Low-Cost Microwave System Development for Analysis of Classification and Adulteration of Honey

Abstract

This article presents a novel, real-time, portable, and low-cost microwave sensing system for the classification and adulteration detection of natural honey using a log-periodic circular microstrip filter integrated with Wilkinson power dividers (WPDs) and an ADALM-PLUTO software-defined radio. The filter and WPDs operate at $\text{2.4}\;\text{GHz}$ and incorporate a copper sample chamber designed to accommodate $100\; \mu \mathrm{L}$ of liquid samples. Six types of unadulterated honey: 1) Mustard; 2) Wild; 3) Lychee; 4) Garlic; 5) Spiced; and 6) Cacao were tested, and each exhibited unique attenuation signatures ranging from −36.0$\mathrm{}$–26.5 $\text{dB}$. These measurements were used to define classification thresholds in a Python-based real-time GUI. To evaluate the detection of adulteration, common additives, such as fructose, glucose, sucrose, corn syrup, rice water, and sugar were mixed with each honey at concentrations of 10% to 50%. The system showed a consistent attenuation shift of approximately 2.5–$\text{3.0}\;\text{dB}$ (8–11% relative change) across adulteration levels, with a resolution of $\text{0.5}\;\text{dB}$ and a coefficient of variation below 0.4%. These results demonstrate the feasibility of electromagnetic sensing for rapid, nondestructive, and repeatable detection of honey adulteration.