System and method to extract satellite doppler curves from waterfall spectrograms data

A computer-implemented method for automated satellite Doppler curve orbit fitting, the computer-implemented method comprising:

• receiving, by one or more computer processors, waterfall spectrograms;
• extracting, by one or more computer processors, satellite data from the waterfall spectrograms;
• removing, by the one or more computer processors, a first noise from the satellite data, wherein the first noise includes horizontal lines and small clusters;
• agglomerating, by the one or more computer processors, the satellite data into one or more clusters using adaptive clustering, wherein the adaptive clustering creates a cluster list wherein each element of the cluster list is a list of integer indices;
• fitting, by the one or more computer processors, a Doppler curve model to each cluster of the one or more clusters;
• removing, by the one or more computer processors, noise clusters from the one or more clusters based on a second noise, wherein the second noise has a Root Mean Square Error (RMSE) below a predetermined threshold;
• and
• determining, by the one or more computer processors, one or more orbital elements of a satellite for each remaining cluster of the one or more clusters.

A system for automated satellite Doppler curve orbit fitting, the system comprising:

• one or more non-transitory computer readable storage media;
• and
• one or more computer processors, the one or more computer processors are further configured to:
• receive waterfall spectrograms;
• extract satellite data from the waterfall spectrograms;
• remove a first noise from the satellite data, wherein the first noise includes horizontal lines and small clusters;
• agglomerate the satellite data into one or more clusters using adaptive clustering, wherein the adaptive clustering creates a cluster list wherein each element of the cluster list is a list of integer indices;
• fit a Doppler curve model to each cluster of the one or more clusters;
• remove noise clusters from the one or more clusters based on a second noise wherein the second noise has a Root Mean Square Error (RMSE) below a predetermined threshold;
• and
• determine one or more orbital elements of a satellite for each remaining cluster of the one or more clusters.