Device and method for training a model

A method of computer-implemented training a model, the method comprising:

• first classifying a plurality of data packets using the model, wherein a first class is assigned to each data packet of a plurality of data packets, wherein the first class is associated with a receiver of a plurality of receivers, wherein each receiver of the plurality of receivers comprises a complexity level of a plurality of complexity levels, and wherein each data packet of the plurality of data packets requires a receiver of the plurality of receivers with a minimum complexity level to be processed without an error;
• second classifying the plurality of data packets, wherein a second class is assigned to each data packet of the plurality of data packets, wherein the second class is associated with a receiver of the plurality of receivers, wherein the second classifying comprises:
• processing each data packet of the plurality of data packets to determine the respective receiver with the minimum complexity level required to process the respective data packet without an error, wherein the processing comprises processing the respective data packet by a first receiver comprising a first complexity level and, in the case that the first receiver generates an error, processing the data packet by a second receiver comprising a second complexity level higher than the first complexity level;
• and
• assigning to each data packet of the plurality of data packets the second class associated to the respectively determined receiver comprising the minimum complexity level;
• and
• training the model using the plurality of first classes and the plurality of second classes assigned to the plurality of data packets.