As space domain together with its increasing mission complexity evolves, it is now more important than ever for autonomous space assets be equipped with onboard navigation and timing that are resilient in operationally challenged and/or degraded GPS environments. In this topic, both theoretical and practical fronts of nonlinear, nonstationary-based signal decomposition approaches are cornerstones for the realization of environmental sensing, where feature extractions are adapted in responding to ubiquitous kinds of signals, including radio frequency (RF) and non-RF signals of opportunity. Also relevant are feature construction frameworks with appropriate support of data fusion operations that are beneficial to representing natural characteristics of the signals of interest; e.g., frequency band variations, statistic time-frequency variations, and correlations in sparse feature domains, etc. despite of the inherent challenges in different sizes and formats of signal features. Innovations and advances in planning and executing resilient satellite navigation and timing operations are also being sought for operationally transient situations from which deep learning based signal classification breakthroughs are foundational to reconfiguring positioning, navigation and timing (PNT) payloads and software-defined radios in accordance of the constructed features from friendly and adversarial impacts to the overall PNT performance and capabilities. Testing and evaluation of feature extraction and classification performance are essential when facing the challenge of dynamic ranges of signal-to-noise ratios.

References

1. G. Chen, K. D. Pham, and E. Blasch, "An Integrated Classification and Clustering System for RF Signal Detection and Classification," IEEE Aerospace Conference, 2019

2. D. Shen, et.al., "Adaptive Markov Inference Game Optimization for Rapid Discovery of Evasive Satellite Behaviors," Proceedings of SPIE, Sensors and Systems for Space Applications XII, 2019

key words

Deep Learning; Feature Extraction; Feature Construction; Data Fusion; Classification; Performance Evaluation; satellite navigation and timing

Citizenship:  Open to U.S. citizens

Level:  Open to Postdoctoral and Senior applicants