Antenna design with traditional numerical methods is a highly computationally expensive task. In this letter, a complex-valued graph neural network (GNN) with residual connections is proposed to efficiently predict radiation pattern for antenna arrays with different geometric structures. The model uses GNN as backbone, and the topological structure and excitation of antenna array are encoded into graph representation. Complex weights and residual connections are designed to improve the accuracy of the model. Experiments are conducted on five kinds of arrays under amplitude-phase control. The experimental results show that the average mean absolute percentage error (MAPE) and the average root mean square error (RMSE) of the model on five types of antenna array are 1.86% and 0.032 respectively, while the time efficiency of the model is 6 orders of magnitude faster than that of traditional full-wave simulation.

A novel two-dimension metamaterial structure with center symmetric ``T" shaped unit is designed, fabricated and measured in this paper. The proposed metamaterial comprises of a 2×2 array of area 4.3 mm×4.3 mm, etched on single side of FR4 sub-strate. The simulation and measurement show that the novel metamaterial has a double negative range frequency 20.16 GHz - 37.47 GHz with absolute width of 17.31 GHz, relative width of 60.07 %, and small unit loss of 0.16 dB, small size of 0.07 $\lambda_0$ in two dimensions. Furthermore, the double negative region of the pro-posed metamaterial operates through K band and Ka band, it can be widely used in satellite communication and broadband digital transmission.