Deep-learning for distributed channel feedback and precoding
Some embodiments of the present disclosure provide a deep-learning-based framework for designing components of a downlink precoding system. The components of such a system include downlink training pilots and channel estimation based on receipt of the downlink training pilots. Another component involves channel measurement and feedback strategy at the user equipment. The components include a precoding scheme designed at the base station based on the feedback from the user equipment.
1. A method comprising:
receiving an instruction, wherein the instruction includes a specification of a first deep neural network for processing reference signals, wherein the specification of the first deep neural network includes an indication of a selected deep neural network from among a set of deep neural networks;
receiving a reference signal;
obtaining a feedback message by processing, using the first deep neural network, the received reference signal, the feedback message configured for use in deriving a precoding matrix;
transmitting the feedback message; and
receiving a data signal, wherein the data signal has been subjected to the precoding matrix, wherein the precoding matrix has been derived using a second deep neural network that has received, as input, the feedback message and feedback messages from other devices.
2. The method of claim 1 wherein the specification of the first deep neural network further comprises an indication of a structure and an indication of a set of coefficients.
3. The method of claim 1 wherein the receiving the instruction employs Radio Resource Control signaling.
4. The method of claim 1 wherein the receiving the instruction employs broadcast signaling.
5. The method of claim 1 wherein the transmitting the feedback message comprises employing a physical uplink shared channel (PUSCH).
6. The method of claim 1 wherein the transmitting the feedback message comprises employing a physical uplink control channel (PUCCH).
7. The method of claim 1 wherein the receiving the data signal comprises employing a physical downlink shared channel (PDSCH).
8. An apparatus comprising:
a memory storing instructions; and
a processor configured, by executing the instructions, to:
receive an instruction, wherein the instruction includes a specification of a first deep neural network for processing reference signals, wherein the specification of the first deep neural network includes an indication of a selected deep neural network from among a set of deep neural networks;
receive a reference signal;
obtain a feedback message by processing, using the first deep neural network, the received reference signal, the feedback message configured for use in deriving a precoding matrix;
transmit the feedback message; and
receive a data signal, wherein the data signal has been subjected to the precoding matrix, wherein the precoding matrix has been derived using a second deep neural network that has received, as input, the feedback message and feedback messages from other devices.
9. The apparatus of claim 8 , wherein the specification of the first deep neural network further comprises an indication of a structure and an indication of a set of coefficients.
10. The apparatus of claim 8 , wherein the processor employs Radio Resource Control signaling to receive the instruction.
11. The apparatus of claim 8 , wherein the processor employs broadcast signaling to receive the instruction.
12. The apparatus of claim 8 , wherein the processor employs a physical uplink shared channel (PUSCH) to transmit the feedback message.
13. The apparatus of claim 8 , wherein the processor employs a physical uplink control channel (PUCCH) to transmit the feedback message.
14. The apparatus of claim 8 , wherein the processor employs a physical downlink shared channel (PDSCH) to receive the data signal.
15. A non-transitory computer-readable medium storing instructions, the instructions, when executed by a processor, causing the processor to:
receive an instruction, wherein the instruction includes a specification of a first deep neural network for processing reference signals, wherein the specification of the first deep neural network includes an indication of a selected deep neural network from among a set of deep neural networks;
receive a reference signal;
obtain a feedback message by processing, using the first deep neural network, the received reference signal, the feedback message configured for use in deriving a precoding matrix; and
transmit the feedback message; and
receive a data signal, wherein the data signal has been subjected to the precoding matrix, wherein the precoding matrix has been derived using a second deep neural network that has received, as input, the feedback message and feedback messages from other devices.
16. The non-transitory computer-readable medium of claim 15 , wherein the specification of the first deep neural network further comprises an indication of a structure and an indication of a set of coefficients.
17. The non-transitory computer-readable medium of claim 15 , wherein the instructions cause the processor to employ Radio Resource Control signaling to receive the instruction.
18. The non-transitory computer-readable medium of claim 15 , wherein the instructions cause the processor to employ broadcast signaling to receive the instruction.
19. The non-transitory computer-readable medium of claim 15 , wherein the instructions cause the processor to employ a physical uplink shared channel (PUSCH) to transmit the feedback message.
20. The non-transitory computer-readable medium of claim 15 , wherein the instructions cause the processor to employ a physical uplink control channel (PUCCH) to transmit the feedback message.
21. The non-transitory computer-readable medium of claim 15 , wherein the instructions cause the processor to employ a physical downlink shared channel (PDSCH) to receive the data signal.