Node Localization Performance Evaluation and Tracking using Kalman Filter in Wireless Sensor Networks

Wireless Sensor Networks (WSNs) are crucial in smart environments where precise node location is necessary for efficient data processing and communication. Present-day localization methods frequently provide an average accuracy of about 11.3%, which degrades network performance as a whole. The proposed research work presents a hybrid method to increase localization accuracy by merging the Kalman Filter (KF) with Random Fourier Feature (RFF) and Support Vector Machine (SVM). By using the RFF for feature mapping and the KF for effective tracking, this technique greatly lowers localization errors. The Mean Absolute Error(MAE) reduced from 7.74 to 3.87 for the first trajectory, 8.41 to 4.31 for the second, and 9.47 to 8.31 for the third trajectory. By improving WSN's accuracy and dependability, the suggested method increases their usefulness in intelligent applications.