Passive and Active Spectrum Sharing (PASS)

PASS is an NSF-funded project (ECCS-2030233)[1]. This page provides access to the noise-floor survey SigMF[2][3] data sets generated as part of the project.

The University of Colorado Boulder (CU) Passive and Active Spectrum Sharing (PASS) project directly addresses the related problems of protecting passive users while enabling secure, dynamic spectrum sharing between passive and active systems. The PASS project is an interdisciplinary collaboration with deep expertise in spectrum science, spectrum sharing, wireless systems, and system security.

Radio frequency (RF) spectrum has become a scarce resource. The Passive and Active Spectrum Sharing (PASS) project will systematically survey and characterize RF noise, and explore and evaluate alternatives for enabling spectrum sharing between passive and active systems. The PASS research will be a significant step forward in understanding and mitigating RF noise, and will enable more dynamic and efficient sharing of spectrum.

Our purpose behind a data standard is to enable data sharing and collaboration. Along with developing the standard and gathering data on RF noise in various environments, the PASS project will develop a prototype database for storing and retrieving RF data. Several sites exist which provide SigMF-based RF data for download. For example, researchers at the UC Berkeley SETI Institute have made available SigMF files from observations made with the Allen Telescope Array, an array of 42 dish antennas at the Hat Creek Observatory in California [4]. Additional datasets are available from Breakthrough Listen observations using the Green Bank Telescope. Deepsig.io has datasets in multiple formats, including hdf5, and Python ‘pickle’ files [5].

[1] PASS NSF Award Page, https://www.nsf.gov/awardsearch/showAward?AWD_ID=2030233&HistoricalAwards=false

[2] GNURadio/SigMF, Signal Metadata Format (SigMF), https://github.com/gnuradio/sigmf

[3] GNURadio: the free and open software radio ecosystem, https://www.gnuradio.org/

[4] SETI Institute, http://seti.berkeley.edu:8000/open-data/

[5] DeepSig, Inc., RF Datasets for Machine Learning, https://www.deepsig.io/datasets