Carrier acquisition in satellite communications
A computer comprises a processor and a memory. The memory stores instructions executable by the processor to input a frequency spectrum distribution to a machine learning program to obtain carrier data as output from the machine learning program satellite communication. The machine learning program is trained with a plurality of frequency spectrum distributions of a wireless satellite communication signal and metadata specifying one or more satellite communication carriers for respective ones of the frequency spectrum distributions, wherein the metadata for each satellite communication carrier includes a respective center frequency and a respective symbol rate.
1. A computer, comprising a processor and a memory, the memory storing instructions executable by the processor to:
input a frequency spectrum distribution to a machine learning program to obtain carrier data, including a symbol rate and a center frequency, and output the obtained carrier data from the machine learning program,
wherein the machine learning program is trained with a plurality of frequency spectrum distributions of a wireless satellite communication signal and metadata specifying one or more satellite communication carriers for respective ones of the frequency spectrum distributions, wherein the metadata for each satellite communication carrier includes a respective center frequency and a respective symbol rate.
2. The computer of claim 1 , wherein respective frequency spectrum distributions are a result of a frequency spectrum analysis of the received wireless satellite communication.
3. The computer of claim 2 , wherein the frequency spectrum analysis includes performing a Fourier Transform.
4. The computer of claim 1 , wherein the instructions further include instructions to determine training data for the machine learning program based on user input specifying one or more satellite carriers for each of the one or more frequency spectrum distributions.
5. The computer of claim 4 , wherein the one or more frequency spectrum distributions are results of simulation or measurement of satellite signal.
6. The computer of claim 1 , wherein the carrier data further includes a confidence measure for each identified carrier.
7. The computer of claim 1 , wherein the instructions further include instructions to determine that the machine learning program is trained upon determining that a detection accuracy of the machine learning program exceeds a threshold.
8. The computer of claim 1 , wherein the machine learning program is a convolutional neural network including a dense layer and a flatten layer.
9. The computer of claim 1 , wherein the instructions include further instructions to update a stored carrier list with carrier data output from the machine learning system.
10. The computer of claim 9 , wherein the instructions include further instructions to lock to a carrier from the updated carrier list.
11. The method of claim 1 , further comprising updating a stored carrier list with carrier data output from the machine learning system.
12. The method of claim 11 , further comprising locking to a carrier from the updated carrier list.
13. A method, comprising:
input a frequency spectrum distribution to a machine learning program to obtain carrier data, including a symbol rate and a center frequency, and output the obtained carrier data from the machine learning program,
wherein the machine learning program is trained with a plurality of frequency spectrum distributions of a wireless satellite communication signal and metadata specifying one or more satellite communication carriers for respective ones of the frequency spectrum distributions, wherein the metadata for each satellite communication carrier includes a respective center frequency and a respective symbol rate.
14. The method of claim 13 , wherein respective frequency spectrum distributions are a result of a frequency spectrum analysis of the received wireless satellite communication.
15. The method of claim 14 , wherein the frequency spectrum analysis includes performing a Fourier Transform.
16. The method of claim 13 , further comprising determining training data for the machine learning program based on user input specifying one or more satellite carriers for each of the one or more frequency spectrum distributions.
17. The method of claim 16 , wherein the one or more frequency spectrum distributions are results of simulation or measurement of satellite signal.
18. The method of claim 13 , wherein the carrier data further includes a confidence measure for each identified carrier.
19. The method of claim 13 , further comprising determining that the machine learning program is trained upon determining that a detection accuracy of the machine learning program exceeds a threshold.
20. The method of claim 13 , wherein the machine learning program is a convolutional neural network including a dense layer and a flatten layer.