摘要
The terahertz( THz) frequency band is a key candidate frequency band for the sixth generation( 6G) of mobile communication. Its continuously available bandwidth can address the contradiction between the rapidly growing service demands and the shortage of spectrum resources. The potential application scenarios of THz communication systems were comprehensively summarized in this paper. Regardless of the application scenario,the wireless channel constitutes a fundamental component for the development,performance evaluation,and optimization of THz communication systems. Therefore,an in-depth analysis of the electromagnetic wave propagation characteristics in the THz band and their implications for channel behavior was conducted in this paper. Furthermore,state-of-the-art methodologies for THz channel modeling are summarized,including artificial intelligence( AI)-based predictive modeling and map-based hybrid modeling approaches. The significance of deterministic channel modeling techniques,particularly those employing ray-tracing( RT),is also discussed in detail. Based on this,a novel joint error-corrected electromagnetic parameter extraction method using a hybrid flower pollination algorithm-neural network( FPA-NN) framework is proposed. To validate the FPA-NN,an indoor office environment is modeled and simulated,with the resulting data utilized to characterize the channel properties. These findings provide valuable inputs for high-precision modeling of the THz band.
