Source Localization Using Leaky Wave Antenna By - Nantes, France - Yide WANG

Description

• Réf

• Sujet de Thèse 23/01/2023
• Financement public/privé
• Yide WANG
• Lieu de travail
• NANTES
• Pays de la Loire
• France
• Intitulé du sujet
• Source localization using Leaky Wave Antenna by subspacebased methods
• Champs scientifiques
• Sciences de l'ingénieur
• Télécommunications

• Array signal processing, DOA finding, subspace based methods, high resolution methods
  Description du sujet:

This research project proposed jointly by IETR/Polytech Nantes and Sorbonne University, Paris (GeePs laboratory) is supported financially by the ANR project: BeSensiCom.

LWAs are a class of traveling wave antennas that can efficiently replace antenna arrays to achieve directional properties (i.e., narrow beams) with a simple feeding structure.

An interesting feature is that LWAs naturally exhibit beam scanning properties with frequency and can therefore be used to dynamically estimate the DOA of different sources.

• The estimation of DOA and power of incoming sources with an LWA, is equivalent to work in the beam space, generated by the LWA frequency scanning. However, since the equivalent directional matrix does not exhibit a Vandermonde structure, several efficient methods classically developed with linear arrays cannot be used. The research work will therefore comprise the following steps.
• Firstly, we will review and implement some classical DOA estimation techniques which can be directly adapted to LWA, such as beamforming methods and the highresolution spectral MUSIC method. The maximum likelihood estimator (MLE) will also be implemented for performance benchmark. Once DOAs are estimated, the power associated to the identified incoming sources will be determined with classical leastsquares approaches to obtain the required power angular spectrums.
• Secondly, we will investigate different approaches to transform the directional matrix or manifold matrix of the different LWAs to a Vandermonde matrix, so that several numerically efficient highresolution techniques such as Root-MUSIC or ESPRIT can be applied. In addition, this transformation is required to estimate coherent signal sources, which is necessary in multipath situations. With this transformation, the classical spatial smoothing techniques and the PUMA method can also be used for localizing coherent sources and/or singlesnapshot signal source.
• Finally, to further improve the performance of the proposed methods in the second step, we will try to exploit the statistical properties of the transmitted pilot sequences found in existing 4/5G reference signals. Aiming at the same goal, we will also optimize the transmitted pilot sequence by exploiting the noncircularity, cyclostationarity, and/or higher order statistics, in the above proposed methods. Under the same context, we will also optimize the transmitted pilot sequence to further improve the performance and to draw several guidelines on the design of pilot sequences for future standards (e.g., 6G).

• 28/08/2023
  Nature du financement:
• Financement public/privé

• ANR
  Présentation établissement et labo d'accueil:
• Yide WANG

• Doctorat en traitement du signal et télécommunication
  Pays d'obtention du doctorat:
• France

• Nantes Université
  Ecole doctorale:
• Mathématiques et Sciences et Technologies de l'Information et de la communication (MATHSTIC)Traitement du signal multidimensionnel