# US 12,003,323 B2 · AI model with error-detection code for fault correction in 5G/6G

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

- Number: US 12,003,323 B2
- Document ID: US12003323B2
- Status: Granted
- Kind: B2
- Application number: 18/502,054
- Filed: 2023-11-05
- Published: 2024-02-22
- Granted: 2024-06-04
- Earliest claimed: 2022-09-06
- Applicants: David E. Newman; R. Kemp Massengill
- Inventors: David E. Newman; R. Kemp Massengill
- Web page: https://patentgrep.com/patent/US12003323B2

## CPC

- H04L 1/0046: {Code rate detection or code type detection (H04L1/0038 takes precedence; detection of the data rate H04L25/0262; for packet format H04L1/0091)}
- H03M 13/09: Error detection only, e.g. using cyclic redundancy check [CRC] codes or single parity bit
- H03M 13/6597: {Implementations using analogue techniques for coding or decoding, e.g. analogue Viterbi decoder}
- H04L 1/0061: {Error detection codes}
- H04L 1/0064: {Concatenated codes}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- H03M: CODING; DECODING; CODE CONVERSION IN GENERAL (using fluidic means F15C4/00; optical analogue/digital converters G02F7/00; coding, decoding or code conversion, specially adapted for particular applications, see the relevant subclasses, e.g. G01D, G01R, G06F, G06T, G09G, G10L, G11B, G11C, H04B, H04L, H04M, H04N; ciphering or deciphering for cryptography or other purposes involving the need for secrecy G09C)

## Abstract

Message faulting is a critical unsolved problem for 5G and 6G. Disclosed herein is a method for combining an AI-based analysis of the waveform data of each message element, plus the constraint of an associated error-detection code (such as a CRC or parity construct of the correct message) to localize and, in many cases, correct a limited number of faults per message, without a retransmission. For example, the waveform data may include a deviation of the amplitude or phase of a particular message element, relative to an average of the amplitudes or phases of the other message elements that have the same demodulation value. The outliers are thereby exposed as the most likely faulted message elements. In addition, using the error-detection code, the AI model can determine the most likely corrected message, thereby avoiding retransmission delays and power usage and other costs.

## Claims

### Claim 1

A method for a wireless receiver to demodulate a message, the method comprising:
a) receiving a message and an error-detection code associated with the message;
b) determining, according to the error-detection code, that the message is corrupted;
c) providing the message and the error-detection code as inputs to an artificial intelligence (AI) model trained to mitigate corrupted messages by correlating message data with the error-detection code; and
d) determining, as output from the AI model, a corrected version of the message.

### Claim 2 (depends on 1)

The method of claim 1, wherein the message is received according to 5G or 6G technology.

### Claim 3 (depends on 1)

The method of claim 1, wherein:
a) the message comprises message elements, each message element comprising a single resource element of a resource grid according to symbol-times in time and subcarriers in frequency;
b) wherein each message element contains a waveform signal, the waveform signal comprising a waveform amplitude and a waveform phase; and
c) wherein the waveform amplitude and the waveform phase are modulated, by a transmitter, according to a modulation scheme comprising integer Nstate modulation states, each modulation state being amplitude modulated according to integer Namp amplitude modulation levels or phase modulated according to integer Nphase phase modulation levels.

### Claim 4 (depends on 3)

The method of claim 3, wherein, for each message element, the message data comprises an amplitude deviation, the amplitude deviation comprising a difference between the waveform amplitude as-received and a closest amplitude modulation level of the Namp amplitude modulation levels.

### Claim 5 (depends on 4)

The method of claim 4, further comprising:
a) for each particular amplitude modulation level, of the Namp modulation levels, determining a subset of the message elements wherein the particular amplitude modulation level is the closest amplitude modulation level to each message element in the subset;
b) for each particular amplitude modulation level, of the Namp modulation levels, determining an average amplitude of the waveform amplitudes of the message elements in the subset;
c) for each message element, determining an amplitude displacement comprising a difference between the waveform amplitude of the message element and the average amplitude of the subset corresponding to the message element; and
d) providing, as further input to the AI model, the amplitude displacement of each message element.

### Claim 6 (depends on 3)

The method of claim 3, wherein, for each message element, the message data comprises a phase deviation, the phase deviation comprising a difference between the waveform phase as-received and the closest phase modulation level of the Nphase phase modulation levels.

### Claim 7 (depends on 6)

The method of claim 6, further comprising:
a) for each particular phase modulation level, of the Nphase modulation levels, determining a subset of the message elements wherein the particular phase modulation level is the closest phase modulation level to each message element in the subset;
b) for each particular phase modulation level, of the Nphase modulation levels, determining an average phase of the waveform phase of the message elements in the subset;
c) for each message element, determining a phase displacement comprising a difference between the waveform phase of the message element and the average phase of the subset corresponding to the message element; and
d) providing, as further input to the AI model, the phase displacement of each message element.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
a) determining a phase rotation angle by combining the average phases of the Nphase subsets; and
b) subtracting the phase rotation angle from the waveform phase of each message element.

### Claim 9

Non-transitory computer-readable media in a wireless receiver, the non-transitory computer-readable media containing instructions that, when executed by a computing environment, cause a method to be performed, the method comprising:
a) receiving or developing or arranging to use an artificial intelligence (AI) model trained to correct corrupted messages;
b) receiving a corrupted message and an error-detection code associated with the corrupted message;
c) providing, as input to the AI model, the corrupted message or data about the corrupted message, as-received;
d) providing, as further input to the AI model, the error-detection code or data about the error-detection code, as received; and
e) determining, as output from the AI model, a most likely corrected version of the corrupted message.

### Claim 10 (depends on 9)

The non-transitory computer-readable media of claim 9, wherein:
a) the corrupted message comprises message elements, each message element comprising a single resource element of a resource grid, each message element comprising a waveform signal, the waveform signal comprising a waveform amplitude and a waveform phase; and
b) wherein each waveform signal is modulated, by a transmitter, according to a modulation scheme, the modulation scheme comprising amplitude modulation or phase modulation or both, wherein the amplitude modulation comprises integer Namp amplitude modulation levels and the phase modulation comprises integer Nphase equally spaced-apart phase modulation levels.

### Claim 11 (depends on 10)

The non-transitory computer-readable media of claim 10, the method further comprising:
a) for each message element, determining an amplitude deviation comprising a difference between the waveform amplitude of the message element and a closest amplitude modulation level of the Namp amplitude modulation levels; and
b) providing, as further input to the AI model, the amplitude deviations of the message elements.

### Claim 12 (depends on 11)

The non-transitory computer-readable media of claim 11, the method further comprising:
a) for each message element, determining a phase deviation comprising a difference between the waveform phase of the message element and a closest phase modulation level of the Nphase phase modulation levels; and
b) providing, as further input to the AI model, the phase deviations of the message elements.

### Claim 13 (depends on 12)

The non-transitory computer-readable media of claim 12, the method further comprising:
a) determining, as further output from the AI model, according to the amplitude deviations or the phase deviations or both, a number Nfault of likely faulted message elements; and
b) when Nfault exceeds a predetermined threshold, requesting a retransmission of the corrupted message.

### Claim 14 (depends on 12)

The non-transitory computer-readable media of claim 12, the method further comprising;
a) determining, as further output from the AI model, according to the amplitude deviations or the phase deviations or both, an estimated probability that the most likely corrected version of the message is correct; and
b) when the estimated probability is below a predetermined threshold, requesting a retransmission of the corrupted message.

### Claim 15

A method for a wireless receiver to correct a corrupted message comprising an error-detection code, the method comprising:
a) determining a series of message elements comprising the message, each message element comprising a waveform signal;
b) providing, as input to an artificial intelligence (AI) model, the waveform signals or data derived from the waveform signals;
c) providing, as further input to the AI model, the error-detection code or data derived from the error-correction code; and
d) determining, as output from the AI model, a corrected version of the message.

### Claim 16 (depends on 15)

The method of claim 15, the method further comprising:
a) determining, as further output from the AI model, which message elements are likely faulted; and
b) determining, as further output from the AI model, a most probable corrected value of each of the likely faulted message elements.

### Claim 17 (depends on 15)

The method of claim 15, the method further comprising:
a) determining, as further output from the AI model, a number of likely faulted message elements; and
b) when the number of likely faulted message elements exceeds a predetermined threshold, requesting a retransmission of the message.

### Claim 18 (depends on 15)

The method of claim 15, the method further comprising:
a) receiving, proximate to the message, a demodulation reference comprising reference elements, each reference element comprising a single resource element of the resource grid, each reference element comprising a reference waveform; and
b) providing, as further input to the AI model, data about each reference element and each reference waveform of the demodulation reference.

### Claim 19 (depends on 15)

The method of claim 15, the method further comprising:
a) determining, for each message element, an amplitude offset comprising a difference between an amplitude of the waveform of the message element and an average waveform amplitude determined from a plurality of other message elements in the message; and
b) providing, as further input to the AI model, the amplitude offset of each message element.

### Claim 20 (depends on 15)

The method of claim 15, the method further comprising:
a) determining, for each message element, a phase offset comprising a difference between a phase of the waveform of the message element and an average waveform phase derived from a plurality of other message elements of the message; and
b) providing, as further input to the AI model, the phase offset of each message element.