5G Transmission Point Selection Using Map Image Extraction

The patent introduces image information as an intermediary data source that bridges the gap between simple mathematical modeling and complex ray tracing simulations. By extracting reception points from image information (such as building layouts, obstacle positions) and using them to generate transmission point candidates, the system achieves accurate RF signal prediction without requiring full-scale ray tracing computations. This intermediary approach maintains calculation efficiency while improving accuracy for high frequency signals.

The patent performs preliminary extraction of reception points from image information before conducting RF signal analysis. By pre-identifying key locations (building corners, obstacle positions, reception points) from map images, the system creates a simplified geometric model that captures essential channel characteristics. This preliminary action enables subsequent transmission point selection to achieve accurate channel analysis results without requiring time-consuming full ray tracing simulations, thus reducing simulation time while maintaining accuracy.

The patent applies local quality by assigning different priorities to different transmission points based on their specific characteristics and locations. Instead of treating all transmission points uniformly, the system evaluates each candidate's suitability for covering specific reception points and assigns priority levels accordingly. This allows the network to consider multiple transmission points for improved coverage and reliability while simplifying the selection process through priority-based differentiation, reducing the complexity of managing large transmission point groups.