Precise Identifying Assets Inside a Metal Cabinet Using RFID and Machine Learning Method

The demand for CLOUD services has increased in recent years. Data centers have been equipped with more information technology (IT) assets to meet this trend; therefore, an efficient inventory management system is essential to track growing assets. The purpose of the study was to propose a novel RFID...

Full description

Saved in:
Bibliographic Details
Published in2022 25th International Conference on Mechatronics Technology (ICMT) pp. 1 - 4
Main Authors Chen, Fu-Kuei, Lin, Hsin-Piao, Zhou, Sheng-Kai
Format Conference Proceeding
LanguageEnglish
Published IEEE 18.11.2022
Subjects
Data centers
Feature extraction
Machine learning
Mechatronics
Metals
multi-element antenna
Predictive models
RFID
Support vector machines
Online AccessGet full text

Cover

More Information
Summary:The demand for CLOUD services has increased in recent years. Data centers have been equipped with more information technology (IT) assets to meet this trend; therefore, an efficient inventory management system is essential to track growing assets. The purpose of the study was to propose a novel RFID-based audit inventory system to precisely identify assets inside the metal cabinet of the data center. A multi-element RFID antenna and a machine learning method were adopted to implement such a system. The experimental results revealed that the extracted features of the Received Signal Strength Indicator (RSSI), Tag Read Count (TRC), combining Model Support Vector Machines (SVM), outperformed the other models with exceeding 99% accuracy in prediction performance. The findings of the study may serve as a guide for other RFID-based systems for static asset tracking in a metal cabinet.
DOI:10.1109/ICMT56556.2022.9997736