.png){kind=icon}
.png){kind=icon}
Return
DNN-Based RFID Antenna Tags Localization
.png){kind=icon}
PRE
AI
DOI：10.1109/I2MTC50364.2021.9460004.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Radio frequency identification technology (RFID) is increasingly becoming an integral part of the Internet-of-Things (IoT). It offers different advantages including battery-free operation, small form-factor, and low cost. This makes the RFID an enticing technology for an indoor localization-based application and services. Geometry based localization approaches often achieve low accuracy due to errors introduced by a multi-path propagation and interference in indoor environments. Many range-based algorithms assume that reader position is known in advance and there are carefully placed reference tags. In contrast, this paper presents a data driven localization methodology for direction-of-arrival (DOA) estimation using a deep neural network processing of signal captured with a reader antenna array. The proposed approach learns the complex mapping of the radio waves interactions in adverse metal environments based on received signal strength indicator (RSSI) values. The RSSI is captured while electrically steering a planar phased array through the area of interest. The proposed methodology is evaluated with multiple tags placed on metallic surfaces. Using readily available measurements, the proposed approach is able to achieve an average DOA error of 5.93 degrees.
Keywords:
machine learning (ML)
deep neural network (DNN)
internet-of-things (IoT)
UHF RFID
localization
phased array antennas
direction-of-arrival (DOA)
Journal
I
IF:
0
Papers:
16
Citations:
0
