Applications for the EM-based classifier in radar sensor network

Tools

Yan, Linjie and Jahangir, Mohammed and Antoniou, Michail and Hao, Chengpeng and Clemente, Carmine and Orlando, Danilo (2025) Applications for the EM-based classifier in radar sensor network. IEEE Sensors Letters, 9 (3). 7001204. ISSN 2475-1472 (https://doi.org/10.1109/lsens.2025.3540732)

[thumbnail of Yan-etal-2025-Applications-for-the-EM-based-classifier-in-radar-sensor-network]
Preview
 Text. Filename: Yan-etal-2025-Applications-for-the-EM-based-classifier-in-radar-sensor-network.pdf
Accepted Author Manuscript
License: Creative Commons Attribution 4.0 logo
Download (664kB)| Preview

Abstract

In this letter, we focus on the application and analysis of the new model-based clustering architectures developed in our recent paper, where the analysis is limited to synthetic simulation results, to data collected by a real radar sensor. Specifically, a more comprehensive analysis of the proposed schemes is carried out in challenging real operating scenarios where the real measurements of multiple moving targets are not perfectly matched with the design assumptions due to real-world effects. Moreover, a new initialization procedure is introduced that accounts for multiple target velocities and the radar sampling time interval required by the specific application. Such a procedure is capable of providing the expectation-maximization (EM) procedure with reliable initial parameter values. The performance assessment confirms the effectiveness of these EM-based clustering algorithms not only on synthetic data, as observed in our companion paper, but also over real-recorded data and in comparison with suitable competitors.

ORCID iDs

Yan, Linjie, Jahangir, Mohammed, Antoniou, Michail, Hao, Chengpeng, Clemente, Carmine ORCID logoORCID: https://orcid.org/0000-0002-6665-693X and Orlando, Danilo;
CORE (COnnecting REpositories)
Sorry the service is unavailable at the moment. Please try again later.