On the Enhancement Anomaly Detection for RF Bio-Sensors by Computing Artificial Networks Using Machine Learning Techniques

Kamil, Raya Adel and Alsabti, Saif Mohamed Baraa and Abdulsattar, Rusul K. and Mohammed, Ammar H. and Elwi, Taha A. (2025) On the Enhancement Anomaly Detection for RF Bio-Sensors by Computing Artificial Networks Using Machine Learning Techniques. INFOCOMMUNICATIONS JOURNAL, 17 (2). pp. 89-95. ISSN 2061-2079

Preview

Text InfocomJournal_2025_2_11.pdf - Published Version Download (2MB) | Preview

Abstract

This work proposes serval sensor designs for a lowcost, highly-sensitive microwave sensor for identifying different liquid samples by monitoring the variation in S21 magnitude. The sensor is developed using an interdigital capacitor (IDC) in series connection with a circular spiral inductor (CSI) and connected directly to a photo-resistor (LDR). To enhance sensor insertion losses, the sensor is introduced to a Hilbert fractal open stub and coupled to an interdigital capacitor to operate at 1.22GHz. The accuracy of the sensor is significantly improved using a back loop trace, eliminating nonlinear effects from multi-layer diffractions. An analytical model based on circuit theory is suggested for the proposed sensor operation. The authors found an observable influence of varying the LDR value on sensor insertion losses, motivating the development of the sensor prototype. The sensor is manufactured and tested experimentally before and after samples introduction, with a human glucose sample mounted on the LDR patch to measure the effects of light intensity.

Actions (login required)

Edit Item