In mobile communication systems, estimating radio propagation characteristics, i.e., signal reception quality, from a base station (BS) to user equipment (UE) with fast and high accuracy can contribute to the optimal placement of base stations, rapid improvements in coverage quality through the dynamic optimization of base station parameters, and reduced power consumption across the entire system. In particular, for the operation of 6G systems capable of meeting various stringent requirements, e.g., extreme high data rate/capacity, extreme high reliability, extreme low energy, and extreme coverage extension, there is a need for faster and more accurate methods for estimating radio wave propagation characteristics when optimizing a vast number of base stations utilizing various frequencies. Until now, the ray tracing method (RTM) has been widely used to estimate radio wave propagation characteristics; however, this method has issues of computation time and estimation accuracy. For example, when evaluating the radio wave propagation characteristics of a single base station covering a 1 km² area in an urban environment, the computation time is more than ten hours even using a high-performance PC.

In this paper, we propose the Color Images Method (CIM) as a fast and high accuracy method for estimating radio wave propagation characteristics based on image processing. In the CIM, to identify the building walls where radio waves scatter, we first assign different RGB colors to all building walls within the evaluation area, then create two RGB images viewed from the transmitter (BS) and receiver points (UE) and compare their colors. If a wall with the corresponding color exists in both images, we identify that wall as the one where the radio waves scatter. Next, the number of pixels on that wall is counted. Since this number of pixels corresponds to the visible area of the wall based on the distance from the transmitter/receiver point to the wall and the wall’s orientation, the radio waves scattering intensity caused by that wall can be calculated by multiplying the number of pixels by the transmit power and antenna gain. Furthermore, based on the relationships among the transmitter/receiver positions, the wall position, and the pixel positions in the images, it is possible to calculate the delay time and angle information of the propagation path. In this paper, to ensure general applicability, we compared the computational time and estimation accuracy of the CIM with those of the RTM, using actual measurement results of propagation losses in different outdoor environments and at various frequencies. We confirmed that, compared to the RTM, the CIM reduces computational time to less than one-hundredth of that required by the RTM and improves estimation accuracy by more than 7 dB in terms of the standard deviation of the estimation error.

Thus, this paper addresses the challenge of achieving the fast and highly accurate estimation of radio propagation characteristics for mobile communication systems by introducing a novel perspective based on image processing. It clearly demonstrates a promising direction toward solving this problem and can be highly regarded as a paper well deserving of the Society’s Best Paper Award.