US2023231685A1PendingUtilityA1

AI-Assisted Selection of Demodulation Reference Type in 5G and 6G

Individually held — no corporate assignee on recordPriority: Jun 14, 2021Filed: Mar 15, 2023Published: Jul 20, 2023
Est. expiryJun 14, 2041(~14.9 yrs left)· nominal 20-yr term from priority
H04L 5/0051H04L 27/38H04L 27/0008H04L 27/2613H04L 27/22H04L 5/0048
73
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

Demodulation of 5G and 6G messages involves complex demodulation reference signals that occupy valuable resource grid area. Disclosed are numerous configurations of short-form demodulation reference types that provide sufficient modulation information to enable a receiver to determine all of the predetermined modulation levels of the modulation scheme, while consuming minimal resources. Selection of the appropriate modulation scheme and demodulation reference type generally depends on many competing factors. Therefore, an AI model may be required. The AI model may be trained on network data to recommend when a different modulation scheme would be beneficial, as the network default or to serve a particular user device. The AI model may be configured to select between the disclosed short-form demodulation references and prior-art demodulation reference signals, thereby optimizing the subsequent network performance as well as individual user satisfaction, while minimizing costs, bandwidth, power, and especially avoiding interference with neighboring cells.

Claims

exact text as granted — not AI-modified
1 . A method for a processor to select a particular demodulation reference scheme according to an artificial intelligence model, the method comprising:
 a) preparing, in a supercomputer, an artificial intelligence model comprising variable parameters, wherein the artificial intelligence model takes, as input, at least a message failure rate, and provides, as output, a suggested demodulation reference scheme selected from a plurality of predetermined demodulation reference types;   b) providing, to the artificial intelligence model, data related to performance of a wireless network, the data acquired while each of the demodulation reference types of the plurality is in use by the wireless network;   c) adjusting one or more of the variable parameters according to the data; and   d) producing an algorithm according to the artificial intelligence model, the algorithm configured for operation by a base station of a wireless network, wherein the algorithm takes as input at least the message failure rate of the base station, and provides as output the suggested demodulation reference type.   
     
     
         2 . The method of  claim 1 , wherein the base station is configured for communications according to 5G or 6G technologies. 
     
     
         3 . The method of  claim 1 , wherein the supercomputer is a quantum computer. 
     
     
         4 . The method of  claim 1 , further comprising storing the adjusted variable parameters in a cloud-based data storage spanning multiple physical storage media. 
     
     
         5 . The method of  claim 1 , further comprising storing the adjusted variable parameters in a blockchain. 
     
     
         6 . The method of  claim 1 , wherein the plurality of predetermined demodulation reference types comprises a particular demodulation reference type comprising exactly four resource elements and a particular modulation scheme, the particular modulation scheme comprising amplitude modulation and a plurality of predetermined amplitude levels, wherein each of the four resource elements is modulated according to one of the predetermined amplitude levels of the particular modulation scheme. 
     
     
         7 . The method of  claim 1 , wherein the plurality of predetermined demodulation reference types comprises a particular demodulation reference type comprising exactly two resource elements and a particular modulation scheme, the particular modulation scheme comprising amplitude modulation and a plurality of predetermined amplitude levels, wherein each of the two resource elements is modulated according to one of the predetermined amplitude levels of the particular modulation scheme. 
     
     
         8 . The method of  claim 1 , wherein the plurality of predetermined demodulation reference types comprises a particular demodulation reference type comprising exactly one resource element and a particular modulation scheme, the particular modulation scheme comprising amplitude modulation and a plurality of predetermined amplitude levels, wherein the one resource element is modulated according to one of the predetermined amplitude levels of the particular modulation scheme. 
     
     
         9 . The method of  claim 1 , wherein the plurality of predetermined demodulation reference types comprises a particular demodulation reference type comprising exactly one resource element and a particular modulation scheme, the particular modulation scheme comprising amplitude modulation and a plurality of predetermined amplitude levels, wherein the one resource element is modulated according to a first branch multiplexed with an orthogonal second branch, wherein the first branch is amplitude modulated according to a maximum amplitude level of the of the plurality of predetermined amplitude levels, and the second branch is amplitude modulated according to a minimum amplitude level of the plurality of predetermined amplitude levels. 
     
     
         10 . A wireless transmitter configured to:
 a) receive or determine at least two predetermined demodulation reference types, wherein:
 i) each predetermined demodulation reference type comprises at most four resource elements; 
 ii) each resource element comprises amplitude modulation according to a particular modulation scheme; 
 iii) the particular modulation scheme comprises a plurality of predetermined amplitude levels; and 
 iv) the predetermined amplitude levels comprise a maximum amplitude level and a minimum amplitude level; 
   b) select a particular demodulation reference type from the at least two predetermined demodulation reference types;   c) prepend, to a beginning of the data message, a leading demodulation reference according to the particular demodulation reference type;   d) append, to an ending of the data message, a trailing demodulation reference according to the particular demodulation reference type, wherein the leading demodulation reference is different from the trailing demodulation reference; and   e) transmit the data message including the leading and trailing demodulation references.   
     
     
         11 . The wireless transmitter of  claim 10 , wherein one demodulation reference, of the leading and trailing demodulation references, comprises the maximum amplitude level, and the other demodulation reference comprises the minimum amplitude level. 
     
     
         12 . The wireless transmitter of  claim 10 , wherein:
 a) a resource grid comprises multiple symbol-times in time, and multiple subcarriers in frequency;   b) each demodulation reference, of the leading and trailing demodulation references, comprises at least two resource elements;   c) when the data message is time-spanning by occupying multiple symbol-times at a single subcarrier, the leading and trailing demodulation references are time-spanning; and   d) when the data message is frequency-spanning by occupying multiple subcarriers at a single symbol-time, the leading and trailing demodulation references are frequency-spanning.   
     
     
         13 . The wireless transmitter of  claim 10 , further configured to:
 a) receive or determine a standard comprising instructions, wherein the instructions indicate the at least two predetermined demodulation reference types;   b) receive a message indicating a particular demodulation reference type from the at least two predetermined demodulation reference types;   c) determine, according to the instructions, a particular format of the leading demodulation reference and a different format of the trailing demodulation reference.   
     
     
         14 . The wireless transmitter of  claim 10 , wherein:
 a) the leading and trailing demodulation references are modulated according to a modulation scheme comprising an I-branch signal multiplexed with an orthogonal Q-branch signal;   b) the leading demodulation reference comprises a first resource element comprising an I-branch signal modulated according to the maximum amplitude level and a Q-branch signal modulated according to the minimum amplitude level; and   c) the trailing demodulation comprises a first resource element comprising an I-branch signal modulated according to the minimum amplitude level and a Q-branch signal modulated according to the maximum amplitude level;   
     
     
         15 . Non-transitory computer-readable media in a user device of a wireless network, the media containing instructions that when implemented in a computing environment cause a method to be performed, the method comprising:
 a) receiving a first demodulation reference comprising at most four resource elements, each resource element modulated according to a particular modulation scheme comprising amplitude modulation, the amplitude modulation comprising integer Namp predetermined amplitude levels, the predetermined amplitude levels comprising a maximum amplitude level and a minimum amplitude level;   b) determining, from one of the at most four resource elements, the maximum amplitude level, and determining, from one of the at most four resource elements, the minimum amplitude level; and   c) calculating, by interpolation, Namp-2 intermediate amplitude levels of the particular modulation scheme.   
     
     
         16 . The non-transitory computer-readable media of  claim 15 , the method further comprising:
 a) receiving a message concatenated with the first demodulation reference, the message comprising message elements, each message element modulated according to the particular modulation scheme;   b) for each message element, measuring one or more measured amplitude values;   c) for each message element, comparing each measured amplitude value to the Namp predetermined amplitude levels;   d) for each measured amplitude value of each message element, determining which of the predetermined amplitude levels is closest to the measured amplitude value; and   e) for each measured amplitude value of each message element, assigning, to the message element, a number or code indicating which of the predetermined amplitude levels is closest to the measured amplitude value.   
     
     
         17 . The non-transitory computer-readable media of  claim 15 , wherein:
 a) the first demodulation reference is configured according to a format selected, by an algorithm, from a plurality of predetermined demodulation reference formats;   b) wherein the algorithm takes, as input, at least a QoS (quality of service) required by the user device, and a measure of current interference at a location of the user device, and provides, as output, a selected format, selected from the predetermined demodulation reference formats.   
     
     
         18 . The non-transitory computer-readable media of  claim 15 , the method further comprising:
 a) receiving, after the first demodulation reference, a message comprising message elements; and   b) receiving, after the message, a second demodulation reference different from the first demodulation reference.   
     
     
         19 . The non-transitory computer-readable media of  claim 18 , the method further comprising:
 a) upon receiving the first demodulation reference, determining that a message follows; and   b) upon receiving the second demodulation reference, determining that the message has ended.   
     
     
         20 . The non-transitory computer-readable media of  claim 18 , the method further comprising:
 a) for each message element, calculating a first distance separating the message element from the first demodulation reference and a second distance separating the message element from the second demodulation reference;   b) for each message element, calculating a weighted average of an amplitude level of the first demodulation reference and an amplitude level of the second demodulation reference, wherein the weighted average is according to the first and second distances; and   c) demodulating the message element according to the weighted average.

Join the waitlist — get patent alerts

Track US2023231685A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.