Single Mobile Base Station Positioning Algorithm Designed for Disaster Emergency Communication
Han Mei, Xiaofeng Zhong, Shidong Zhou, Jie Wei · 2025
In disaster relief operations, both emergency communication and personnel positioning are critical. This paper proposes an enhanced unscented Kalman filter (UKF) positioning algorithm for communication terminals, utilizing random access signals (RACH) and demodulated reference signals (DM-RS). This method can accurately determine the location of on-site terminals without interfering with emergency communication. The algorithm utilizes TOA and AOA two-dimensional observations for unscented Kalman filter. The severity of non-line-of-sight (NLOS) effects at the current location can be quantified by comparing the residual, which are difference between measured and predicted values, with the standard deviation of the measurement data. According to the severity, the residual value undergoes different processing, and then determines whether to reset Kalman filtering. The simulation results demonstrate that under complete NLOS conditions, the proposed coarse estimation method achieves an average root mean square error (RMSE) of 52 meters after 360 time steps (approximately 7 seconds) of measurement processing, representing a 27.3% improvement in positioning accuracy compared to the recent UKF-based approach. Meanwhile, when both coarse and precise estimations are combined, the average RMSE reaches 14 meters, improving performance by 81.3% compared to the recent UKF method.