AI-Enabled Data-Driven Channel Modeling for Future Communications
Mi Yang, Ruisi He, Bo Ai, Huang Chen, Chenlong Wang, Yuxin Zhang, Zhangdui Zhong
Abstract
Wireless channel modeling plays an essential role in the design of wireless communication networks. Especially the future integrated network with various applications and extended-spectrum needs an accurate channel model as the cornerstone. However, with the expansion of scenarios, frequencies, and user requirements, classical channel modeling methods face many limitations, and new approaches must be explored. Artificial intelligence (AI) has become one of the key technologies in the evolution of wireless communication systems. Recent research has applied AI technology to predicting channel characteristics such as path loss. This paper presents an AI-enabled channel modeling framework for future communication networks, which aims to establish a nonlinear model between environmental information and channel characteristics. Then, this paper expounds on the proposed framework ' s architecture and features and analyzes some key technologies involved. At last, technical challenges are pointed out in this paper.