Signal transmission apparatus, parameter determination apparatus, signal transmission method, parameter determination method and recording medium
A signal transmission apparatus ( 1 ) includes: a distortion compensation unit ( 11 ) for performing a distortion compensation processing on an input signal (x) by using a Neural Network ( 112 ) including L+1 arithmetic layers that include L (L is a variable number representing an integer equal to or larger than 1) hidden layer ( 112 M) and an output layer ( 112 O); a storage unit ( 13 ) for storing parameter sets ( 131 ) each of which includes a parameter for Q (Q is a variable number representing an integer equal to or smaller than L) arithmetic layer of the L+1 arithmetic layers; and an application unit ( 142 ) for selecting one parameter set from the parameter sets based on a signal pattern of the input signal and applying the parameter included in the selected one parameter set to the M number of arithmetic layer, a parameter of another arithmetic layer of the L+1 arithmetic layers, which is other than the Q arithmetic layer, is fixed.
1. A signal transmission apparatus comprising:
at least one memory configured to store instructions; and
at least one processor configured to execute the instructions to:
perform a distortion compensation processing on an input signal by using a Neural Network including L+1 number of arithmetic layers, wherein the arithmetic layers include L hidden layers and an output layer, and wherein L is a variable number representing an integer that is equal to or larger than 1;
select, from a storage that is configured to store a plurality of parameter sets that each include a parameter for Q number of arithmetic layers of the L+1 number of arithmetic layers, one parameter set based on a signal pattern of the input signal, wherein Q is a variable number representing an integer that is equal to or smaller than L; and
apply the parameter included in the selected one parameter set to the Q number of arithmetic layers, wherein
a parameter of an arithmetic layer of the L+1 number of arithmetic layers, other than the Q number of arithmetic layers, is fixed.
2. The signal transmission apparatus according to claim 1 , wherein
the Q number of arithmetic layers includes the output layer.
3. The signal transmission apparatus according to claim 2 , wherein
the variable number Q is 1,
the parameter of the L number of hidden layers is fixed.
4. The signal transmission apparatus according to claim 1 , wherein
each of the plurality of parameter sets includes the parameter that is to be applied to the Q number of arithmetic layers when the distortion compensation processing is performed on the input signal having one signal pattern corresponding to each parameter set.
5. The signal transmission apparatus according to claim 1 , wherein
the at least one processor is configured to execute the instructions to:
determine the signal pattern of the input signal; and
select the one parameter set based on a determination result.
6. The signal transmission apparatus according to claim 5 , wherein
the at least one processor is configured to execute the instructions to determine the signal pattern of the input signal based on an electric power characteristic of the input signal.
7. The signal transmission apparatus according to claim 6 , wherein
the electric power characteristic includes at least one of an average electric power and an electric power distribution of the input signal.
8. A signal transmission method performed by a computer and comprising:
performing a distortion compensation processing on an input signal by using a Neural Network including L+1 number of arithmetic layers, wherein the arithmetic layers include L hidden layers and an output layer, and wherein L is a variable number representing an integer that is equal to or larger than 1;
selecting, from a storage that is configured to store a plurality of parameter sets that each include a parameter for Q number of arithmetic layers of the L+1 number of arithmetic layers, one parameter set based on a signal pattern of the input signal, wherein Q is a variable number representing an integer that is equal to or smaller than L; and
applying the parameter included in the selected one parameter set to the Q number of arithmetic layers, wherein
a parameter of an arithmetic layer of the L+1 number of arithmetic layers, other than the Q number of arithmetic layers, is fixed.
9. A non-transitory recording medium storing a computer program executable by a computer to execute perform a signal transmission method comprising:
performing a distortion compensation processing on an input signal by using a Neural Network including L+1 number of arithmetic layers, wherein the arithmetic layers include L hidden layers and an output layer, and wherein L is a variable number representing an integer that is equal to or larger than 1;
selecting, from a storage that is configured to store a plurality of parameter sets that each include a parameter for Q number of arithmetic layers of the L+1 number of arithmetic layers, one parameter set based on a signal pattern of the input signal, wherein Q is a variable number representing an integer that is equal to or smaller than L; and
applying the parameter included in the selected one parameter set to the Q number of arithmetic layers, wherein
a parameter of an arithmetic layer of the L+1 number of arithmetic layers, other than the Q number of arithmetic layers, is fixed.