Intelligent metasurface channel estimation method and system based on model driving

• Main Authors: FENG XIAO, WANG ZHENGHAI, YU LISU, ZHOU HUILIN, SU YANPING
• Format: Patent
• Language: Chinese; English
• Published: 16.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; PHYSICS; TRANSMISSION; TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHICCOMMUNICATION

The invention provides an intelligent metasurface channel estimation method and system based on model driving, and the method comprises the steps: obtaining a plurality of groups of different cascaded channel rough estimation values and real cascaded channel label values, constructing a data pair, obtaining channel known parameters, and building a data set; constructing a deep neural network model, inputting the data set into the deep neural network model, carrying out multi-round iterative optimization training according to a preset loss function, and finally obtaining an optimal deep neural network model and an optimal weight of the optimal deep neural network model; and inputting the initial rough cascade channel estimation value into the optimal deep neural network model to reconstruct an optimal cascade channel. According to the intelligent metasurface channel estimation method and system based on model driving, a physical model is fused, and the generalization ability of an algorithm is improved, so tha