Electronic device for processing input signal of power amplifier and operation method thereof

Patent metadata

Field	Value
Publication number	US-2022385317-A1
Application number	US-202017755517-A
Country	US
Kind code	A1
Filing date	Jun 2, 2020
Priority date	Oct 29, 2019
Publication date	Dec 1, 2022
Grant date	—

Abstract

Provided is a method of processing an input signal of an amplifier in an electronic device, the method including obtaining a pre-distorter configured to pre-distort an input signal of the amplifier by using a pretrained neural network model to pre-distort the input signal of the amplifier based on signals input to and output from the amplifier, which are obtained while the amplifier operates in a plurality of different environments, and a plurality of pieces of environmental information corresponding to the plurality of different environments, obtaining an input signal for the amplifier, obtaining information about an environment of the amplifier, pre-distorting the input signal by using the pre-distorter based on the obtained environmental information to prevent an output signal in response to the input signal to be processed by the amplifier from being distorted, and inputting the pre-distorted input signal to the amplifier.

First claim

1 . A method of processing an input signal of an amplifier in an electronic device, the method comprising:
obtaining a pre-distorter configured to pre-distort an input signal of the amplifier by using a pretrained neural network model to pre-distort the input signal of the amplifier based on signals input to and output from the amplifier, which are obtained while the amplifier operates in a plurality of different environments, and a plurality of pieces of environmental information corresponding to the plurality of different environments; obtaining an input signal for the amplifier; obtaining information about an environment of the amplifier; pre-distorting the input signal by using the pre-distorter based on the obtained environmental information to prevent an output signal in response to the input signal to be processed by the amplifier from being distorted; and inputting the pre-distorted input signal to the amplifier.

2 . The method of claim 1 , wherein the amplifier is an amplification unit including at least one amplifier configured to process at least one input signal and output at least one output signal, and
wherein the neural network model is a model trained in advance based on training data including a plurality of sets of at least one input signal input to the amplification unit obtained in the plurality of different environments, at least one output signal of the amplification unit in response to the at least one input signal, and environmental information corresponding to the different environments, such that the at least one input signal is output from the neural network model when the at least one output signal and the environmental information are input to the neural network model.

3 . The method of claim 2 , wherein, when a signal pre-distorted by the pre-distorter is input to the amplification unit including a plurality of amplifiers and a plurality of signals are output through the plurality of amplifiers in the electronic device, the neural network model is a model trained in advance based on training data including a plurality of sets, each having one input signal_input to the amplification unit obtained in the plurality of different environments, a representative value of a plurality of output signals of the amplification unit in response to the one input signal, and environmental information corresponding to the different environments, such that the input signal is output from the neural network model when the representative value and the environmental information are input to the neural network model.

4 . The method of claim 2 , wherein, when N signals pre-distorted by the pre-distorter are input to the amplification unit including M amplifiers and N signals are output through the M amplifiers in the electronic device, the neural network model is a model trained in advance based on training data including a plurality of sets, each set having N input signals input to the amplification unit obtained in the plurality of different environments, N output signals of the amplification unit in response to the N input signals, and environmental information corresponding to the different environments, such that the N input signals are output from the neural network model when the N output signals and the environmental information are input to the neural network model.

5 . The method of claim 1 , wherein the amplifier is an amplification unit including at least one amplifier configured to process at least one input signal and output at least one output signal, and
wherein the neural network model is a model trained in advance based on training data including a plurality of pairs of learning models in which operations of the amplification unit corresponding to the different environments are modeled, which are obtained according to operations of the amplification unit performed in the plurality of different environments, and environmental information corresponding to the different environments, such that at least one output signal of the modeled learning model is equal to at least one input signal when the at least one input signal and the environmental information are processed and output by the neural network model and the modeled learning model in sequence.

6 . The method of claim 5 , wherein, when a signal pre-distorted by the pre-distorter is input to the amplification unit including a plurality of amplifiers and a plurality of signals are output through the plurality of amplifiers in the electronic device, the neural network model is a model trained in advance based on the training data including a plurality of pairs of learning models in which operations of the amplification unit are modeled, which are obtained according to operations of the amplification unit performed in the plurality of different environments, and environmental information corresponding to the different environments, such that a representative value of a plurality of output signals of the modeled learning model is equal to an input signal when the input signal and the environmental information are processed and output by the neural network model and the modeled learning model in sequence.

7 . The method of claim 5 , wherein, when N signals pre-distorted by the pre-distorter are input to the amplification unit including M amplifiers and N signals are output through the M amplifiers in the electronic device, the neural network model is a model trained in advance based on the training data including a plurality of pairs of learning models in which operations of the amplification unit are modeled, which are obtained according to operations of the amplification unit performed in the plurality of different environments, and environmental information corresponding to the different environments, such that N output signals of the modeled learning model are equal to N input signals when the N input signals and the environmental information are processed and output by the neural network model and the modeled learning model in sequence.

8 . An electronic device for processing an input signal of an amplifier, the electronic device comprising:
at least one processor configured to obtain a pre-distorter configured to pre-distort an input signal of the amplifier by using a neural network model trained in advance to pre-distort the input signal of the amplifier based on signals input to and output from the amplifier, which are obtained while the amplifier operates in a plurality of different environments, and a plurality of pieces of environmental information corresponding to the plurality of different environments, obtain an input signal for the amplifier, obtain information about an environment of the amplifier, and pre-distort the input signal by using the pre-distorter based on the obtained environmental information to prevent an output signal in response to the input signal to be processed by the amplifier from being distorted, and input the pre-distorted input signal to the amplifier; and a communication module configured to wirelessly transmit a signal output from the amplifier.

9 . The electronic device of claim 8 , wherein the amplifier is an amplification unit including at least one amplifier configured to process at least one input signal and output at least one output signal, and
wherein the neural network model is a model trained in advance based on training data including a plurality of sets, each set including at least one input signal input to the amplification unit obtained in the plurality of different environments, at least one output signal of the amplification unit in response to the at least one input signal, and environmental information corresponding to the different environments, such that the at least one input signal is output from the neural networ

Assignees

• Samsung Electronics Co Ltd
• Univ Yonsei Iacf

Inventors

• Jung Sunghoon
• Kim Jaebum
• Ku Jungwoo
• Woo Youngyoon
• Kim Yeonghwan
• Lee Chungyong

Classifications

• G06N3/065

  Analogue means · CPC title

• G06N3/045

  Combinations of networks · CPC title

• H03F3/195

  in integrated circuits · CPC title

• H04B2001/0425

  with linearisation using predistortion · CPC title

• H03F1/3241 Primary

  using predistortion circuits (H03F1/3211, H03F1/3217 take precedence) · CPC title