Leveraging RSSI and RTT for Accurate Distance Prediction in Bluetooth HC-05 with Multivariate Linear Regression Model

A model for the accurate estimation of the distance between Bluetooth devices using a Multivariate Linear Regression (MLR) approach that integrates Received Signal Strength Indicator (RSSI) and Round-Trip Time (RTT) data is presented in this study. Bluetooth technology, specifically the HC-05 module, is employed for wireless communication between devices, where RSSI and RTT serve as independent variables for distance prediction. The present study aims to address the limitations of using these methods separately, as RSSI is susceptible to environmental factors and signal interference, whereas RTT provides a more accurate measurement but often requires more complex calculations. By integrating both methods using an MLR model, a more robust and accurate distance estimate was achieved. The proposed model exhibited a Mean Squared Error (MSE) of 0.0173, indicating a very small average error in distance predictions, while the R-squared (R²) value of 0.9986 demonstrated that the model explained 99.86% of the variance in the actual distance data, highlighting its high accuracy. A Root Mean Squared Error (RMSE) of 0.1316 m, or approximately 13.16 cm, indicates that the model's average prediction error is around 13 cm. This approach significantly improves the reliability of Bluetooth-based localization systems and is highly beneficial for applications that require precise distance measurements.