensors considering limited memory and processing power budget. A novel approach is developed to create both μ-Doppler and μ-range spectrogram of multiple objects concurrently using an extended Kalman filter (EKF) based tracking layer integration. One-dimensional (1D) time sequence features are extracted from both spectrograms per target object, and a 1D convolutional neural network (CNN) based classifier is built to classify multiple target objects (human or non-human) in the same scene accurately.
DOI:10.1109/RadarConf2351548.2023.10149609