Dynamic Spectrum Access Using Audio By: Samuel Henderson Amos Ajo Advised By: Dr. Dietrich Dr. Beex
Project Objective
This project demonstrates the application of Software Define Radio (SDR) and Cognitive Radio (CR) to perform Dynamic Spectrum Access (DSA) using an audio communication system.
What is Wireless Spectrum ?
Licensed User
Radio
Cell Tower
Federal Communications Commission
FCC
Cartoon pic Posted by John Klossner on Apr 18, 2013 at 12:10 PM
http://fcw.com/blogs/fcw-insider/2013/04/klossner_spectrum.aspx
Spectrum data collected using GNU Radio
Cognitive communication Group Project collecting data @ Radford park
Spectrum Utilization
Data collected from VT parking garage across Whitmore building by Cognitive Communication group
Spectrum Hole
Primary User @ 854 MHz Primary User Inactive @ 854 MHz
FFT Plot FFT Plot Spectrum Hole
Dynamic Spectrum Access (DSA)
• DSA: Ability of a Secondary User (SU) radio to opportunistically access spectrum allotted to a Primary User (PU).
• SDR: Are radios which are implemented and controlled in software.
• CR: SDR’s which can sense its environment and use software algorithms to configure and adjust itself dynamically
• Spectrum Sensing: Find spectrum holes
Primary User Link• PU transmitter sends binary data over a sound
wave– This is made possible by a process called digital
modulation.
• PU receiver demodulates this wave to recover the data
Primary User Link (cont.)
• Coded in MATLAB:– PU TX transmits periodically in bursts– Occupy a channel of spectrum centered at 4 kHz
Primary User Performance
• How accurately does PU receiver recover the bits? Bit Error Rate (BER)
• Of 100k bits, 1.59% are in error• Errant Data caused by carrier incoherence
00.05
0.10.15
0.20.25
0.30.35
0.40.45
0.5
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
BER
SAMPLES
Bit Error Rate Plot
Secondary User
• The SU is a transmitter whose task is to– Employ spectrum sensing to:
• Occupy PU channel when available.• Vacate PU channel when PU returns.
– Cognitive Radio! – Implemented in software using GNU Radio– Spectrum sensing and control logic scripted in
Python
System Performance
• The performance of our total system is measured by PU BER
• Factors influencing performance are:– Power of SU– Burst length of PU
Results
With 2 second PU burst length:-Avg. BER of 13.24%
With 3 second SU burst length:-Avg. BER of 5.24%
100k bits of data transmitted and received by PU
Conclusion
• In our research we accomplished the following:– Established a workable model for DSA audio
communication system– PU: Used MATLAB to transmit and receive BPSK
modulated data– SU: Used GNU Radio to implement SDR systems,
and learned to use Python to program implement a cognitive radio
Discussion
• Our research taught us that DSA algorithms, though promising and increasingly necessary, can be difficult to efficiently implement
• Smarter and faster algorithms for spectrum sensing and cognition can be developed to ensure non-intrusive navigation of the spectrum.
Further Work
• Implement proper carrier recovery in PU RX• Add more SU’s to the network• Implement entire system in RF using CORNET
Acknowledgements
• Special Thanks:– Mentorship of Dr. Beex and Dr. Dietrich– Assistance of Tamoghna Roy and Jason Snyder– NSF for sponsoring our research
References
• 1.) "Simulink Exercises for "Digital Communications: A Discrete-Time Approach," by M. Rice." Simulink Exercises for "Digital Communications: A Discrete-Time Approach," by M. Rice. N.p., n.d. Web. 25 July 2013.