A Dual-Kernel Computational Framework for Artificial Intelligence-Based Indoor Communication Analysis

For indoor communication of next-generation wireless networks, the efficient and realistic analyzing tool is in an urgent demand. Artificial intelligence (AI) methods can be employed with simulated or real-world data to provide effective analysis of wireless channel characteristics in a computational way. However, those methods need to prepare a big valid data for model training. It limits the efficiency improvement of AI methods. Here, a dual-kernel AI framework is proposed to reduce the amount of reference data. In order to verify the validity of the proposed framework, a specific scenario is investigated in the framework. Results obtained show that the proposed framework reduces the demand for data preparation, achieved the purpose of improving the calculation efficiency. The proposed AI framework is useful for the numerical analysis of indoor wireless communication.