An example of neural network usage for assigning a modulation-code scheme to a 5G base station scheduler

The article proposes a method for assigning a modulation-code scheme by a base station scheduler based on predicting the value of the signal-to-interference ratio on the mobile user's equipment at the next time slot from a sequence of known values of this ratio in the past. For prediction, a model of a single-layer neural network is built in the work, by the example of which a machine learning process is shown for solving a multiparametric optimization problem using the stochastic gradient method. The trained neural network for the predicted value of the signal/interference ratio allows the scheduler to correctly select the modulation-code scheme for the user, thereby ensuring the level of quality of data transmission in the radio channel required for the provision of the service.