Physical layer security communication method based on heterogeneous graph learning

The invention provides a physical layer security communication method based on heterogeneous graph learning. The method comprises the following steps: constructing a mathematical model of a multi-input single-output (MISO) network system, and obtaining network topology data and channel state data of the MISO network system; constructing a heterogeneous graph neural network based on the network topology data and the channel state data of the MISO network system; and training the heterogeneous graph neural network to obtain a trained heterogeneous graph neural network, inputting data of the to-be-processed MISO network into the trained heterogeneous graph neural network, and outputting a transmission scheme of the to-be-processed MISO network to achieve physical layer security by the trained heterogeneous graph neural network. The method provided by the invention can utilize the advantages of the heterogeneous graph neural network, avoids the problem of too slow operation in the traditional method, can capture