Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR...

20
RCJ SDR - 2/27/08 Introduction to Software Defined Radio Richard C. Jaeger Professor Emeritus ECE Department Auburn University

Transcript of Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR...

Page 1: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Introduction toSoftware Defined Radio

Richard C. JaegerProfessor EmeritusECE Department

Auburn University

Page 2: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Mitola’s Software Defined Radio

ADC

DAC

DigitalSignal

ProcessingIsolator

PA

Software Defined Radio (After Mitola - 1992 and Abidi)

Page 3: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

SDR Advantages• Minimum of “Expensive” Analog Hardware• Maximize Use of VLSI Technology

– High density, low cost, low power• Maximum Flexibility

– Implement new standards or additional modes– Incorporate new modulation/demodulation methods– “Easy” to correct design errors

• High Accuracy Quadrature Mixing

Page 4: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

SDR Practical LimitsDoes One Really Want Implement the Mitola Model?

• Direct Sampling Couples a Signal Out the Antenna (FCC not happy)• Why Sample the “World” Instead of the Desired Band

– Wastes power & dynamic range• High Speed ADCs Consume Large Amounts of Power• A More Realistic Situation is the Direct Conversion Receiver

From the 2007 ARRL Radio Handbook Chapter 16

Page 5: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

SNR and Dynamic Range Limits

MDS = 0.3µVRMS Maximum Interferer = 30mVRMSDynamic Range of 105 (100 dB) → 16 bits plus sign

Page 6: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Page 7: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Page 8: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

P =2SNRBits x f S

FOM P =

214 x 1010

1.2x1012 =137 W

Page 9: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Page 10: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Sample Rate Reduction

• Analog Front End– Bandpass Filtering– Low Noise Amplification– Down Conversion to Reduce Frequency

• Sub-Sampling– Bandpass Sampling, IF Sampling

Page 11: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Analog Front End - Down ConverterLNA, LO, Mixer, Filtering

• Low noise amplifier• Mixer• Local Oscillator• Bandpass Filtering (often several filters)

LNA

Mixer

Local Oscillator

BandpassFilter

fRF fIF = fRF - fLO

fLO

BandpassFilter

fRF

To ADC

Page 12: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Impulse Sampling

Impulse Sampling of Signal s(t) : s t( )↔ S f( )

y t( ) = s t( ) δ t − nTS( )−∞

∑ → Y f( ) =1TS

S f − nfS( )−∞

Sampling Replicates of the Spectrum

Spectral Copies are Created Above and Below the Original

The Sampling Process Can be Utilized to Directly Up- or Down - Convert a Signal

Page 13: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Sub-Sampling Theory(Bandpass, IF Sampling)

Sample Rate Bounds for FM Broadcast Band

m Lower Bound (MHz) Upper bound (MHz)

1 108 176

2 72.0 88.0

3 54.0 58.7

4 43.2 44.0

5 36.0 35.2

6 30.9 29.3

• Only need to sample at greater than twice the signal bandwidth• Sample rates carefully chosen to prevent spectral aliasing• Example - FM broadcast band 88 MHz - 108 MHz

– fC = 98 MHz, B = 20 MHz

B = BandwidthfC = Band Center frequency

2 fC + Bm +1

≤ fS ≤2 fC + Bm

Page 14: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Sub-SamplingFM Broadcast Band Example

FM Broadcast Band 88 - 108 MHz

0 20 40 60 80 100 120-20-40-60-80-100-120

FM Band Sub-Sampled at 44 MHz

0 20 40 60 80 100 120-20-40-60-80-100-120

FM Band Sub-Sampled at 64 MHz

(-108) - (-88) MHz

Down Conversion ExampleHigher Frequency Copies are Not Shown

64 MSPS: 91.1 MHz Signal is Aliased to 100.9 MHz

Page 15: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Analytic Signals

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5x 10-3

-1

-0.5

0

0.5

1

seconds

-5000 -4000 -3000 -2000 -1000 0 1000 2000 3000 4000 50000

0.05

0.1

0.15

0.2

0.25

0.3

0.35

cosωt =ejωt + e−jωt

2 contains two spectral components at ±ω

ejωt = cosωt+ j sinωt (an analytic function) contains one spectral component at +ω

cos 2000πt( )

cos 2000πt( ) + j sin 2000πt( )

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5x 10-3

-1

-0.5

0

0.5

1

seconds

-5000 -4000 -3000 -2000 -1000 0 1000 2000 3000 4000 50000

0.2

0.4

0.6

0.8

1

1.2

1.4

Page 16: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Half-Complex MixerQuadrature Mixing - Real Input Signals

y t( ) = x t( ) cosωot + j sinωot[ ] = I t( ) + jQ t( )From the 2007 ARRL Radio Handbook Chapter 16

Page 17: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Quadrature DemodulationAM & FM Signals

A t( ) = I 2 t( ) + Q2 t( ) φ t( ) = tan−1Q t( )I t( )

ω =

dφ t( )dt

From the 2007 ARRL Radio Handbook Chapter 16

Page 18: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Gnu RadioUniversal Software Radio Peripheral (USRP)

• Open Source Software– Linux, Windows, Mac

• USRP– 4 64 MSPS ADCs– 2 128 MSPS DACs– FPGA

• DDC• USB Interface (32 MB/Sec)

• USRP 2– 128 MSPS ADCs– 256 MSPS DACs– Gb Ethernet Port

Page 19: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

USRP Demonstration

• USRP and PC runningLinux

• AM BC Receiver• FM BC Receiver (WB)• HF AM Receiver• NBFM Receiver

Page 20: Introduction to Software Defined Radio - Auburn …agrawvd/COURSE/E7950_Spr08/SDR Talk.pdfIntroduction to Software Defined Radio ... RCJ SDR - 2/27/08 Mitola’s Software Defined Radio

RCJ SDR - 2/27/08

Some SDR Sources

• Gnu Radio gnuradio.org/trac• HPSDR hpsdr.org• Pentek pentek.com• TenTec tentec.com• Flex Radio flex-radio.com• TI, Analog Devices, …• Many More