https://www.engr.ncsu.edu/news/2021/11/19/nasa-intern-works-on-ai...

https://www.engr.ncsu.edu/news/2021/11/19/nasa-intern-works-on-ais-next-generation/

As part of the Space Communications and Navigation (SCaN) Internship Project (SIP) at the NASA Glenn Research Center, I am currently working on the following topic: Analysis of Artificial Intelligence Applications for Cognitive Communications Systems. My project involves implementing algorithms for CubeSat scheduling and understanding the machine learning environment for ground station development to support the cognitive links team. In the Fall 2021 term, I am working on the following tasks:° Developing a spiking neural network for network topology problems to estimate various port metrics using snnTorch° Understanding how snnTorch interfaces with neuromorphic hardware and assess how SNNs in this framework can perform on various cognitive communications tasks° Developing an anomaly detection algorithm for radio frequency propagation data using AWS SageMaker & Data Wrangler° Wrote user guide for snnTorch to establish user practices at NASAMy contributions during the Summer 2021 term include: ° Developed spiking neural network (SNN) models for communication data on Nengo for cognitive learning/communication channel tasks ° Developed a convolutional neural network (CNN) to perform multiclass classification of Amazon forest image data (achieving 85% fbeta score) to compare against SNN performance° Developed regression (such as XGBoost regression, random forest regression, and Bayesian ridge regression) models to estimate wireless link quality for communication systems using sklearn and AWS SageMaker° Wrote a user guide for Nengo program execution and developing applications in neuromorphic computing to establish user practices at NASA° Learned about the FPGA design environment and development tools (such as Xilinx) ° Learned VHDL through introductory, intermediate, and advanced courses offered by GRC