This paper presents a method to predict potassium percentage and volumetric water content (VWC) in soil by analyzing its electrical properties. It highlights the role of potassium as an important factor in soil salinity and examines how potassium and VWC affect soil permittivity and electrical conductivity (EC) at different frequencies. A dataset is created with diverse potassium levels and VWC percentages, measuring electrical features from 10-295 MHz at 58 discrete frequencies with 5 MHz resolution. Initially constructed with 20 soil samples, the dataset was expanded to 2,000 samples through two-dimensional interpolation for model development, with 90% of the data used in the training phase. Using these measurements, a neural network model is developed to predict VWC and potassium levels accurately. The model's prediction error improved significantly when using data from over 30 frequencies, reducing potassium prediction error to less than 0.05 g/kg and VWC prediction error to less than 0.4%. This study makes its measurements using DAK (Dielectric Assessment Kit) and VNA (Vector Network Analyzer), comparing the outputs of two commercial sensors with actual values and provides valuable insights for the development of advanced soil sensors.

This industry-driven paper presents the development of a novel multi-band soil sensor capable of simultaneously measuring volumetric water content (VWC) and potassium content, which serves as an indicator of soil salinity. The sensor operates by analyzing the soil’s electromagnetic (EM) properties. Unlike conventional Time-Domain Reflectometry (TDR) methods that rely on pulse signals and time-delay measurements, the proposed sensor leverages continuous-wave signals to capture phase differences and amplitude variations across a wide frequency range. By utilizing a frequency vector spanning from 1 to 10 MHz rather than a single operating frequency, the sensor achieves enhanced precision and reliability in data acquisition. A key feature of the proposed sensor is the integration of a machine learning model, which processes the collected raw data to accurately predict both VWC and salinity levels. The machine learning algorithm is trained to interpret complex signal variations, enabling the system to provide highly accurate and consistent measurements across different soil types and environmental conditions. Furthermore, the sensor includes a self-calibration circuit that compensates for external factors such as temperature fluctuations, soil composition variability, and other environmental influences, ensuring reliable performance across diverse agricultural and environmental monitoring applications. This multi-band sensing approach, combined with dual-measurement capability and advanced data processing techniques, offers a promising industry-grade solution for precision agriculture, irrigation management, and soil health monitoring. The proposed sensor system enables real-time, non-destructive, and cost-effective soil analysis, supporting sustainable agricultural practices by enabling data-driven decision-making.

Soil salinity is a critical factor impacting agriculture and the environment. This study develops a frequency-dependent model to predict soil salinity based on the dielectric permittivity and volumetric water content. A targeted soil database was created, incorporating varied salinity and moisture levels from two soil types: sandy and clay. Real and imaginary parts of permittivity were measured across a frequency range of 10–295 MHz using the dielectric assessment kit vector network analyzer system, as permittivity in this band shows the highest variation and sensitivity to moisture and salinity. After analyzing the behavior of permittivity changes and minimizing prediction errors, two separate models were proposed for the real and imaginary parts. These models are significantly more sophisticated and flexible than previous ones, designed to accept inputs that are easily measurable by commercial sensors. The comparison between model outputs and measured data demonstrates a high level of alignment, indicating that the model can improve salinity prediction accuracy and extend its applications beyond agriculture to practical uses in geology and hydrology. The average prediction error was less than 8.2% for the real part and 9.3% for the imaginary part.

Soil salinity is a critical factor influencing agricultural productivity and environmental sustainability, requiring precise monitoring tools. This article focuses on developing a frequency-dependent model to predict soil salinity based on electrical conductivity (EC) and volumetric water content (VWC). A dataset of 40 soil samples with varying levels of salinity and moisture, consisting of two soil types (sandy and clayey), was experimentally measured for EC in the frequency range of 10 to 295 MHz using EC spectroscopy measurement with the dielectric assessment kit–vector network analyzer) system. A new, more comprehensive frequency-dependent model is proposed, surpassing previous models that lacked frequency considerations. This modeling approach was conducted in stages: initially, a frequency-independent model for EC as a function of salinity and VWC was developed. Next, a frequency-dependent model was introduced. Finally, a comparison between pure sandy soil and a sandy–clay mixture led to the final model, which also incorporates effective porosity. The results of the proposed model, comparing measured and predicted values, provide a robust approach to accurately predict soil salinity. Findings demonstrate that the model can enhance salinity prediction accuracy, extending its applicability beyond agriculture to geological and hydrological applications in real-world scenarios.

This industry-driven paper introduces a novel multi-frequency soil sensor for precision agriculture applications. The sensor simultaneously measures Volumetric Water Content (VWC) and salinity by leveraging the soil’s electrical properties. Unlike traditional Time Domain Reflectometry (TDR) sensors, which rely on pulse signals and time-delay measurements, the proposed sensor utilizes continuous waves, phase shifts, and amplitude variations to enhance accuracy and reduce system complexity. Operating simultaneously at frequencies of 70 MHz, 140 MHz, and 210 MHz , the multi-frequency approach mitigates the errors inherent in single-frequency measurements, ensuring improved precision even in heterogeneous soils. A cost-effective frequency generation method utilizing a single voltage-controlled oscillator (VCO) and its harmonics was implemented, significantly reducing hardware requirements while maintaining high performance. Additionally, machine learning models, trained on a dedicated calibration dataset, were used to accurately interpret raw sensor data and predict both VWC and salinity values. Machine learning models, including a Neural Network (NN) and Support Vector Regression (SVR) with a radial basis function (RBF) kernel, were developed and compared to estimate soil volumetric water content and salinity from phase and amplitude data. Experimental results demonstrated the proposed MMSS sensor achieves high accuracy in predicting soil parameters, with mean absolute error (MAE) of 0.4% for VWC and 0.05 g/kg for potassium concentration, outperforming existing soil moisture sensors, and demonstrating robust performance for precision agriculture applications and environmental monitoring. Its accuracy, scalability, and cost-effectiveness make it a valuable tool for optimizing resource usage and enhancing crop yields.