Pseudoranger_trials

This code was made to test various noise cancellation filters for smoothing orbital data during application process of Google Summer of Code 2017.

Linear Regression

The X's and Y's are estimated separately. And then plotted against the jittery orbit. The results however uptill now are not satisfactory.

Convolutional Filter

The Convolutional averaging filter performs much better as compared to the previous regression based model.

Exponential Smoothing

The model performs exceptionally for some values of smoothing constant.

Credits

I would like to thank @hornig for orbit generation code which can be found in this repository -> a03-LonePseudorangerOrbitPosition