Electronic device for feeding back channel state information by using autoencoder based on modified split learning, and method of operating the same
A method of operating a base station, includes: transmitting a reference signal to an electronic device comprising a first model and a third model; receiving a first intermediate output from the electronic device; obtaining a second intermediate output by inputting the first intermediate output to a partial model excluding an output layer from a second model comprising a second neural network for data decompression; transmitting the second intermediate output to the electronic device; receiving a first gradient value from the electronic device and updating weight parameters of the partial model; and generating a second gradient value different from the first gradient value and transmitting the second gradient value to the electronic device.
1 . A method of operating an electronic device, the method comprising:
obtaining a channel estimation result, based on a reference signal received from a base station;
obtaining a first intermediate output by inputting the channel estimation result to a first model comprising a first neural network for data compression;
transmitting, to the base station, the first intermediate output;
receiving, from the base station, a second intermediate output different from the first intermediate output;
obtaining a decompressed channel estimation result by inputting the second intermediate output to a third model corresponding to an output layer of a second model comprising a second neural network for data decompression;
calculating a loss between the obtained channel estimation result and the decompressed channel estimation result by comparing the obtained channel estimation result with the decompressed channel estimation result;
transmitting, to the base station, a first gradient value for minimizing the loss;
receiving, from the base station, a second gradient value different from the first gradient value; and
updating the first model based on the second gradient value.
2 . The method of claim 1 , further comprising:
identifying that a difference between the channel estimation result and the decompressed channel estimation result is less than or equal to a threshold value; and
based on the identifying that the difference between the channel estimation result and the decompressed channel estimation result is less than or equal to the threshold value, transmitting the third model to the base station.
3 . The method of claim 1 , wherein the base station comprises the second model.
4 . The method of claim 3 , wherein the second intermediate output corresponds to an output of the second model of the base station.
5 . The method of claim 3 , wherein the second gradient value is generated based on back-propagation of the second model of the base station.
6 . The method of claim 1 , wherein an encoder corresponding to the first model corresponds to at least one of a convolution neural network (CNN), a recurrent neural network (RNN), a stacking-based deep neural network (S-DNN), and a long short-term memory (LSTM) network.
7 . The method of claim 1 , wherein a decoder concatenating the second model with the third model corresponds to at least one of a convolution neural network (CNN), a recurrent neural network (RNN), a stacking-based deep neural network (S-DNN), and a long short-term memory (LSTM) network.
8 . A method of operating a base station, the method comprising:
transmitting a reference signal to an electronic device comprising a first model and a third model;
receiving a first intermediate output from the electronic device;
obtaining a second intermediate output by inputting the first intermediate output to a partial model excluding an output layer from a second model comprising a second neural network for data decompression;
transmitting the second intermediate output to the electronic device;
receiving a first gradient value from the electronic device and updating weight parameters of the partial model; and
generating a second gradient value different from the first gradient value and transmitting the second gradient value to the electronic device.
9 . The method of claim 8 , wherein the first model comprises a first neural network for data compression, and
wherein the third model corresponds to the output layer of the second model.
10 . The method of claim 9 , wherein the first intermediate output corresponds to data output by the first model using, as an input, a channel estimation result measured based on the reference signal.
11 . The method of claim 9 , wherein the first gradient value is generated based on a back-propagation algorithm to minimize a loss calculated by comparing a channel estimation result measured by the electronic device based on the reference signal with a decompressed channel estimation result output by the third model using the second intermediate output as an input.
12 . The method of claim 11 , further comprising receiving the third model from the electronic device,
wherein the third model is transmitted based on identifying that a difference between the channel estimation result and the decompressed channel estimation result is less than or equal to a threshold value.
13 . The method of claim 8 , wherein an encoder corresponding to the first model corresponds to at least a convolution neural network (CNN), a recurrent neural network (RNN), a stacking-based deep neural network (S-DNN), and a long short-term memory (LSTM) network.
14 . The method of claim 8 , wherein a decoder concatenating the second model with the third model corresponds to at least one of a convolution neural network (CNN), a recurrent neural network (RNN), a stacking-based deep neural network (S-DNN), and a long short-term memory (LSTM) network.
15 . A method of operating a wireless communication system comprising an electronic device comprising an encoder for data compression and a base station comprising a decoder for data decompression, the method comprising:
transmitting, by the base station, a reference signal to the electronic device;
obtaining, by the electronic device, a channel estimation result based on the reference signal;
inputting, by the electronic device, the channel estimation result to the encoder and obtaining an intermediate output;
transmitting, by the electronic device, the obtained intermediate output to the base station; and
inputting, by the base station, the intermediate output to the decoder, and obtaining, by the base station, a decompressed channel estimation result,
wherein the method further comprises:
training, by the electronic device, the encoder, and
training, by the electronic device, an output layer of the decoder.
16 . The method of claim 15 , further comprising training, by the base station, the remaining layers of the decoder other than the output layer of the decoder.
17 . The method of claim 15 , further comprising transmitting the output layer of the decoder to the base station based on a completion of the training of the output layer of the decoder.
18 . The method of claim 15 , wherein the encoder corresponds to at least one of a convolution neural network (CNN), a recurrent neural network (RNN), a stacking-based deep neural network (S-DNN), and a long short-term memory (LSTM) network.
19 . The method of claim 15 , wherein the decoder corresponds to at least one of a convolution neural network (CNN), a recurrent neural network (RNN), a stacking-based deep neural network (S-DNN), and a long short-term memory (LSTM) network.