Meet SHOEWU OMOTAYO OLUWAGBEMIGA, an Academic Staff of Lagos State University.
Electronics And Telecommunication
Associate Professor / Reader
Electronics and Computer Engineering
At the Electronics And Computer Engineering department office
Appointment on Visitation important
Topic: Pathloss Model Development In A Vineyard Environment Using Machine Learning Methods
Description:
Accurate pathloss is critical for deploying IoT, precision agriculture, and private 5G networks in vineyards. Traditional empirical models like Okumura-Hata and COST -231 fail in vineyard environments due to unique clutter: row-structured vines, seasonal canopy density and undulating terrain. This research proposes a dat-driven path loss model for vineyard environment using machine learning. Adequate and extensive narrowband IoT measurements at 800MHz to 2.4GHz across three vineyard sites in Nigeria over two growing seasons. A data set of 42000 samples with features including distance, frequency, canopy height, row orientation,, soil moisture and season was collected. Four machine learning models were trained and compared: Random Forest, Gradient Boosting, Support Vector Regression and a 3-layer ANN. Results show that xGboost outperformed empirical models with RMSE = 3.87dB vs 9.47dB for COST 321, and R2 = 0.94. Feature importance analysis revealed canopy destiny and row alignment contribution 48% to pathloss variance. The proposed ML model enables 31% more accurate link budget planning for vineyards WSN deployments.
| # | Certificate | School | Year |
|---|---|---|---|
| 1. | Ph.D (Electrical Engineering Electronics and Communications Engineering ) | Department of E;lectrical/Electronics Engineering, Michael Okpara University of Agriculture, Umudike | 2021 |
Ab initio Quantum Physics Based Dielectric function for irregular Geometry Using Conservative Finite Difference Method
Effect of Electromagnetic waves on Off-shore and On-shore in Vegetational Environment. Plasmonic devices with accelerated computing in Dielectric Material coupled with Electromagnetic Signals. Ab initio Quantum Physics-Based Dielectric function for Irregular Geometry Using Conservative Finite Difference Method Due to non-comprehensive numerical technique for modelling and simulation of plasmonic device applications and particularly small-scale devices as a result of scaling of semiconductor devices from macro to micro up to nano region, there is an urgent need for exploration and investigation of different models applicable in this field of study. In this work, efforts will be geared towards addressing this lacuna by applying a novel numerical technique, the Conservative Finite Difference Method to the modelling and simulation of frequency-dependent dielectric function in plasmonic and small-scale device applications for the next generation. This research comprises analytical derivations, ab initio modelling as well as computer simulations. The approaches adopted in this study include analytical and theoretical formulations, Standard Finite Difference Method, and Conservative Finite Difference Method CFDM for the simulation of devices. The merit of using a conservative numerical scheme such as CFDM is that it preserves the original properties of the differential operator involved in discretization, Computer simulations using MATLAB and other software tools will be presented in evaluating the developed models.
SHOEWU OLUWAGBEMIGA is a Associate Professor / Reader at the Department of Electronics and Computer Engineering
SHOEWU has a Ph.D in Electrical Engineering Electronics and Communications Engineering from Department of E;lectrical/Electronics Engineering, Michael Okpara University of Agriculture, Umudike
