Abstract—The high-precision Indoor Positioning System (IPS) is a captivating area of research that has made significant advancements in recent years due to the increasing demand for its applications. Our study proposes an innovative approach to improve indoor positioning accuracy by integrating Principal Component Analysis (PCA), weighted k-nearest Neighbors (WKNN), and Linear Regression (PCAWLR). This hybrid strategy enables the system to leverage the unique characteristics of each model, capturing intricate patterns and correlations in the data. Experimental evaluations on a publicly available dataset demonstrate the superiority of our hybrid approach. The Root Mean Squared Error (RMSE) achieved is 1.97 meters, and the mean distance error is 2.23 meters. Remarkably, the ensemble outperforms individual methods in other studies on the same dataset, showing 10.8% to 17.2% improvement in accuracy. Notably, our proposed hybrid approach significantly reduces training time from 581.3599 seconds to 8.8814 seconds, representing an impressive reduction of approximately 98.47%. Similarly, testing time is reduced from 10.1721 seconds to 0.0176 seconds, indicating a substantial decrease of around 99.82%. These significant reductions in training and testing times underscore the efficiency and effectiveness of our proposed ensemble model, making it highly practical for real-time applications.