A Geometry-Based Stochastic Wireless Channel Model using Channel Images

Due to the high complexity of geometry-deterministic wireless channel modeling and the difficulty in its implementation, geometry-based stochastic channel modeling (GBSM) approaches have been used to evaluate wireless systems. This paper introduces a new method to model any GBSM by training a generative neural network using images formed by channel parameters. In this work, we obtain channel parameters from the ray-tracing simulation in a specific area and process them in the form of images to train any generative model. Through a case study, we confirm that the use of channel images completes the training of the generative model within 10 epochs. In addition, we show that the trained generative model based on channel images faithfully represents the distributions of the original data under the assigned conditions. Therefore, we argue that the proposed data-to-image methods will facilitate the modeling of GBSM using any generative neural network under general wireless conditions.