A 3-D Ray Inversion Method for Regional Channel Impulse Response Restoration

A deep understanding of the Channel Impulse Response (CIR) is crucial for the design and optimization of modern communication systems. CIR not only provides a comprehensive description of channel characteristics but also plays a key role in various aspects, including channel estimation, system design, and Multiple-Input Multiple-Output (MIMO) system development. Traditional channel modeling and measurement methods for obtaining regional CIR typically require point-by-point scanning of the region, which significantly increases time and equipment overhead. In this paper, we propose a three-dimensional (3-D) Ray Inversion (RI) method based on the Image Method (IM) in Ray Tracing (RT) method. The RI method does not require knowledge of the geometric information of the surrounding environment and can reconstruct regional CIR based on a limited number of observed CIR. The RI method discretizes the spatial domain into tetrahedral elements, where the CIR at the nodal points is the observed data. The multipath components of the observed CIR are classified first, and then the location of the secondary wave sources are determined from the ray propagation paths, thereby reconstructing the regional CIR within the tetrahedrons. To optimize computational efficiency while maintaining solution accuracy, an adaptive refinement algorithm based on the longest-edge bisection criterion is implemented to ensure the generated subtetrahedra have small and uniform edge lengths. Experimental results show that the RI method maintains high accuracy in both indoor and outdoor scenarios, and the computational efficiency is improved by more than 20 times compared to the IM method, and the efficiency gains are greater as the sampling distance decreases.