Software Defined Radio and Cognitive Radio · Software Defined Radio and Cognitive Radio:...

Software Defined Radio and Cognitive Radio: Implementations & Test-Beds

Alexander M. Wyglinski, Ph.D.

Associate Professor of Electrical and Computer Engineering

IEEE Vehicular Technology Society Distinguished Lecturer

Worcester Polytechnic Institute

Presentation Outline

• Introduction

• The Information Age

• Wireless Innovation Laboratory

• Spectrally Agile Waveforms

• SDR and CubeSats

• Concluding Remarks

• More Information

Introduction


Definition: Cognitive radio

A cognitive radio is a kind of two-way radio that automatically changes its transmission or reception parameters, in such a way that the entire wireless communication network - of which it is a node -communicates efficiently, while avoiding interference with licensed or unlicensed users.

A cognitive radio, as defined by the researchers at Virginia Polytechnic Institute and State University, is "a software defined radio with a cognitive engine brain".

http://en.wikipedia.org/wiki/Cognitive_radio


So what is Cognitive Radio?

+ ?


Intelligence

Optim

izat

ion

Adaptation

Environmental Awareness

AgileFlexible

Spectrally Efficient

Learning

Cooperative

Op

po

rtu

nis

tic

Autonomous

Dynamic Spectrum Access

Cognitive radio means …

The Information Age


Several Key Innovators

Marconi Shannon ShockleyBardeen Brittain

Wireless Transmission

Digital Communications

Transistors

Source: Wikipedia


Progress of Technology


Evolution of Wireless Systems“C

ognitiv

e Rad

io C

om

munic

atio

ns

and N

etw

ork

s: P

rinci

ple

s an

d P

ract

ice”

By

A.

M.

Wyg

linsk

i, M

. N

ekove

e, Y

. T.

Hou (

Els

evie

r, D

ecem

ber

2009)


Quick Survey

How many of you:─Own a cell phone?─Use a laptop with WiFi?─Use an ATM?─Fly on a plane?─Traveled in a car?

11


Increasing Demand262 Million Subscribers!

Source: CTIA


Increasing Demand

1.1 Trillion Minutes!

Source: CTIA


US Spectrum Scarcity!

So

urc

e:

NT

IA


Across the Pond in the UK!

So

urc

e:

Ro

ke M

an

or


Oh Canada!S

ou

rce:

Ind

ust

ry C

an

ad

a


Apparent Scarcity

• Measurement studies have shown that in both the time and frequency domains that spectrum is underutilized

Spectrum measurement across the 900 kHz –1 GHz band (Lawrence, KS, USA)

Spectrum Holes


Potential Solution

Spectrum measurement across the 900 kHz –1 GHz band (Lawrence, KS, USA)

• Dynamic Spectrum Access (DSA)

Fill with secondary

users


Enabling Transmission Agility

PROGRAMMABLE FIXED


Sample SDR Platforms

Universal Software Radio Peripheral 2 (USRP2) Unit.

COSMIAC FPGA board currently being retrofitted for better memory access, to add USB functionality and

to make the board SPA compatible.


Mitola & Cognitive Radio

Mitola


Cognitive Radio: A Black Box Model

What you want

What you see

What you can do

What you can tune


Flexible RF Front Ends Needed

88 MHz

5.8 GHz

Can I do this with just one

RF front end?


RF MEMS Can Help!

• A single RF front end would not normally be able to support a very wide frequency range of operations─ Radio Frequency

Micro-Electro-Mechanical Systems (RF MEMS) can be used to “tune” the RF front end to the corresponding frequency Source: Wireless ICs and MEMS

Laboratory, McGill University

Close-Up of MEMS Tunable LC Filter.

CapacitorInductor


RF MEMS

• RF MEMS can be used to implement:─ Antennas (e.g., fractal antennas)─ Filters (e.g., tunable RF bandpass filters)─ Oscillators

• Real-time operations very difficult to support─ Time needed to physically change configurations on the

order of seconds Compared to the rate at which data is transmitted, this is

considered to be ages!

Wireless Innovation Laboratory


• Associate Professor, WPI ECE

• Director, Wireless Innovation Laboratory─ 8 Ph.D. students, 5 M.S. students

• Distinguished Lecturer, IEEE Vehicular Technology Society (2012-2014)

• Technical Editor, IEEE Communications Magazine

• Editor, IEEE Transactions on Wireless Communications

• General Co-Chair, IEEE VTC 2015-Fall (Boston, MA, USA)

• ~35 journal publications, ~75 conference papers, 9 book chapters, 2 books

• Served or is serving as PI/co-PI for several federal and industrial grants

Who is Alex Wyglinski?


~60 km

Where is WPI?


• Founded in 1865; 3rd oldest US polytechnic

─ Model for most engineering schools in US

─ Nationally ranked as 64th Best College in U.S. (2011 US News Rankings)

─ Voted one of the Top 10 Best U.S. Colleges for “Young Einsteins” by Unigo (together with MIT, CalTech, Princeton, Dartmouth, Stanford, Johns Hopkins, Case Western, Georgia Tech, and Cornell)

• 3800 students & 220 faculty─ ECE: 318 undegraduate, 250

graduate, 21 faculty

What is WPI?


What is “Lehr und Kunst”?

• Project-based education at the core of “The WPI Plan”

• Many ECE students conduct their senior capstone design projects (called “Major Qualifying Projects”or MQPs) at off-campus locations:─ MIT Lincoln Laboratory─ MITRE (Bedford Campus)─ General Dynamics (Groton Campus)─ Silicon Valley─ Wall Street─ Etc …


Wireless Innovation Laboratory

• 2 USRP (Version 1) software-defined radio platforms

• 14 USRP (Version 2) software-defined radio platforms

• 15 USRP (Version N210) software-defined radio platforms

• 1 Agilent CSA N1996A 0-3 GHz spectrum analyzer (with battery packs)

• 1 Mini-discone antenna (100 – 1600 MHz, with 3’ tripod)

• 1 WG horn antenna (0.7 – 18.0 GHz, with tripod)

• 1 Xilinx Virtex 5 HW-V5-ML506-UNI-G Prototyping Board

• 25 complete licenses of MATLAB and Simulink with associated toolboxes and blocksets

• 2 OPNET licenses


WI Lab External Sponsorship


SDR Activities at WPI

Photograph of a supervised laboratory session for ECE4305 “Software-Defined

Radio Systems and Analysis” during February 2011.

Screen capture of a functioning ECE4305 course design project in MATLAB showing four SDR units

forming an ad hoc wireless network.

http://www.sdr.wpi.edu/

Spectrally Agile Waveforms


Opportunistic Spectrum Access

• Opportunistic spectrum access (OSA) is a significant paradigm shift in the way wireless spectrum is accessed─ Instead of PUs possessing exclusive access to licensed

spectrum, SUs can temporarily borrow unoccupied frequency bands

─ SUs must respect the incumbent rights of the PUs with respect to their licensed spectrum

• OSA enables greater spectral efficiency and facilitates greater user and bandwidth capacity


• The utilization efficiency of “prime” wireless spectrum has been shown to be poor

A snapshot of PSD from 88 MHz to 2686 MHz measured on July 11th 2008 in Worcester, MA (N42o16.36602, W71o48.46548)

OSA Motivation

A. M. Wyglinski, M. Nekovee, Y. T. Hou (Eds.). “Cognitive Radio Communications and Networks: Principles and Practice.” (Chapter 6) Academic Press, December 2009.

empty empty emptyempty


Leveraging the Electrospace

Several dimensions of the electrospace include space, time,

and frequency, although there do

exist others such as code, polarization,

and directional.

“Cognitiv

e Rad

io C

om

munic

atio

ns

and N

etw

ork

s: P

rinci

ple

s an

d P

ract

ice”

By

A.

M.

Wyg

linsk

i, M

. N

ekove

e, Y

. T.

Hou (

Els

evie

r, D

ecem

ber

2009)


Several Possible Approaches

• Secondary transmission in licensed spectrum can be classified into three categories:– Cooperative Approach

• Primary and secondary users coordinate with each other regarding spectrum usage

– Underlay Approach• Secondary signals transmitted at very low power

spectral density; undetected by primary users• e.g., ultra wideband (UWB)

– Overlay Systems• Secondary signals fill in the spectrum unoccupied by

primary users


Spectral Opportunities!



empty empty emptyempty


Underlay Solution



underlay transmissions


Overlay Solution



overlay transmissions


Multicarrier-Based OSA

• Multicarrier modulation is a variant of the conventional frequency division multiplexing (FDM)─ Orthogonal Frequency Division Multiplexing

(OFDM) an efficient form of multicarrier modulation

• In order to utilize unused portions of licensed spectrum, several subcarriers can be turned OFF to avoid interfering with the primary signals

• Each subcarrier experiences flat-fading and hence high data-rates are possible if several unused bands of secondary spectrum are available


Multicarrier Overlay Solution



multicarrier overlay transmissions


Spectral Agility In Action!

PU signal!

multicarrier overlay SU transmission wraps around PU

As seen in this close-up of the multicarrier overlay transmission, subcarriers located within the vicinity of

a PU can be deactivated in order to avoid interference

with that signal.


Spectrally Agile Multicarrier

H.

Boguck

a, A

. M

. W

yglin

ski, S

. Pa

gadar

ai, A

. Klik

s. “

Spec

trally

Agile

Multic

arr

ier

Wav

eform

s fo

r O

pport

unis

tic

Wirel

ess

Acc

ess”

. IE

EE C

om

munic

atio

ns

Mag

azin

e, J

une

2011.


Major Issue: Out-of-band Emission

• Out-of-band (OOB) interference problem with OFDM-based cognitive radios

• Power spectral density of the transmit signal over one subcarrier:

• Mean relative interference to a neighboring legacy system subband:


Sinc Pulses Have High OOB Levels!

-6 -4 -2 0 1 2 4 6-50

-45

-40

-35

-30

-25

-20

-15

-10

-5

0

5

Subcarrier Index

Nor

mal

ized

pow

er s

pect

rum

(in

dB

) OFDM carrier spacing

Interference power to the first adjacent

sub-band

−10 −5 0 5 10 15 20 25 30

0

0.5

1

Other

Transmissions

Other

Transmissions

Other

Transmissions

Subcarrier index

Nor

mal

ized

am

plitu

de

−10 −5 0 5 10 15 20 25 30

−60

−40

−20

0

Subcarrier index

Nor

mal

ized

pow

er in

dB

m

OtherTransmissions

OtherTransmissions

OtherTransmissions

OOB

OOB


Several Solutions

• Cancellation Carriers─ Non-data bearing subcarriers whose phase and amplitude

values cancel OOB

• Modulated Filter Banks─ Attenuates OOB in stopband region

• Combine cancellation carriers (CCs) with modulated filter banks (MFBs) to attenuate OOB emissions


Cancellation Carriers


Filtering with CCs

−4 −2 0 2 4 6 8 10 12 14 16

−0.2

0

0.2

0.4

0.6

0.8

1

Subcarrier Index

Nor

mal

ized

Am

plitu

de

Raised−Cosine Filter

CC1 CC2


Hardware Experimentation

Photograph of a spectrally agile wireless transceiver test-bed at Poznan

University of Technology, Poznan, Poland.

Photograph of a spectrally agile wireless transceiver test-bed at Worcester Polytechnic Institute,

Worcester, MA, USA.


Spectrally Agile Waveform ResultsP.

Kry

szki

ewic

z, H

. Boguck

a, A

. M

. W

yglin

ski. "

Prote

ctio

n o

f Pr

imar

y U

sers

in D

ynam

ically

Vary

ing R

adio

Envi

ronm

ent:

Pr

actica

l Solu

tions

and C

halle

nges

." Acc

epte

d for

public

atio

n in

the

EU

RASIP

Journ

al o

n W

irel

ess

Com

munic

atio

ns

and

Net

work

ing

Dec

ember

23

2011


Performance with notches of different size

53


Performance with different number of notches

54


Digital Pre-Distortion Results

55

Output PSD of PA model

0 0.05 0.1 0.15 0.2 0.250

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

PA Input Magnitude

Nor

mal

ized

Out

put M

agni

tude

Nonlinear PA modelLinearized PA model

−1 −0.5 0 0.5 1−160

−140

−120

−100

−80

−60

−40

−20

0

20

Normalized Frequency (x π rad/sample)

Pow

er S

pect

ral D

ensi

ty [d

B]

1st ILA iteration2nd ILA iteration3rd ILA iteration

AM/AM

Zhu Fu, Lauri Anttila, Mikko Valkama and Alexander M. WyglinskiA. M. Wyglinski. "Digital Pre-distortion of Power Amplifier Impairments in Spectrally Agile Transmissions." Proceedings of the 2012 IEEE Sarnoff

Symposium, Newark, NJ, USA, May 2012.

CubeSat SDR

56


Motivation

• Challenges associated with time-varying satellite communication links

• Leverage existing technologies in order to enhance transmission quality via an efficient, low-cost approach─Software-defined radio (SDR) employed

to achieve highly responsive space platforms


Proposed Solution

• CubeSat-based SDR platform prototype developed─ Custom designed FPGA board possessing Spartan 3A

XC3S1400A device incorporated into CubeSat platform ─ Contains customized variant of commercially available

Universal Software Radio Peripheral (USRP) SDR platform

• Several satellite communication baseband modules implemented for FPGA─ Migrated to COSMIAC Small Form Factor CubeSat FPGA

Board (CCFB) possessing customized USRP SDR


Project Vision

• Creation of adaptive satellite communications link based on a combination of reconfigurable FPGA-based software-defined radio (SDR) platforms and cognitive radio concept─ Enables “data throttling” for maximizing data

throughput─CubeSat SDR platforms aware of environmental

conditions


COSMIAC FPGA

1”

COSMIAC Xilinx Spartan 3A FPGA


COSMIAC CubeSat Modules

COSMIAC CubeSat FPGA Board with Sensor and Power Daughtercards (no RF daughtercards are present in this photo)

FPGA Board

Optical SensorBoard

Power Board


FPGA-Based Embedded Controller Board

Spartan 3AFPGA

4 GbNAND Flash

Memory

2 GbDDR2 SDRAM

Memory

LEDs &Switch

JTAG

RS232Level Shift

EthernetLAN8700

1 MbNOR Flash

Config Memory

USB2USB3300

AT90 (SPA-U)Microcontroller

SPA-UConnector

SPA-1Header

Connectors to Daughter Card


MicroBlaze

USB2 Controller

SPI Controller

I2C Controller

Modified USRP

Flash Memory Controller

DDR2 Memory Controller

Ethernet ControllerRS232 UARTRS232 UART

(AT90)

USB Data

RFDaughte

Card


Laboratory Testing

Current SatCom laboratory setup at COSMIAC

Concluding Remarks


These are interesting times!

• Numerous advances in cognitive radio, dynamic spectrum access, and software-defined radio have recently occurred─ Secondary access of digital TV spectrum─ Ratification of IEEE 802.22, IEEE 802.11af standards

• Today’s wireless landscape is quickly changing due to new capabilities of wireless transceiver devices─ Largely due to smaller, faster processing devices resulting

from applications such as smart phones


Still room for improvement

• There still exists a substantial amount of research that is needed to make future wireless devices such as cognitive radio more reliable─ Ensuring minimal interference to other wireless

transmissions─ Enabling real-time decision-making and transmission

operations─ Making RF spectrum access more reliable for everyone

involved

More Information


Contact Info

Professor Alexander Wyglinski

Department of Electrical and Computer EngineeringWorcester Polytechnic Institute

Atwater Kent Laboratories, Room AK230

508-831-5061

[email protected]

http://www.wireless.wpi.edu/


Cognitive Radio Textbook

Available since December 2009 (Academic Press)

20 chapters

End-of-chapter problems (with solutions guide)

Presentation slides for most chapters

Covers physical and network layers, in addition to current platforms and standards


New SDR Textbook

• January 2013 Publications Date (Artech House Publishers)

• 9 comprehensive chapters ─ Fundamentals in signals &

systems, probability, and digital communications

─ “Hands on” approach to learning digital communication concepts using SDR and Simulink

─ End-of-chapter problems─ Corresponding course

lecture slides

http://www.sdr.wpi.edu/

Software Defined Radio and Cognitive Radio · Software Defined Radio and Cognitive Radio:...

Documents

Transcript of Software Defined Radio and Cognitive Radio · Software Defined Radio and Cognitive Radio:...