| Title: |
An Autonomous Approach to Deinterleave andRecover Radar Pulse Sequences in an UnknownMaritime Environment |
| Authors: |
Martin, Guillaume; Bouzid, Salah Eddine; Alanic, Jean-Luc; Charge, Pascal |
| Contributors: |
Institut d'Électronique et des Technologies du numéRique (IETR); Université de Rennes (UR)-Institut National des Sciences Appliquées - Rennes (INSA Rennes); Institut National des Sciences Appliquées (INSA)-Institut National des Sciences Appliquées (INSA)-CentraleSupélec-Centre National de la Recherche Scientifique (CNRS)-Nantes Université - pôle Sciences et technologie; Nantes Université (Nantes Univ)-Nantes Université (Nantes Univ); AMG Microwave (AMG Microwave) |
| Source: |
International Radar Conference ; https://hal.science/hal-04185150 ; International Radar Conference, Nov 2023, Sydney, Australia. Paper ID: 279, 2023 |
| Publisher Information: |
HAL CCSD |
| Publication Year: |
2023 |
| Collection: |
Université de Rennes 1: Publications scientifiques (HAL) |
| Subject Terms: |
intelligent transponders; explainable artificial intelligence; radar signal processing; deinterleaving; unsupervised algorithms; homogeneity and completeness score; [SPI.TRON]Engineering Sciences [physics]/Electronics |
| Subject Geographic: |
Sydney; Australia |
| Description: |
International audience ; In a high-density maritime environment, therecovery of radar pulse sequences presents a challenge totransponders because of interleaved pulses. Moreover, due toa few manufacturers of civil radars, the lack of diversity insignal signatures makes the task harder. In this paper, we presenta new approach to deinterleave radar pulse sequences withoutspecific knowledge. Our proposed solution, based on a three-stepapproach, enables transponders to recognize their environmentand interact with emitters more effectively without any trainingstep. The effectiveness of this approach is demonstrated throughsimulations. Obtained results show that predicted clusters havea homogeneity score higher than 98%, a completeness scorehigher than 92%, and a Rand index higher than 97%. Oursolution can deinterleave and recover radar pulse sequences inan unknown maritime environment in real-time without priorknowledge, while providing a thorough interpretation of results. |
| Document Type: |
conference object; still image |
| Language: |
English |
| Availability: |
https://hal.science/hal-04185150; https://hal.science/hal-04185150/document; https://hal.science/hal-04185150/file/2023_RADAR_CONF_Poster_GMARTIN.pdf |
| Rights: |
info:eu-repo/semantics/OpenAccess |
| Accession Number: |
edsbas.BF72E0D7 |
| Database: |
BASE |