A Design Workflow for Software Defined Radios Frederick Weidling
[email protected] Presented August 22, 2007 to: Dr. Gary J. Minden Dr. Joseph B. Evans Dr. Alexander M. Wyglinski
Acknowledgements • Defense Committee • Dr. Gary J. Minden (Chair) • Dr. Joseph B. Evans • Dr. Alexander M. Wyglinski
• KU Agile Radio Group • Leon Searl and Dan DePardo • Jordan Guffey, Rory Petty, Tim Newman, Brian Cordill, Dinesh Datla, Rakesh Rajbanshi, Qi Chen, and Megan Peck
2
Outline • Background • Software Defined Radios (SDR) • Dynamic Spectrum Access (DSA)
• Proposed Research • A workflow for SDRs
• Validation • Apply design workflow to KUAR • Generate systems
• Extension • Generic SDR API • Hardware Agnostic Cognitive Network
• Conclusion 3
What is an SDR? •
Ideal SDR • • •
•
Direct conversion between digital samples and analog waveform Use only general purpose processors (GPP) Unrealistic for high frequencies – cost, processor speed, algorithm complexity
Practical SDR • •
Baseband processing, analog components translate to transmission band Optimizations using digital signal processors (DSP) and field programmable gate arrays (FPGA)
4
Why do we need SDRs? • FCC “Command-andControl” Policy Limits • What can be transmitted • Who can transmit • Where they can transmit
• Results in apparent spectrum scarcity • Restricts research and novel services • New York 13.1% utilization 3.0 MHz – 3.0 GHz • TV utilization often less than 50%
• Solution: SDRs 5
Dynamic Spectrum Access • •
SDR detects unused frequencies (whitespace) and utilizes them Cognitive Radio (CR) • •
•
Radio which adapts to its environment Cognitive algorithms implemented on SDR platforms
Dynamic Spectrum Access (DSA) • •
Avoid licensed users Utilize whitespace
Image From “Spectrum Pooling an Innovative Strategy for the Enhancement of Spectrum Efficiency,” by T. Weiss and F. Jondral
6
Current SDR Technologies • Universal Serial Radio Peripheral (USRP) • Generic RF front-end • Modular transmit/receive bands
• GNU Radio • Open source SDR software • Only supports general purpose processors • Based around USRP
• Global Mobile (GloMo) API • Generic radio modem API
• Software Communications Architecture (SCA) • CORBA based communications model for SDR modules
• Joint Tactical Radio System (JTRS) • SDR platform commissioned by Defense Advanced Research Projects Agency (DARPA) • Built using SCA 7
The KU Agile Radio (KUAR) • Digital Board • • • • • •
1.4 GHz Pentium M 1 GB RAM & 8 GB microdrive PCI Express, USB, 1 Gb ethernet 160 MSPS Digital-to-Analog Converter (DAC) 105 MSPS Analog-to-Digital Converter (ADC) Xilinx Virtex II Pro 30 FPGA
• RF Front-End • • • • • •
Modular 5.25 GHz – 5.85 GHz UNII Research Band Receiver sensitivity -100 dBm Transmitter power +25 dBm Quadrature modulation & demodulation Baseband bandwidth of 30 MHz
8
SDR: A Federation of Components • Variety of components must work in unison to transmit/receive • • • • •
General purpose processors Special purpose processors (FPGA, Microcontrollers) Frequency synthesizers Modulators/Demodulators Attenuators
• Different types of design problems on SDR • Cognitive network design • Real time data management • Communication systems
9
Development Domains • Break the problem into domains • Reconfigurable hardware domain aimed at Communications Engineers • Embedded software domain aimed at System Engineers • Radio management domain aimed at Network Engineers and general radio maintenance
• Each domain has its own workflow requirements • Different set of development tools • Each domain’s toolset must support designing, implementing, and verifying modules • Domains interact but implementations are independent 10
Reconfigurable Hardware Domain • Implement physical layer communication systems • Direct access to ADC/DAC • Signal processing optimizations possible in FPGA
• Generic memory elements need to be provided • Bus controller • Buffers (FIFOs, RAMs) • Registers (Status, Control)
11
Embedded Software Domain • Manage hardware interfaces and data streams • Components • Dynamic Hardware Interface: configure/communicate with reconfigurable hardware • Static Hardware Interface: communicate with RF modules and sensors • Spectrum Sensing: perform spectral sweeps • Waveform Protocols: physical layer protocols
12
Radio Management Domain • Radio Management • Diagnostic tools • Network protocols • Spectrum access protocols
• User interface enables network tests • Control multiple radio nodes • Define co-operative tests • Display results
13
Implementation on the KUAR •
Reconfigurable Hardware Domain • •
•
Xilinx Virtex II FPGA PCI, PCIe, or USB bus
Embedded Software Domain •
•
1.4 GHz Pentium M with 1 GB RAM
Radio Management Domain •
Remote KUAR interface AD8347 I-Q DEMODULATOR 800 MHz - 2.7 GHz
(-100dBm)
LTC2284 Dual 14-Bit 105 Msps
+8dBi 1.850-2.450 GHz
5.250-5.850 GHz
BASEBAND I
-3dB LNA -4dB
+19dB
14 LNA
R
0 90
I L
+19dB
-5dB
-10dB
-5dB GAIN CONTROL
3.4 GHz
ADC
BASEBAND Q
I DET
DPB Input
Q DET
Jumper Select
SPI Bus
RX IF GAIN CONTROL
Audio Output
PCI
256k x 32 bit
Keyboard Output Memory
16.0 MHz
CPU/Mis. Memory
SDRAM 1 GB
PCI-eXpress 1 Lane
Xilinx Virtex II Pro FPGA
256k x 32 bit
ADF4113 -5dB
USB
RX LO 1
LO 3
+9dB
Audio Input
Li Battery Input Memory
ADF4360-1 2.150-2.450 GHz
AD5601 6 BIT DAC
3 dB POWER DIVDER -2dB
RTC backup power
14
ADF4360-2 1.850-2.150 GHz
MC68HC08 Microcontroller
I²C
Front Panel
Diff.
BW =30 MHz
Diff.
Active RX Antenna Module
Kontron ETXexpress 1.4 Ghz Pentium-M
512k x 32 bit
USB Mouse USB GPS VGA Monitor
SMV3300A 1000BaseT Network TX LO 2
Microstrip
Lumped
ADF4360-2 1.850-2.150 GHz
AD5601 6 BIT DAC
Active TX Antenna Module
TX IF GAIN CONTROL
(+25dBm) 3.4 GHz
+8dBi
AD8349 I-Q MODULATOR 700 MHz - 2.7 GHz
ADT7461 Temperature Sensor
ADF4360-1 2.150-2.450 GHz
(+21dBm)
(+15dBm) (+9dBm)
(-2dBm) (-8dBm)
(-3dBm) (+7dBm) R
+17dB
-5dB OPTIONAL
Σ
I
-10dB
+9dB
-5dB
0 90
I 2C I2 C
USB Peripheral Controller
Flash Carrier Cnfg
29
16 Diff.
BW =30 MHz
DAC BASEBAND Q
For use with passive antenna
Cypress CY7C68013A
IDE
(+3dBm)
L
PA -4dB
FPGA Configuration
AD9777 Dual 16-Bit
BASEBAND I (+17dBm)
12
USB
16 SPI
CF Connector
3 Diff. USB
5.250-5.850 GHz
USB 1.850-2.450 GHz
CF+ 6 GB uDrive
I2 C I2C
14
Communication Systems • Development tools • Develop using Matlab/Simulink • Implement using Xilinx VHDL • Verify with Simulink, Modelsim, and hardware test bench
• Test cases translated down flow • Verified components added to shared library • Systems Built • BPSK, MQAM, OFDM, Spectrum Analyzer, …
15
Support Libraries •
Development tools • •
•
FPGA Library • •
•
Configure bit file Read from and write to hardware configurations
RF Control Library • • •
•
Implement using GCC & GNU Makefile User validation & cUnit
Manage the RF front-end Set and get frequencies and gains Query hardware capabilities
Monitor Library •
Monitor system temperatures
16
The KUAR Control Panel • Control multiple radios from one interface • Radio status • Configure radio profile • Control RF front-end
• Define network and single radio tests • Test definitions XML based
• Extendable interface • Java based remote KUAR API • Programmable test window
17
KUAR Control Panel: Visualizations
18
Achievements • Developed design workflows to handle • Reconfigurable hardware • Embedded software • Radio management
• Implemented on the KUAR • Communications systems – BPSK, QPSK, OFDM • KUAR API – RF Control, FPGA Control, Monitoring libraries • Analytical systems – Whitespace detector, spectrum analyzer
19
Enabling Generic SDR Development • Physical layer implementations often tailored to system • Dependent on sampling rates • Optimized for parallel and signal processing • Hard real-time deadlines
• Cognitive algorithms are complex but more generic • Wide variety of issues: whitespace, battery life, channel characteristics, noise power, … • Complex solutions: frequency selective modulation, expert systems, genetic algorithms, ontology based systems, …
• Should be able to re-use cognitive algorithms 20
Hardware Agnostic Network Stack • SDR Platform API handles the physical layer implementations • Unifying Layer exposes generic SDR interface • Traffic Scheduler manages access to shared RF front-end • Cognitive Network Layer is independent of SDR implementation platform
21
Unifying Layer • Hardware Manager • List hardware capabilities • Control RF front-end
• Protocol Manager • List waveform protocols • Manage waveform protocols
• Spectrum Sensor • Handle spectral sweeps
• Data Stream Manager • Similar to GloMo API • Interact with configured waveform protocols
22
Traffic Scheduler •
Schedule packets to channels • • •
•
Channel is a hardware digital/analog chain Scheduled packet includes protocol, frequency, power, and time frame Each channel has an allocation list
Packets are scheduled in a sliding window • •
Packets may be scheduled periodic or single-shot Call back interface for error states and packet completion
23
Traffic Scheduler: State Diagram
24
Conclusion • Achievements • • • •
Developed a design workflow for SDRs Workflow used to develop systems for KUAR Proposed generic SDR API (Unifying Layer) Described infrastructure to support hardware agnostic cognitive networks
• Future work • Implement hardware agnostic network stack • Develop more robust cognitive algorithms for KUAR • Apply workflow to another SDR platform
25
Resources •
SDR Resources • •
•
Publications •
•
•
•
•
SDR Forum: http://www.sdrforum.org/ Wikipedia Article: http://en.wikipedia.org/wiki/Software-defined_radio G. J. Minden, J. B. Evans, L. Searl, D. DePardo, R. Rajbanshi, J. Guffey, Qi Chen, T. Newman, V. R. Petty, F. Weidling, M. Lehnherr, B. Cordill, D. Datla, B. Barker, A. M. Wyglinski, A. Agah, "An Agile Radio for Wireless Innovation" IEEE Communications Magazine, Vol. 45, Issue 5, pp 113-121, May, 2007. V. R. Petty, R. Rajbanshi, D. Datla, F. Weidling, D. DePardo, P. J. Kolodzy, M. J. Marcus, A. M. Wyglinski, J. B. Evans, G. J. Minden, J. A. Roberts, "Feasibility of Dynamic Spectrum Access in Underutilized Television Bands" in Proceedings of the 2nd IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks (DySpan 2007), (Dublin, Ireland), April 2007. G. J. Minden, J. B. Evans, L. Searl, D. DePardo, V. R. Petty, R. Rajbanshi, T. Newman, Q. Chen, F. Weidling, J. Guffey, D. Datla, B. Barker, M. Peck, B. Cordill, A. M. Wyglinski and A. Agah, "KUAR: A Flexible Software-Defined Radio Development Platform" in Proceedings of the 2nd IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks (DySpan 2007), (Dublin, Ireland), April 2007. F. Weidling, D. Datla, V. Petty, P. Krishnan, and G. J. Minden, “A Framework for RF Spectrum Measurements and Analysis,” in Proceedings of the 1st IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks (DySpan 2005), (Baltimore, MD, USA), pp. 573– 576, Nov. 2005.
Questions?
26
