Machine learning-assisted reconfigurable dual-band wearable patch antenna for body-centric IoT applications
|
Full Text |
Pdf
|
|
Author |
P. Santosh, B. Mahalakshmi, T. Sri Deepika, Y. M. S. R. Krishna, K. Yeswanth and Shaik Ansar Bibi
|
|
e-ISSN |
1819-6608 |
|
On Pages
|
678-689
|
|
Volume No. |
21
|
|
Issue No. |
10
|
|
Issue Date |
July 20, 2026
|
|
DOI |
https://doi.org/10.59018/052673
|
|
Keywords |
wearable antenna, machine learning, reconfigurable antenna, body-centric IoT, dual-band antenna, adaptive communication.
|
Abstract
Due to fluctuating wireless settings and signal strength degradation caused by human body impacts, reconfigurable antennas using machine learning methods are seeing more applications in body-centric Internet of Things devices. For the purpose of reconfiguring frequency bands, this study presents the construction of a miniature wearable dual-band patch antenna with a PIN diode. Antennas of this kind work well in the 2.44 GHz and 5.09 GHz ranges. By using signal metrics like RSSI and SNR, an adaptive switching circuit is designed using a machine learning approach. This model is then utilised to identify the optimal frequency range. We can see that the gain is within 7 dBi, the specific absorption rates are within safety limits, and the impedance matching is within an acceptable range (S11 > -15 dB) from the simulation results. The developed wearable antenna can withstand bending conditions, making it ideal for use in wearable applications.
Back