Cognitive Radio Evolution from Agile Platforms to Omniscient Networks: the Road from Dreams to Prototypes Charles W. Bostian Virginia Tech Radio Hardware Awareness Sensing and Modeling Adapting Evolution and Optimization Learning Building and Retaining Knowledge

Acknowledgements This project is supported by Award No IJ- CX-K017 awarded by the National Institute of Justice, Office of Justice Programs, US Department of Justice. The opinions, findings, and conclusions or recommendations expressed in this publication/program/exhibition are those of the author(s) and do not necessarily reflect the views of the Department of Justice. This material is based upon work supported by the National Science Foundation under Grant No. CNS Any opinions, findings and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation (NSF). This work is also supported by Air Force Institute of Technology (AFIT). The views expressed in this article are those of the author and do not reflect the official policy or position of the Air Force, Department of Defense or the U.S. Government. Center for Wireless Telecommunications The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of the Defense Advanced Research Projects Agency or the U.S. Government. Defense Advanced Research Projects Agency Strategic Technology Office DARPA Order AF89-00

Acknowledgment: The VT Team

An (old) radio guy’s vision of cognitive radio: A universal transceiver (all modes and all frequencies) capable of discovering radios like itself and working cooperatively to negotiate frequencies, waveforms, and protocols to optimize performance subject to user needs based on the radio’s awareness of its environment and its past experience.

The VT Public Safety Cognitive Radio Recognize any P25 Phase 1 waveforms Identify known networks Interoperate with legacy networks Provide a gateway between incompatible networks Serve as a repeater when necessary – useful when infrastructure has been destroyed or does not exist.

In developing this prototype, we have solved some hard problems in rapid reconfiguration of a radio platform and in signal recognition and synchronization. Configure this in real time and operate it. Find a signal of interest

Cognitive Engine + SDR = Cognitive Radio

The relatively easier part – realization of the cognitive engine General Implementations: A restricted implementation: the VT Public Safety Cognitive Radio

FCC Worries: Code correctness, insecure memory accesses, tamper resistance. Off-line unit testing and formal verification plus light-weight yet effective anti- tampering methods to ensure that any module replacement is compliant. Ensures that any replacement of the modules, including over-the-air updates is done by trusted parties.

The harder part – building a “universal” radio platform

The GPP Problem – Latency and Inability to Control Timing OK for narrowband waveforms with simple timing requirements. A real problem for wideband waveforms and MACs requiring precise timing.

The solution that we are developing now: A hybrid architecture containing fixed and reconfigurable subsystems. Embedded GPP performs cognitive functions and determines radio configuration Reconfigurable FPGA and ASICs perform PHY and MAC layer operations Accelerators implement application layer functions

System Overview of PSCR (hybrid implementation)‏ FPGA DSPGPPAnalog RF ADC/DAC Spectrum Sweeper Signal Classifier Waveform Recognition RF front-end PGAADCDDC Waveform Knowledge Base Case-based Waveform Solution Maker GUI & Center Controller configure FilterGain Demod FEC Decoder De-packet MAC Carrier Sense Algorithm MAC Layer Protocol FilterGain Mod FEC Encoder Packet RF front-end PGAADCDDC RF front-end PGADUC DAC VTSDCSS Binary Source Binary Data RX TX

My student Ying Wang will demonstrate some of our current spectrum, waveform identification, and radio configuration technology as part of this meeting. She and my student Qinqin Chen invented the system we will demonstrate and many others in our group contributed to the implementation.

What is wrong with this picture? One radio platform can’t do it all. Focuses on interactions of two nodes. Ignores network issues. Ignores applications that the radio will run.

The reality: Multiple networks Multiple protocols Multiple applications Dynamic Spectrum Access All this leads to the concept of an application and network driven integrated architecture for a cognitive node

Architecture for An Application and Network Driven Integrated Cognitive Node Conceived by my student Feng Andrew Ge to capture the overall cognitive radio efforts of our group.

An important part of the implementation: The Universal Cognitive Gateway, dissertation topic of my student Qinqin Chen

Another application: dissertation work of my student Mark Silvius

Dynamic Cellular Cognitive Radio (Ying Wang) 700M Hz Application Scenario Basic Concept Software Structure PPCN PCN CMT

Contact Information Charles W. Bostian Alumni Distinguished Prof. Virginia Tech