ISART 2016: Spectrum Forensics — Synopsis of Prequel and Panel Topics

Prequel Tutorial: Spectrum Forensics Case Studies

The prequel tutorial provides essential background for ISART participants who are not familiar with the FCC Enforcement Bureau’s criteria and process for both civil and criminal spectrum interference investigations and enforcement actions.

Moderator: Rebecca Dorch, NTIA/ITS.D, Boulder Labs
Margaret Egler – former FCC Enforcement Bureau Western Region Counsel
Greg Hermes – FCC Enforcement Bureau, Equipment Development Group Director
Jim Higgins – NASA, Spectrum Management Engineer, former FCC EB Northeast Deputy Regional Director

Policy Considerations

This panel will explore the implications of technological developments on spectrum policy as they relate to the forensic analysis of radio interference. As telecommunications systems increase in complexity and quantity, and intentional misuse of spectrum is democratized through ready access to inexpensive software defined radios and comprehensive code repositories, the future frequency, severity, and consequence of interference is of increasing concern. This panel will consider commercial wireless industry approaches and existing FCC and NTIA frameworks for identification and mitigation of harmful interference through both legal and technical channels. Considering past experiences with the Internet, future looking perspectives on the intentional criminal misuse of spectrum will also be considered. As exemplified in other regulatory frameworks, spectrum forensics will require the development of new disciplines to collect and interpret spectrum related artifacts to pinpoint sources of interference, initiate corrective action and, in some cases, prosecute offenders.

Moderator: Eric Nelson, NTIA/ITS.M, Boulder Labs

Greg Shannon - OSTP
Julius Knapp – FCC Office of Engineering and Technology Chief
Paul Anuszkiewicz - CTIA VP of Spectrum Planning
Wayne Phoel – DARPA
Peter Tenhula – NTIA/OSM

Spectrum Monitoring – Purposeful Data Collection

There are economic and technical constraints that limit the scope of data available for spectrum forensics. Good architectural design can maximize limited resources and provide novel processes for data acquisition, integration, management, and use. The Spectrum Monitoring panel will be comprised of spectrum data architects and users. They will discuss, evaluate, and weigh ideas for a national infrastructure that (1) supports data aggregation from diverse sources and (2) establishes a logical separation between infrastructure and information, in order to promote specialization and economies of scale in areas such as sensor design, data visualization, and information extraction. The panel will start with a strawman spectrum monitoring architecture. It will evaluate ideas, options, and recommendations for design requirements, business models, and next steps.

Moderator: Ken Baker, NTIA/ITS.T, Boulder Labs
Roger Hislop - Internet Solutions, South Africa
Jesse Caulfield - Key Bridge, President
Dennis Roberson – Roberson and Associates
Steven Gripppando – NOAA
John Chapin - Carnegie Mellon University
Mathieu Gemme – CRC
Mike Cotton – ITS

Enabling Technologies and Standards

The technical enablers of spectrum forensics span monitoring instrumentation, computational resources, and standards for recording spectrum measurements.  Monitoring instrumentation includes not only the downconverters, digitizers, and signal analyzers capable of analyzing the wide bandwidths of current and future wireless broadband systems, but also antenna arrays and related technologies for direction finding and localization of emitters.  In addition, storage architectures will be needed to record vast amounts of data and make them readily available for subsequent analysis.  Computational resources will be needed to detect anomalous spectrum behavior in real time as well as to analyze captured data off-line.  Finally, standard formats for signal waveforms and associated metadata would bring the benefits of open architectures to future spectrum forensics solutions.  This panel will report on the state-of-the-art in the technologies and standards that make spectrum forensics possible and identify gaps where future development and standardization are needed.

Moderator: Michael Souryal, NIST/CTL
Paul Denisowski – Rohde & Schwarz
Robert Normoyle – JHU/APL, VITA-49.0
Nebu John Mathai – Cognitive System Corp.
Apurva Mody – BAE Systems, IEEE 802.22
Pablo Tacconi – MAXSiA

Data Analytics

This panel will focus on the mathematical techniques for analyzing captured signal data, with an emphasis on machine learning, data mining and artificial intelligence techniques. Classical pattern recognition and other statistically-based methods may also be included. The limitations and potential pitfalls of each technique will be considered along with the advantages and capabilities. Related to this topic is the question of what type and quantity of data need to be captured and retained in order to use each technique. Time and frequency resolution and appropriate quantization are examples of this. The panel will not only discuss current research results but will help identify specific topics and techniques where future research is still needed.

Moderator: Tim Hall, NIST/CTL
Tim O’Shea – Virginia Tech
Tom Rondeau – DARPA
Al Gasiewski – Science Band, University of Colorado
Mitch Kokar – Northeastern University
Dirk Grunwald – University of Colorado