Machine Learning-Optimized Wearable Antenna for LoRa Localization

Citation

Yahya, Muhammad S. and Soeung, Socheatra and Abdul Rahim, Sharul Kamal and Tan, Kim Geok and Musa, Umar (2024) Machine Learning-Optimized Wearable Antenna for LoRa Localization. IEEE Access. p. 1. ISSN 2169-3536

[img] Text
Machine Learning-Optimized Wearable Antenna for LoRa Localization.pdf - Published Version
Restricted to Repository staff only

Download (5MB)

Abstract

This paper presents the design and implementation of a compact, wearable, printed monopole antenna for Long-Range (LoRa) location tracking applications, covering the 915 MHz and 923 MHz LoRa bands. Fabricated on a Rogers Duroid RO3003™ substrate, the antenna integrates meandered monopoles to enable miniaturization, alongside a matching stub and a partial ground plane for enhanced impedance matching and performance efficiency. It features a bidirectional radiation pattern in the E-plane and an omnidirectional pattern in the H-plane at both frequencies, achieving a peak gain of 2 dBi and a radiation efficiency of 98% across its operational bands. With a total dimension of 50 × 50 mm2 (0.153 λ0 × 0.153 λ0), where λ0 represents the free-space wavelength at 915 MHz), it represents the most compact LoRa wearable antenna known to the authors. The design, simulation, and analysis were carried out using CST Microwave Studio (MWS) software. Bending investigations showed excellent efficiency with minimal impact on bandwidth and gain, while Specific Absorption Rate (SAR) analysis indicated compliance with safety limits set by the Federal Communications Commission (FCC) and International Commission on Non-Ionizing Radiation Protection (ICNIRP) standards. SAR values at 915 MHz were measured at 0.94 W/kg for 1 gram of tissue and 0.85 W/kg for 10 grams, affirming the antenna’s suitability for wearable applications. Correspondingly, at 923 MHz, SAR values stood at 0.92 W/kg for 1 gram of tissue and 0.82 W/kg for 10 grams of tissue. Furthermore, real-world experimental validation employing the proposed antenna within a wearable LoRa tracking system revealed superior performance in LoRa localization. The average improvement in Received Signal Strength Indicator (RSSI) of 3 dBm, compared to a conventional antenna, was observed within a range of 1.5 km. This comprehensive assessment solidifies the proposed antenna’s potential as a promising and dependable solution for efficient and reliable LoRa-based location tracking across various scenarios.

Item Type: Article
Uncontrolled Keywords: IoT, optimization, machine learning
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television
Divisions: Faculty of Engineering and Technology (FET)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 01 Oct 2024 05:58
Last Modified: 01 Oct 2024 05:58
URII: http://shdl.mmu.edu.my/id/eprint/13028

Downloads

Downloads per month over past year

View ItemEdit (login required)