Skip to main content
U.S. flag

An official website of the United States government

Dot gov

Official websites use .gov
A .gov website belongs to an official government organization in the United States.


Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.


Application of deep learning algorithms to 3.5 GHz spectrograms to characterize incumbent federal radar emissions.


NVIDIA® DGX™ Station, an AI Supercomputer
As federal agencies collaborate with industry to refine requirements for detection of federal incumbent radar signals in the 3550-3650 MHz portion of the Citizens Broad Radio Service (CBRS) band, there is a need to understand potential performance limitations of Environmental Sensing Capability (ESC) detectors and to develop sound methods for performance evaluation.  Building on the completed Waveform Measurements of Radars Operating in the 3.5 GHz Band project, we are studying radar detection with spectrograms and time-domain waveforms.
Specifically, using advanced methods of such as deep learning, NASCTN is investigating detection algorithms for federal incumbent radar signals in the 3550-3650 MHz band.  Recent work using spectrograms was reported in a journal paper Deep Learning Classification of 3.5 GHz Band Spectrograms with Applications to Spectrum Sensing and an accompanying technical note Spectrum Occupancy and Ambient Power Distributions for the 3.5 GHz Band Estimated from Observations at Point Loma and Fort Story.  The results demonstrated that training deep learning algorithms with real-world measurements can outperform traditional energy detection methods.  Moreover, they demonstrated the utility of machine learning for estimating spectrum occupancy and ambient power distributions.   
The outcomes of this project facilitate the development of technical solutions and policies for spectrum sharing in the 3.5 GHz band, including standards specifications and implementation strategies considered for approval by the Wireless Innovation Form Spectrum Sharing Committee (WINNF SSC). For more information, please see the NIST CBRS project webpage and the NIST press release on recent publications. 

For this and future NASCTN projects involving machine learning, NASCTN has acquired a high-performance workstation (pictured) specifically designed for deep learning applications.

Created January 30, 2018, Updated April 30, 2019