Implementing Wireless Sensor Network Through Machine Learning Techniques

Introduction: A regional WSN of independent SNs collects environmental and physical data. Sensors for specific applications measure light, pressure, temperature, humidity, etc. WSNs can transmit real-time data without infrastructure via radio frequencies. Smart cities, agriculture, industrial automation, environmental monitoring, and healthcare use WSN data. Many SNs monitor the network architecture's physical environment. Methods: This design lets SNs talk to neighbors and base stations. Due to power limits, WSNs must prioritise energy efficiency. ML enhances data processing, energy efficiency, and WSN performance. WSN sensors assist ML in predicting events. Machine learning improves data compression, anomaly detection, adaptive network management, WSN predictive modelling, industrial WSN performance, energy efficiency, data analysis, etc. In this work, the dataset, which has parameters such as SN’s positions, energy, distance from cluster head, and SN lifetime is provided as an input to the following ML models Linear Regression, Random Forest, SVM, Naive Bayes, PCA, K Nearest Neighbour, XG Boost, and NN. Results and Discussion: The result showed that the Linear Regression, Random Forest, SVM, Naive Bayes Classifier, PCA, K Nearest Neighbour, and XG Boost were all examined on the training and testing data. Conclusion: The training data accuracies are as follows: Linear Regression (46.64%), Random Forest (99.67%), XG Boost (99.99%), SVM (32.10%), K Nearest Neighbor (98.56%), Naïve Bayes Classifier (96.45%), Principle Component Analysis (97.88%) and NN (34.86%). The testing data accuracies are as follows: Linear Regression (22.00%), Random Forest (97.18%), XG Boost (96.94%), SVM (21.75%), K Nearest Neighbor (97.66%), Naïve Bayes Classifier(64.49%), PCA(99.38%) and NN(28.94%). Linear Regression, Random Forest, SVM, Naive Bayes Classifier, PCA, K Nearest Neighbour, and XG Boost were all examined.