# US 2010/0169254 A1 · System and method for calibrating radio frequency power of communication devices

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2010/0169254 A1
- Document ID: US20100169254A1
- Status: Published application
- Kind: A1
- Application number: 12/509,141
- Filed: 2009-07-24
- Published: 2010-07-01
- Earliest claimed: 2008-12-29
- Assignees: Chi Mei Communication Systems, Inc.
- Inventors: YI-JEN SU
- Web page: https://patentgrep.com/patent/US20100169254A1

## CPC

- H03G 3/3042: {in modulators, frequency-changers, transmitters or power amplifiers}
- H03G: CONTROL OF AMPLIFICATION

## Abstract

A radio frequency (RF) calibrating system and a method for calibrating RF power of communication devices are provided. The method collects RF signals transmitted from the communication devices, and generates a group of training samples by retrieving measurement data from the RF signals. The method further constructs a neural network according to the group of training samples, calibrate RF power of the communication devices using the neural network, and generate corresponding calibration results of the RF power. In addition, the method generates a frequency spectrum of the RF power according to the calibration results of the RF power, and displays the frequency spectrum on a display device of the RF calibrating system.

## Claims

### Claim 1

A computing system for calibrating radio frequency (RF) power of communication devices, the computing system comprising:
an RF signal collecting device operable to collect RF signals transmitted from a plurality of communication devices;
an RF calibrating device, comprising:
a training sample generating module operable to receive the RF signals from the RF signal collecting device, and obtain a plurality of measurement data from the RF signals to generate a group of training samples;
a neural network constructing module operable to construct a neural network according to the group of training samples;
a power calibrating module operable to calibrate RF power of the communication devices using the neural network, and generate calibration results of the RF power; and
a calibration result outputting device operable to generate a frequency spectrum of the RF power according to the calibration results, and display the frequency spectrum on a display device.

### Claim 2 (depends on 1)

The computing system according to claim 1, wherein the neural network constructing module is further operable to generate a plurality of nerve cells of the neural network, calculate a weight value of each of the nerve cells according to the group of training samples, and construct the neural network based upon the nerve cells with corresponding weight values.

### Claim 3 (depends on 1)

The computing system according to claim 1, wherein the neural network comprises an input layer, a hidden layer, and an output layer.

### Claim 4 (depends on 3)

The computing system according to claim 3, wherein the power calibrating module calibrates the RF power by transmitting the RF signals through the hidden layer of the neural network.

### Claim 5 (depends on 3)

The computing system according to claim 3, wherein the power calibrating module outputs the calibration results of the RF power via the output layer of the neural network.

### Claim 6 (depends on 1)

The computing system according to claim 1, wherein the frequency spectrum is a curve diagram that represents a relationship between digital-to-analog converter (DAC) power and the RF power.

### Claim 7

A computer-enabled method for calibrating radio frequency (RF) power of communication devices, the method comprising:
(a) collecting a plurality of RF signals transmitted from a plurality of communication devices;
(b) generating a group of training samples by retrieving a plurality of measurement data from the RF signals;
(c) constructing a neural network according to the group of training samples;
(d) calibrating RF power of the communication devices using the neural network;
(e) generating calibration results of the RF power; and
(f) generating a frequency spectrum of the RF power according to the calibration results of the RF power, and displaying the frequency spectrum on a display device.

### Claim 8 (depends on 7)

The method according to claim 7, wherein the frequency spectrum is a curve diagram that represents a relationship between digital-to-analog converter (DAC) power and the RF power.

### Claim 9 (depends on 7)

The method according to claim 7, wherein block (b) comprises:
calculating a normalization value of each of the RF power by performing a normalization operation;
selecting part of the normalization values as input values of the neural network, and obtaining all the normalization values as output values of the neural network; and
generating the group of training samples according to the input values and the output values.

### Claim 10 (depends on 9)

The method according to claim 9, wherein the normalization operation is a operation function that is denoted as Pi=(Pi−Pmin)/(Pmax−Pmin), where “ Pi” is the normalization value of a RF power, “Pi” is a RF power value, “Pmax” is a maximum value of the RF power, and “Pmin” is a minimum value of the RF power.

### Claim 11 (depends on 7)

The method according to claim 7, wherein block (c) comprises:
generating a plurality of nerve cells of the neural network;
calculating a weight value of each of the nerve cells according to the group of training samples; and
constructing the neural network based upon the nerve cells with corresponding weight values.

### Claim 12 (depends on 7)

The method according to claim 7, wherein block (d) comprises:
inputting part of the measurement data from an input layer of the neural network;
outputting target data from an output layer of the neural network after the part of measurement data are transmitted through a hidden layer of the neural network;
calculating calibrated data by performing a reversed normalization operation on the target data; and
generating the calibrated RF power according to the calibrated data.

### Claim 13

A medium having stored thereon instructions that, when executed by a processor of a computing device, cause the computing device to perform a method for calibrating radio frequency (RF) power of communication devices, the method comprising:
(a) collecting a plurality of RF signals transmitted from a plurality of communication devices;
(b) generating a group of training samples by retrieving a plurality of measurement data from the RF signals;
(c) constructing a neural network according to the group of training samples;
(d) calibrating RF power of the communication devices using the neural network;
(e) generating calibration results of the RF power; and
(f) generating a frequency spectrum of the RF power according to the calibration results of the RF power, and displaying the frequency spectrum on a display device.

### Claim 14 (depends on 13)

The medium according to claim 13, wherein the frequency spectrum is a curve diagram that represents a relationship between digital-to-analog converter (DAC) power and the RF power.

### Claim 15 (depends on 13)

The medium according to claim 13, wherein block (b) comprises:
calculating a normalization value of each of the RF power by performing a normalization operation;
selecting part of the normalization values as input values of the neural network, and obtaining all the normalization values as output values of the neural network; and
generating the group of training samples according to the input values and the output values.

### Claim 16 (depends on 15)

The medium according to claim 15, wherein the normalization operation is a operation function that is denoted as Pi(Pi−Pmin)/(Pmax−Pmin) where “ Pi” is the normalization value of a the RF power, “Pi” is a RF power value, “Pmax” is a maximum value of the RF power, and “Pmin” is a minimum value of the RF power.

### Claim 17 (depends on 13)

The medium according to claim 13, wherein block (c) comprises:
generating a plurality of nerve cells of the neural network;
calculating a weight value of each of the nerve cells according to the group of training samples; and
constructing the neural network based upon the nerve cells with corresponding weight values.

### Claim 18 (depends on 13)

The medium according to claim 13, wherein block (d) comprises:
inputting part of the measurement data from an input layer of the neural network;
outputting target data from an output layer of the neural network after the part of measurement data are transmitted through a hidden layer of the neural network;
calculating calibrated data by performing a reversed normalization operation on the target data; and
generating the calibrated RF power according to the calibrated data.
