Comparative Study of Hybrid Machine Learning Models for Indoor Localization Using SVM, GPR, and GCN with Augmented Techniques

Indoor positioning technology is vital for numerous applications, from intelligent building systems to emergency services. This research presents an innovative hybrid approach to indoor localization, combining Support Vector Machines (SVM), Gaussian Process Regression (GPR), and Graph Convolutional Networks (GCNs). We enhance these models' effectiveness through advanced data augmentation methods, including the use of Generative Adversarial Networks (GANs). Our methodology is tested using the UJI Indoor Localization dataset, demonstrating notable enhancements in positioning precision. The GCN-augmented technique, in particular, yields superior performance metrics compared to other approaches. This study provides insights into the practical applicability and generalization capabilities of our hybrid model in real-world indoor environments.