Message Fault Recovery Without Retransmission in 5G and 6G
A receiver can determine that a received message is corrupt according to an associated error-detection code in 5G/6G. The receiver can then calculate the modulation quality of each message element of the message by comparing a modulation value of the message element with a set of predetermined modulation levels, wherein a larger deviation from the closest predetermined modulation level corresponds to a lower modulation quality. Noise and interference usually generate random changes in the modulation as-received, and this generally results in a larger deviation relative to the predetermined modulation levels and hence a lower modulation quality. The receiver can count the number of faulted message elements with modulation quality below a threshold, and if the number of faults is less than some limit, the receiver can attempt to recover the message by altering the worst-quality message elements, testing each version for consistency.
1 . A method for a wireless receiver to recover a faulted message, the method comprising:
a) receiving a message comprising message elements, each message element modulated according to a modulation scheme, the modulation scheme comprising amplitude modulation multiplexed with phase modulation, wherein the amplitude modulation comprises integer Namp predetermined amplitude levels spaced apart by an amplitude step, and the phase modulation comprises integer Nphase predetermined phase levels spaced apart by a phase step;
b) determining, according to an error-detection code associated with the message, that the message is corrupted;
c) for each message element, determining a message amplitude and a message phase;
d) for each message element, determining an amplitude deviation comprising an amplitude difference divided by the amplitude step, wherein the amplitude difference comprises a difference between the message amplitude and a closest amplitude level of the predetermined amplitude levels;
e) for each message element, determining a phase deviation comprising a phase difference divided by the phase step, wherein the phase difference comprises a difference between the message phase and a closest phase level of the predetermined phase levels; and
f) for each message element, determining a modulation quality according to the amplitude deviation and the phase deviation.
2 . The method of claim 1 , wherein the message is configured according to 5G or 6G technology.
3 . The method of claim 1 , wherein:
a) for each message element, the modulation quality is an inverse function of the amplitude deviation and the phase deviation, such that a larger amplitude deviation or phase deviation corresponds to a lower modulation quality, and a smaller amplitude deviation or phase deviation corresponds to a higher modulation quality.
4 . The method of claim 1 , wherein:
a) for each message element, the determining the modulation quality comprises calculating a square root of a sum of the amplitude deviation squared plus the phase deviation squared.
5 . The method of claim 1 , wherein:
a) for each message element, the determining the modulation quality comprises calculating a sum of a magnitude of the amplitude deviation plus a magnitude of the phase deviation.
6 . The method of claim 1 , wherein:
a) for each message element, the determining the modulation quality further comprises determining a SNR (signal-to-noise ratio or signal-to-interference-and-noise ratio) of the message amplitude or the message phase or both.
7 . The method of claim 6 , wherein:
a) for each message element, the determining the modulation quality further comprises using an algorithm that takes as input the amplitude deviation, the phase deviation, and the SNR of the message element, and provides, as output, an estimate of the modulation quality.
8 . The method of claim 7 , wherein the algorithm is derived according to an AI (artificial intelligence) model.
9 . The method of claim 8 , further comprising:
a) receiving a retransmitted version of the message;
b) determining that the retransmitted version is not corrupted; and
c) determining, by comparing corresponding message elements of the message and the retransmitted version, which message elements of the message are faulted.
10 . The method of claim 9 , further comprising:
a) providing, as input to the AI model, the message, the retransmitted version, and an indication of which message elements of the message are faulted.
11 . Non-transitory computer-readable media in a wireless receiver, the media containing instructions that when implemented in a computing environment cause a method to be performed, the method comprising:
a) receiving a message and determining that the message is corrupted;
b) calculating, for each message element of the message, a modulation quality according to a modulation of the message element;
c) determining a number Nbad of bad-modulation elements that have modulation quality below a predetermined threshold; and
d) upon determining that Nbad is zero, requesting a retransmission of the message.
12 . The non-transitory computer-readable media of claim 11 , further comprising:
a) upon determining that Nbad is greater than a predetermined limit Nlimit, requesting a retransmission of the message.
13 . The non-transitory computer-readable media of claim 12 , further comprising:
a) upon determining that Nbad is greater than zero and at most equal to Nlimit, sequentially producing a plurality of message versions, wherein each message version includes one of the bad-modulation message elements altered in modulation; and
b) determining whether all of the message versions are corrupted.
14 . The non-transitory computer-readable media of claim 13 , further comprising:
a) upon determining that all of the message versions are corrupted, sequentially producing a plurality of further message versions, wherein each of the further message versions includes two or more of the bad-modulation elements altered in modulation; and
b) determining whether all of the further message versions are corrupted.
15 . The non-transitory computer-readable media of claim 14 , further comprising:
a) upon determining that all of the further message versions are corrupted, requesting a retransmission of the message.
16 . A processor in a wireless receiver, the processor comprising an AI (artificial intelligence) model, the processor configured to:
a) receive a message comprising message elements modulated according to QAM (quadrature amplitude modulation) comprising a first branch signal and am orthogonal second branch signal, each branch signal amplitude-modulated according to a set of predetermined amplitude levels;
b) determine that the message is corrupted by comparing the message, or a hash or digest thereof, to an error-detection code associated with the message; and
c) determine, for each message element, using the AI model, a modulation quality according to a first deviation and a second deviation, the first deviation comprising a difference between the first branch signal and a closest one of the predetermined amplitude levels, and the second deviation comprising a difference between the second branch signal and a closest one of the predetermined amplitude levels.
17 . The processor of claim 16 , further configured to:
a) select a selected message element having lowest modulation quality;
b) using the AI model, produce a plurality of versions by successively altering the first and second branch signals of the selected message element to each of the predetermined amplitude levels in a nested grid search that includes all combinations of the two branch signals and all of the predetermined amplitude levels; and
c) determine whether each of the versions is corrupted.
18 . The processor of claim 17 , further configured to:
a) upon determining that one of the versions, or a hash or digest thereof, agrees with the error-detection code, transmit an acknowledgement.
19 . The processor of claim 17 , further configured to:
a) upon determining that all of the versions, or a hash or digest thereof, disagrees with the error-detection code, select two selected message elements having lowest modulation quality; and
b) using the AI model, produce additional versions by successively altering the first and second branch signals of the two selected message elements to each of the predetermined levels in a nested grid search that includes all combinations of the two branch signals of the two selected message elements, and all of the predetermined amplitude levels.
20 . The processor of claim 19 , further configured to:
a) upon determining that one of the additional versions, or a hash or digest thereof, agrees with the error-detection code, transmit an acknowledgement.