• Developed Deep Learning algorithm to predict all six orbital parameters of LEO satellites for each specific epoch. • Innovated sequential prediction technique, using predicted parameters as test data for predicting other parameters.• Mitigated the error propagation problem of opportunistic positioning over time using RNN specifically LSTM.• Developing algorithm to perform pseudo range positioning with Iridium, Orbcomm and Starlink constellations.

• Reduced the Root Mean Squared Error (RMSE) in satellite position and velocity estimates by up to 54% and 73% respectively. • Explored Satellite Laser Ranging (SLR) data for LEO satellites to directly validate prediction results from the LSTM model.• Developing a novel approach for Orbital Parameter prediction with LSTM focusing on predicting error propagation.

o Developing a single GPS/Beidou receiver based on Vector Tracking Loop.o Analyzed research papers and materials to obtain core level understanding of the topic.o Analyzed results from Conventional Tracking and Vector Tracking to compare performance.