Enhancing the Prediction of Underwater Wireless Communication using Machine Learning Techniques and Related Issues
M. Saravanan, V. Jagan Naveen, Siddhartha Kumar, Dilli Ganesh, S Sivashankar, K. Vijaya Bhaskar
Abstract
This research aims to enhance the prediction of underwater wireless communication for underwater wireless sensor networks using Logistic Regression compared with Linear Regression Algorithms. To forecast the functionality of underwater wireless sensor networks Linear Regression and Novel Logistic Regression algorithms were used with different training and testing splits with Underwater Acoustic Wireless Sensor. A sample size of 20 (Each 10 from Group 1 and Group 2) is calculated by fixing a G-power of 0.8, alpha and beta values of 0.05 and 0.2, and a confidence interval of 95%. With a statistical significance value of 0.001 (p<0.05), Novel Logistic Regression (87.77% & 12.22%) outperforms Linear Regression (80.17% & 19.32%) in terms of accuracy and Loss which is statistically significant. The Underwater Wireless Communication efficiency were compared with the obtained results for Linear Regression, Logistic Regression proves to have higher accuracy.