Modulation format estimation device, and modulation format estimation method

A modulation format estimation method comprising:

• estimating an amount of a frequency shift of a baseband signal acquired from a received signal of which a modulation format is unknown and correcting the baseband signal based on the estimated amount of the frequency shift;
• generating a plurality of frequency errors from a range set based on an error in the estimated amount of the frequency shift;
• acquiring a plurality of learning baseband signals in which each of a plurality of source signals modulated by different modulation formats is frequency-shifted by each frequency error;
• inputting the baseband signal that has been corrected based on the estimated amount of the frequency shift to a first machine learning model created by machine learning using learning data including the plurality of learning baseband signals and labels indicating the different modulation formats of the learning baseband signals;
• and
• estimating a modulation format of the received signal based on an output of the first machine learning model.

A modulation format estimation device comprising:

• at least one memory storing instructions;
• and
• at least one processor configured to execute the instructions to:
• generate a plurality of frequency errors from a range set based on an error in an estimated amount of frequency shift of a broadband signal acquired from a received signal of which a modulation formation is unknown;
• acquire a plurality of learning baseband signals in which each of a plurality of source signals modulated by different modulation formats is frequency-shifted by each frequency error;
• and
• create a first machine learning model by machine using learning data including the plurality of learning baseband signals and labels indicating the different modulation formats of the learning baseband signals.