.png){kind=icon}
.png){kind=icon}
Return
Joint Localization and NLOS Identification Exploiting Reconfigurable Intelligence Surface
.png){kind=icon}
PRE
AI
DOI：10.1109/TWC.2024.3441556.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Reconfigurable intelligent surface (RIS) is regarded as a promising technology to potentially enhance communications and localization capabilities. In this paper, we investigate a challenging received signal strength (RSS) based localization problem utilizing RIS. In particular, we consider a mixed line-of-sight and non-line-of-sight (LOS/NLOS) scenario with multiple sources and sensors as well as one RIS. Firstly, we obtain a closed-form Cramer-Rao lower bound (CRLB) expression and formulate a joint optimization problem to minimize the CRLB. Due to the non-convex objective function and coupled relationship between variables, it is difficult to solve directly. To overcome such issues, a novel two-stage alternating localization and passive beamforming (PBF) scheme is proposed, where the optimization problems corresponding to two stages, namely the joint localization and NLOS identification (JLNI) stage and the PBF optimization (PBFO) stage are alternately addressed. Specifically, for the JLNI stage, we adopt alternating optimization, parametric sparse Bayesian dictionary learning and block coordinate descent approaches to jointly estimate the locations of sources and LOS/NLOS path condition. Subsequently, for the PBFO stage, we optimize the PBF vector by exploiting semi-definite relaxation, difference-of-convex programming and successive convex approximation approaches. Finally, compared with the state-of-the-art methods, numerical simulations demonstrate the superiority and effectiveness of the proposed scheme in terms of CRLB, localization error and NLOS identification accuracy.
Keywords:
Location awareness
Sensors
Optimization
Vectors
Bayes methods
Millimeter wave communication
Array signal processing
Reconfigurable intelligent surface (RIS)
multiple sources localization
received signal strength (RSS)
Cramer-Rao lower bound (CRLB)
sparse Bayesian learning and semi-definite relaxation
Journal
IF:
10.7
Papers:
1.3W
Citations:
5.3W