A 5G Adaptive Positioning Method in Environments with Significant Noise Changes

As the demand for positioning grows, precision and reliability are becoming increasingly important. While Global Navigation Satellite Systems (GNSS) offer outdoor positioning, their limitations in urban settings necessitate alternative solutions. 5G positioning emerges as a complementary approach, leveraging disruptive technologies like millimeter-wave and massive MIMO antennas, enabling high precision. Current 5G positioning methods, mainly Extended Kalman Filter (EKF)-based, suffer from time-varying noise in 5G measurements, degrading accuracy. To overcome this, adaptive methods like Adaptive Kalman Filter (AKF) have been introduced, but suboptimal performance persists. This work introduces Variational Bayesian Adaptive Kalman Filter (VBAKF) for 5G positioning. The work includes simulation verification, VBAKF performance evaluation under varying ƛ factors, and comparative analyses with conventional adaptive methods. Simulations reveal VBAKF's superior performance, positioning it as an effective choice for 5G positioning, especially in challenging scenarios.