Deep learning based beamforming method and apparatus

A beamforming control apparatus using a deep neural network, the deep neural network comprising an input layer, L hidden layers (L is a natural number greater than or equal to 1), and an output layer, and the beamforming control apparatus comprising a processor and a memory storing instructions executable by the processor, wherein when executed by the processor, the instructions cause the processor to:

• obtain channel information h between a base station and K terminals (K is a natural number greater than or equal to 1) and a transmit power limit value P of the base station, and input h and P into the input layer;
• and
• perform beamforming on signals to be transmitted to the K terminals using beamforming vectors derived using the output layer and at least one activation function,
• wherein the base station transmits the signals to the K terminals using M transmit antennas (M is a natural number greater than or equal to 1).

A cloud radio access network (C-RAN) system including a centralized processor (CP) and a plurality of remote radio heads (RRHs), the CP including a deep neural network comprising a an input layer, L hidden layers (L is a natural number greater than or equal to 1), and an output layer,

• wherein the CP is configured to:
• obtain channel information h between the plurality of RRHs and K terminals (K is a natural number greater than or equal to 1) and transmit power limit values P of the plurality of RRHs, input h and P into the input layer, derive beamforming vectors for signals to be transmitted to the K terminals by using the output layer and at least one activation function, and deliver the derived beamforming vectors to the plurality of RRHs,
• wherein the plurality of RRHs are configured to:
• perform beamforming on the signals to be transmitted to the K terminals by using the beamforming vectors, and transmit the beamformed signals to the K terminals.