Sensing information-assisted superimposed pilot-based channel estimation for UAV communication systems

The pervasive integration of unmanned aerial vehicles (UAVs) into demanding industrial scenarios, including power line inspection and mine hoisting, poses critical challenges for next-generation wireless networks in ensuring robust connectivity and high spectral efficiency under rapidly time-varying channel conditions. Although superimposed pilot schemes offer a promising solution to improve spectral efficiency by sharing time-frequency resources, these methods inevitably introduce severe pilot-data mutual interference. This interference degrades channel estimation accuracy and symbol detection reliability, thereby threatening mission-critical UAV operations. To tackle this issue, a sensing information-assisted superimposed pilot channel estimation method is proposed in UAV orthogonal frequency division multiplexing systems. In the proposed method, high-precision kinematic parameters, including position and velocity, acquired from onboard UAV sensing receivers are exploited to derive deterministic, enhanced prior bounds in the delay and Doppler domains. Based on the sensed prior information, a two-stage delay-Doppler denoising scheme is designed to truncate data symbol interference via adaptive thresholding. Subsequently, a sensing-assisted iterative decision-directed mechanism is employed to refine the estimation accuracy. Simulation results demonstrate that the proposed method eliminates the severe error floors of conventional superimposed pilot-based channel estimation methods. Furthermore, it achieves rapid convergence and optimal symbol detection performance at a low pilot power ratio, effectively improving spectral efficiency in dynamic UAV communication scenarios.