Information transmission method and apparatus
This application provides an information transmission method and apparatus. The method includes: receiving first indication information from a network device, where the first indication information indicates an association relationship between a plurality of branch networks in an artificial intelligence AI network and channel state information CSI measurement configuration information; and obtaining first quantization information based on a first branch network and channel information, where the first branch network is associated with current CSI measurement configuration information, and the first branch network belongs to the plurality of branch networks in the AI network.
1 . A method, comprising:
receiving first indication information, wherein the first indication information indicates an association relationship between a plurality of branch networks in an artificial intelligence (AI) network and channel state information (CSI) measurement configuration information; and
obtaining first quantization information based on a first branch network and channel information, wherein the first branch network is associated with current CSI measurement configuration information, and the first branch network belongs to the plurality of branch networks in the AI network.
2 . The method according to claim 1 , wherein obtaining the first quantization information based on the first branch network and the channel information comprises:
obtaining a first matrix based on the first branch network and the channel information; and
obtaining the first quantization information based on a first quantization network and the first matrix, wherein the first quantization network is associated with the current CSI measurement configuration information or the first branch network.
3 . The method according to claim 1 , wherein the first indication information further indicates a quantization manner, and the quantization manner comprises a quantization direction or a quantization dimension, and wherein obtaining the first quantization information based on a first quantization network and a first matrix comprises:
processing the first matrix by using the first quantization network in the quantization manner, to obtain the first quantization information.
4 . The method according to claim 1 , wherein the first quantization information comprises second quantization information and third quantization information, and obtaining the first quantization information based on a first quantization network and a first matrix comprises:
dividing the first matrix into a second matrix and a third matrix in a quantization manner;
obtaining the second quantization information based on the first quantization network and the second matrix; and
obtaining the third quantization information based on a second quantization network and the third matrix, wherein the second quantization network is associated with the current CSI measurement configuration information or the first branch network.
5 . The method according to claim 1 , wherein obtaining the first quantization information based on a first quantization network and a first matrix comprises:
obtaining a fourth matrix based on a second branch network and the first matrix, wherein the second branch network is associated with the current CSI measurement configuration information, and the second branch network belongs to the plurality of branch networks in the AI network; and
obtaining the first quantization information based on the fourth matrix.
6 . The method according to claim 5 , wherein obtaining the first quantization information based on the fourth matrix comprises:
obtaining the first quantization information based on a third quantization network and the fourth matrix, wherein the third quantization network is associated with the current CSI measurement configuration information or the second branch network.
7 . The method according to claim 1 , wherein the method further comprises:
sending the first quantization information.
8 . The method according to claim 7 , wherein sending the first quantization information comprises:
sending the first quantization information in a preconfigured sequence.
9 . The method according to claim 8 , wherein the preconfigured sequence comprises:
a line before column sequence, a column before line sequence, or a sequence of identifiers (IDs) of the branch networks.
10 . A method, comprising:
sending first indication information, wherein the first indication information indicates an association relationship between a plurality of branch networks in an artificial intelligence (AI) network and channel state information (CSI) measurement configuration information; and
receiving first quantization information, wherein the first quantization information is obtained based on a first branch network and channel information, the first branch network is associated with current CSI measurement configuration information, and the first branch network belongs to the plurality of branch networks in the AI network.
11 . The method according to claim 10 , wherein the first quantization information is obtained based on a first quantization network and a first matrix, the first matrix is obtained based on the first branch network and the channel information, and the first quantization network is associated with the current CSI measurement configuration information or the first branch network.
12 . The method according to claim 10 , wherein the first indication information further indicates a quantization manner, and the quantization manner comprises a quantization direction or a quantization dimension, and the first quantization information is obtained based on a first quantization network, the quantization manner, and the first matrix.
13 . The method according to claim 10 , wherein the first quantization information comprises second quantization information and third quantization information, and the second quantization information is obtained based on a first quantization network and a second matrix, the third quantization information is obtained based on a second quantization network and a third matrix, the second matrix and the third matrix are combined to form the first matrix, and the second quantization network is associated with the current CSI measurement configuration information or the first branch network.
14 . An apparatus, comprising:
a transceiver configured to receive first indication information, wherein the first indication information indicates an association relationship between a plurality of branch networks in an artificial intelligence (AI) network and channel state information (CSI) measurement configuration information;
one or more processors; and
a non-transitory computer-readable storage medium storing a program to be executed by the one or more processors, the program including instructions to:
obtain first quantization information based on a first branch network and channel information, wherein the first branch network is associated with current CSI measurement configuration information, and the first branch network belongs to the plurality of branch networks in the AI network.
15 . The apparatus according to claim 14 , wherein instructions to obtain the first quantization information include instructions to:
obtain a first matrix based on the first branch network and the channel information; and
obtain the first quantization information based on a first quantization network and the first matrix, wherein the first quantization network is associated with the current CSI measurement configuration information or the first branch network.
16 . The apparatus according to claim 15 , wherein the first indication information further indicates a quantization manner, the quantization manner comprises a quantization direction or a quantization dimension, and the instructions to obtain the first quantization information include instructions to:
process the first matrix by using the first quantization network in the quantization manner, to obtain the first quantization information.
17 . The apparatus according to claim 16 , wherein the first quantization information comprises second quantization information and third quantization information, and the instructions to obtain the first quantization information include instructions to:
divide the first matrix into a second matrix and a third matrix in the quantization manner; and
obtain the second quantization information based on the first quantization network and the second matrix, and obtain the third quantization information based on a second quantization network and the third matrix, wherein the second quantization network is associated with the current CSI measurement configuration information or the first branch network.
18 . The apparatus according to claim 14 , wherein the first quantization information is obtained based on a first quantization network and a first matrix, and the instructions to obtain the first quantization information include instructions to:
obtain a fourth matrix based on a second branch network and the first matrix, wherein the second branch network is associated with the current CSI measurement configuration information, and the second branch network belongs to the plurality of branch networks in the AI network; and
obtain the first quantization information based on the fourth matrix.
19 . The apparatus according to claim 18 , wherein the instructions to obtain the first quantization information include instructions to:
obtain the first quantization information based on a third quantization network and the fourth matrix, wherein the third quantization network is associated with the current CSI measurement configuration information or the second branch network.
20 . The apparatus according to claim 14 , wherein the transceiver is further configured to send the first quantization information.