Overview
Next generation wireless systems need to operate in harsh environments, where various types of interference increase the system level challenges. Wireless receivers have numerous applications in systems that require efficient spectrum management. In this session, we will demonstrate how to apply techniques Deep Learning and Machine Learning networks for a range of wireless communications systems.
We will look at the trade-offs between machine learning and deep learning workflows. We will also demonstrate ways to perform data collection and labeling from off-the-shelf software-defined radios and radars to train and test classifiers. Our focus will be on data synthesis to train networks including efficient ways to work with communications baseband I/Q signals to improve classification results.
Highlights
Demonstrate the concepts and workflows using several application examples including waveform modulation ID, RF Fingerprinting, and 5G channel estimation
Understand trade-offs between machine learning and deep learning techniques for baseband signals
Pre-process and label baseband data
Synthesize data to train networks
About the Presenters
Dr. Houman Zarrinkoub is a senior product manager at bat365 responsible for wireless communications products. During his 20-year tenure at bat365, he has also served as a development manager and has been responsible for multiple signal processing and communications software tools. Prior to bat365, he was a research scientist working on mobile and voice coding technologies in the Wireless Group at Nortel Networks. He has been awarded multiple patents on topics related to computer simulations of signal processing applications. Houman is the author of the book Understanding LTE with MATLAB: From Mathematical Modeling to Simulation and Prototyping. He holds a B.Sc. degree in electrical engineering from McGill University and M.Sc. and Ph.D. degrees in telecommunications from the University of Quebec, in Canada.
Florent Busnoult is a Senior Application Engineer at bat365 focusing on signal processing and wireless communications products. He holds a Master’s degree in telecommunications engineering from Telecom Bretagne, in France.
