Model training method, channel adjustment method, electronic device, and computer readable storage medium

A channel adjustment method, comprising:

• collecting historical samples, with the historical samples comprising first scheduling information and first information corresponding to a historical data transmission, the first information representing the result of the cyclic redundancy check, and the first scheduling information comprising first intermediate variable information in an adaptive modulation and coding process;
• performing model training on a first prediction model according to the historical samples to obtain a trained first prediction model, wherein during the model training, the first scheduling information is used as an input of the first prediction model, the first information is converted into second information corresponding to the historical data transmission to be used as an output of the first prediction model;
• collecting second scheduling information corresponding to a current data transmission, with the second scheduling information comprising second intermediate variable information in an adaptive modulation and coding process;
• inputting the second scheduling information into the trained first prediction model to obtain second information corresponding to the current data transmission, with the second information representing a probability value of a result of cyclic redundancy check;
• in response to that the second information corresponding to the current data transmission is less than a first threshold, or greater than a second threshold, adjusting the second intermediate variable information according to the second information corresponding to the current data transmission to obtain adjusted second intermediate variable information;
• and
• replacing the second intermediate variable information with the adjusted second intermediate variable information, and continuing to execute the inputting the second scheduling information into a trained first prediction model to obtain second information corresponding to the current data transmission until a loop exit condition is met,
• wherein the loop exit condition being met comprises any one of:
• a total number of adjustment times is greater than or equal to a preset maximum number of the adjustment times;
• the second information corresponding to the current data transmission is greater than or equal to the first threshold and less than or equal to the second threshold;
• or
• an accumulated adjustment amount is greater than or equal to a preset maximum accumulated adjustment amount, with the accumulated adjustment amount being a sum of adjustment amounts corresponding to previous N adjustments, N being a current number of the adjustment times, and an adjustment amount corresponding to each adjustment being a difference between the adjusted second intermediate variable information corresponding to the adjustment and the second intermediate variable information before the adjustment,
• wherein the first threshold is less than the second threshold.

A channel adjustment method, comprising:

• collecting historical samples, with the historical samples comprising first scheduling information and first information corresponding to a historical data transmission, the first information representing the result of the cyclic redundancy check, and the first scheduling information comprising first intermediate variable information in an adaptive modulation and coding process;
• performing model training on a first prediction model according to the historical samples to obtain a trained first prediction model, wherein during the model training, the first scheduling information is used as an input of the first prediction model, the first information is converted into second information corresponding to the historical data transmission to be used as an output of the first prediction model;
• collecting second scheduling information corresponding to a current data transmission, with the second scheduling information comprising second intermediate variable information in an adaptive modulation and coding process;
• traversing all values of the second intermediate variable information, and inputting the second scheduling information into the trained first prediction model to obtain the second information corresponding to the current data transmission which corresponds to each value of the second intermediate variable information, with the second information representing a probability value of a result of cyclic redundancy check;
• and
• selecting the value, corresponding to a difference with a minimum absolute value in differences between the second information corresponding to the current data transmission and a target value of a block error rate, from all the values of the second intermediate variable information.