Software Defined Radio: Radios of Tomorrow Behrouz Farhang ECE Department, Univ of Utah farhang@ece.utah.edu.

Software Defined Radio:

Radios of Tomorrow

Behrouz Farhang

ECE Department, Univ of Utah

farhang@ece.utah.edu

Organization

1. A Brief History of Radio Communication

2. Background: Fourier Series Expansion

3. Important Concepts: Filtering and Modulation

4. MATLAB Demos

5. From Analog to Digital

6. Cognitive Radios

7. Research Activities at Wireless Communications

History of the Radio: James C. Maxwell

1861-1865: James Clerk Maxwellexperiments with electromagnetic waves

1873: describes the propagation of electromagnetic waves in his paper “A Dynamical Theory of the Electromagnetic Field”

Wireless Radio Begins: Nikola Tesla

1893: Tesla gives public demonstration of wireless radio communication at St. Louis, Missouri.

He describes in detail the principles of radio communication.

Wireless Radio Begins: Guglielmo Marconi and K. F. Braun

1896: G. Marconi is awarded a patent for radio:Improvements in Transmitting Electrical Impulses and Signals and in Apparatus There-for

1897: Marconi establishes a radio station on the Isle of Wight (England)

1909: Marconi and Karl Ferdinand Braun win Nobel Prize in Physics for "contributions to the development of wireless telegraphy"

The History of the Radio

Better Sound Quality: FM Radio

1933: Edwin H. Armstrong patents FM radio (better quality than AM)

Television

1940s: analog television transmissions start in North America and Europe.

Television

1940s: analog black and white television transmissions start in North America and Europe.

1950s: analog color television transmissions start in North America and Europe.

Other stuff

Internet

Digital Audio Broadcasting (DAB)

Digital Video Broadcasting (DVB)HDTV

Satellite Radio

etc…

Going from analog “payload” to digital “payload”

However, radio itself still “analog” up to the decoder stage

The Fundamental Blocks in a Radio

Move signal from RF to baseband (original signal)HOW?

Remove frequency componentsHOW?

Increase amplitude of signalmultiply by constant

Fourier Series Expansion

The Theory Behind Radio Technology: Fourier Series

Jean Baptiste Joseph Fourier (1768-1830)French mathematician and physicist

Idea:“any” periodic function can be decomposed into an (infinite) sum of sines and cosines

How Radios Work?

€

f ( t) = F(ω1)cos(ω1t)+ F(ω2 )cos(ω2t)+ F(ω3 )cos(ω3t)+L

• According to Fourier any signal f(t) can be expanded as a sum of

sine-waves:

This leads to the notion of signal spectrum

€

F(ω1)

€

F(ω2)

€

F(ω3)

€

ω

€

L

Example: Square Pulse, infinite # harmonics!

With 250 harmonics:

Gibbs phenomenonat discontinuity

Example: Sawtooth Wave

Example: Triangle Wave

Some TD signals are suited better for transmission than others (less harmonics)

Rect-wave: poor TD quality with 25 harmonics

Sawtooth-wave: poor TD quality with 25 harmonics

Triangle-wave: pretty good TD quality with 10 harmonics

Modulation

Modulation is established by multiplying a signal with a carrier cos(ωct):

€

x( t) = f (t)cos(ωct) = F(ω1)cos(ω1t)cos(ωct)+ F(ω2 )cos(ω2t)cos(ωct)+L

=1

2F(ω1)[cos((ωc −ω1)t)+ cos((ωc +ω1)t)]

+1

2F(ω2 )[cos((ωc −ω2 )t)+ cos((ωc +ω2 )t)]+L

The spectrum is shifted from frequencies ω1, ω2, ω3, … to frequencies ωc+ω1, ωc-ω1, ωc+ω2, ωc-ω2, ωc+ω3, ωc-ω3, …

€

cosA cosB =1

2[cos(A− B)+ cos(A+ B)]

Demodulation

Now let us take the modulated signal f(t)cos(ωct) and

modulate it again with the same carrier.

€

y( t) = x(t)cos(ωct)

= f ( t)cos(ωct)cos(ωct) =1

2f ( t)[cos((ωc −ωc )t)+ cos((ωc +ωc )t)]

=1

2f ( t)+

1

2f (t)cos(2ωct)

By passing this signal through a lowpass filter (a system that cancels the second term on the right hand side) and amplifying the result by a gain of 2, we get f(t).

Next let us consider the modulated signal

€

x( t) = f1(t)cos(ωc1t)+ f2(t)cos(ωc2t)

And demodulated versions of it:

€

y1( t) = x(t)cos(ωc1t)

=1

2f1( t)+

1

2f1(t)cos(2ωc1t)

+1

2f2( t)cos((ωc1 −ωc2 )t)+

1

2f2( t)cos((ωc1 +ωc2 )t)

and

€

y2( t) = x(t)cos(ωc2t)

=1

2f2( t)+

1

2f2(t)cos(2ωc2t)

+1

2f1( t)cos((ωc2 −ωc1)t)+

1

2f1( t)cos((ωc2 +ωc1)t)

Demodulation

What is Filtering?

Input Signal

Filter Response

Output Signal

A Radio?

A Conventional Radio: all in Analog

New Paradigm: Software Defined Radio (SDR)

RX Signals are digitized immediately after the LNAand then processed entirely in software, flexible

TX vice versa

Software Defined Radio:

A Radio!

Cognitive Radios

• Have the capability to be aware of their surrounding environment

• Can change PHY depending on environment

• Can change PHY depending on traffic needs

• Can alter higher layer behavior as needed

• Learn from past experiences

Capable of complex adaptation

Cognitive Radios: Primary and secondary users

Primary (licensed) and secondary (unlicensed) users coexist and share the same spectrum

PUs have priority and thus SUs must back-off as soon as PUs begin a communication

This requires channel sensing

PU

PUPU

PU

SU

SUSU

SU

SDR Activities at the Wireless Communications Lab

1. A course title “Software Radio” is offered (graduate level class)

2. We are actively involved in the SDR Forum, a nonprofit international industry association that supports the development and deployment of SDR technologies

3. A team of our students is finalist in the Smart Radio Challenge; a worldwide competition that challenges student teams to design, develop and test an SDR / a cognitive radio (CR) system.

4. We also develop MIMO Communication systems; a technology that promises data rates in order of Giga bits persecond.