Selection of Best Frequency Range for Soil Urea Prediction
Sulaxana R.Vernekar1, Jivan S. Parab2

1Sulaxana R. Vernekar*, Assistant Professor Department of Computer Science, GVM’s GGPR College of Commerce, Ponda, Goa
2Jivan S. Parab, Assistant Professor, Department of Electronics, Goa University. Goa
Manuscript received on December 14, 2019. | Revised Manuscript received on December 22, 2019. | Manuscript published on January 10, 2020. | PP: 1934-1936 | Volume-9 Issue-3, January 2020. | Retrieval Number: C8904019320/2020©BEIESP | DOI: 10.35940/ijitee.C8904.019320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: Soil is a complex system and its nature determines the types of crops that can be cultivated. Soil testing plays an important role in determining its ability to grow crops. Conventional soil testing methods are found to be cumbersome and expensive and also time consuming. Hence there is a need for methods which can give fast results and help in effective crop cultivation. Soil exhibits both spatial and temporal variability. The traditional methods of soil testing do not take into account the variability of the soil and uniform application of external inputs is done. This leads to over or under use of fertilizers, which in turn results into soil turning infertile, ground water getting contaminated etc. Precision Farming technique makes use of new technologies to take into consideration the variability exhibited by soil and is also called as site specific management. Soil nutrient testing is an important aspect of soil testing which helps in finding out the available soil nutrients and which in turn determines the crops that can be cultivated. Various techniques have been developed to determine the soil nutrients but most of these techniques are found to be time consuming. Hence there is a need to develop techniques which can give real time measurements of soil nutrients. This paper discusses about the use of RF spectra for predicting the soil nutrients. The RF spectra are obtained using a cell which is designed based on the principle of dielectricity. Samples were prepared in the laboratory by mixing five different components namely urea, potash, phosphate, lime and salt in distilled water. RF spectra of different samples having varying concentrations of the components were recorded. Multivariate analysis based on the Partial Least Square Regression technique was used to predict the amount of urea in a sample. The prediction of urea was done using two different frequency ranges i.e 10MHz-500MHz and 500MHz – 1000MHz and analysis of the results was done to determine which frequency range gives better results. The results show that percentage error of urea prediction is better in the frequency range of 500MHz-1000MHz as compared to 10MHz-500MHz. 
Keywords:  RF, Urea, PLSR, Multivariate.
Scope of the Article: Frequency selective surface