Self-calibration of wireless sensor networks using adaptive filtering techniques

This paper presents an approach for the relative self-calibration of a wireless sensor that utilizes both acoustic and RF signals through adaptive filtering techniques. By integrating an Extended Kalman Filter (EKF) and a complementary filter, we developed an algorithm that utilizes the Time Difference of Arrival (TDOA) and Direction of Arrival (DOA) measurements to enhance sensor node localization. Our method employs a mobile beacon to transmit synchronized acoustic and RF signals, allowing the calculation of precise distances and angles between the beacon and the sensor nodes. The EKF, which is known for its robustness in noisy environments, excels in accurately estimating node positions and orientations, whereas the complementary filter offers computational simplicity and efficiency, particularly under low-noise conditions. The simulation results demonstrate the superior performance of the EKF in terms of convergence speed and accuracy under various noise levels compared to the complementary filter's effectiveness in rapid processing scenarios. This study highlights the importance of selecting appropriate filtering techniques based on specific operational requirements, contributing to the advancement of wireless sensor network technologies, and paving the way for practical applications in autonomous systems, urban planning, and smart city implementations. • Novel algorithm integrates EKF and complementary filter for sensor localization. • Utilizes acoustic and RF signals for accurate node calibration. • Demonstrates EKF's superior performance in noisy environments. • Highlights complementary filter's efficiency in low-noise conditions. • Advances self-calibration techniques for autonomous systems and smart cities.