Transcript Modem Design, Implementation, and Testing Using NI’s LabVIEW Prof. Brian L. Evans

Modem Design, Implementation,
and Testing Using NI’s LabVIEW
http://www.wncg.org
http://www.ece.utexas.edu
Prof. Brian L. Evans
Embedded Signal Processing Laboratory
The University of Texas at Austin
[email protected]
Contributions by Vishal Monga, Zukang Shen, Ahmet Toker, and Ian Wong, UT Austin
Outline

Real-Time DSP Course

Single Carrier Transceiver
 Sinusoidal Generation
 Digital Filters
 Data Scramblers
 Pulse Amplitude Modulation
 Quadrature Amplitude Modulation

Multicarrier Transceiver

Conclusion
2
Real-Time DSP Course: Overview

Objectives of junior-level class
Over 500
 Build intuition for signal processing concepts
served
 Translate signal processing concepts into
since 1997
real-time digital communications software

Lecture: breadth (three hours/week)
 Digital signal processing algorithms
 Digital communication systems
 Digital signal processor architectures

Laboratory: depth (three hours/week)
 Deliver voiceband modem
 “Design is the science of tradeoffs” (Yale Patt)
 Test/validate implementation
http://www.ece.utexas.edu/~bevans/courses/realtime/
3
Real-Time DSP Course: Which DSP?





Students are third-year undergraduates
Fixed-point DSPs for high-volume products
 Battery-powered: cell phones, digital still cameras …
 Wall-powered: ADSL modems, cellular basestations …
Fixed-point issues
 Using non-standard C extensions for fractional data
 Converting floating-point programs to fixed-point
 Manual tracking of binary point prone to error
Floating-point DSPs
 Feasibility for fixed-point DSP realization
 Shorter prototyping time
Program TI TMS320C67x DSP in C
 Code Composer Studio 2.2
4
Real-Time DSP Course: Textbooks

C. R. Johnson, Jr., and W. A.
Sethares, Telecommunication
Breakdown, Prentice Hall, 2004.
 Intro to digital communications
and transceiver design
 Matlab examples

Rick Johnson Bill Sethares
(Cornell)
(Wisconsin)
S. A. Tretter, Comm. System Design using
DSP Algorithms with Lab Experiments for
the TMS320C6701 & TMS320C6711, 2003.
 Assumes DSP theory and algorithms
 Assumes access to C6000 reference manuals Steven Tretter
(Maryland)
 Errata/code: http://www.ece.umd.edu/~tretter
5
Lab 1. QAM Transmitter Diagram
Lab 4
Rate
Control
LabVIEW demo by Zukang Shen (UT Austin)
Lab 6
QAM
Encoder
Lab 2
Passband
Signal
Lab 3
Tx Filters
6
Lab 1. QAM Transmitter Diagram
LabVIEW
Control
Panel
QAM
Passband
Signal
Eye
Diagram
LabVIEW demo by Zukang Shen (UT Austin)
7
Lab 1. QAM Transmitter Diagram
square root raise cosine, roll-off = 0.75, SNR = 
passband signal for 1200 bps mode
raise cosine, roll-off = 1, SNR = 30 dB
passband signal for 2400 bps mode
8
Lab 2. Sine Wave Generation



Aim: Evaluate three ways
to generate sine waves
 Function call
 Lookup table
 Difference equation
Three output methods
 Polling data transmit register
 Software interrupts
 Direct memory access (DMA) transfers
Expected outcomes are to understand
 Signal quality vs. implementation complexity tradeoff
 C6701 EVM board’s stereo codec operation
 Interrupt mechanisms and DMA transfers
9
Lab 2. Sine Wave Generation

Evaluation procedure
 Validate sine wave frequency on scope, and test for
various sampling rates (14 sampling rates on board)
 Method 1 with interrupt priorities
 Method 1 with different DMA initialization(s)
Spring 2004
Fall 2003
HP 60 MHz
C6701
Digital Storage
Oscilloscope
LabVIEW DSP
Test Integration Code Composer
Toolkit 2.0
Studio 2.2
10
Lab 3. Digital Filters


Aim: Evaluate four ways to implement
discrete-time linear time-invariant filters
 FIR filter: convolution in C and assembly
 IIR Filter: direct form and cascade of biquads, both in C
IIR filter design gotchas: oscillation & instability
 In classical designs, poles sensitive to perturbation
 Quality factor measures sensitivity of pole pair:
classical

Q
Elliptic analog lowpass IIR filter dp = 0.21 at wp =
20 rad/s and ds = 0.31 at ws = 30 rad/s [Evans 1999]
poles
zeros
Q
poles
zeros
1.7
-5.3533±j16.9547
0.0±j20.2479
0.68
-11.4343±j10.5092
-3.4232±j28.6856
61.0
-0.1636±j19.9899
0.0±j28.0184
10.00
-1.0926±j21.8241
-1.2725±j35.5476
optimized
Q  [ ½ ,  ) where Q = ½ dampens and Q =  oscillates
11
Lab 3. Digital Filters

IIR filter design for implementation
 Butterworth/Chebyshev filters special
cases of elliptic filters
 Minimum order not always most efficient


Filter design gotcha: polynomial inflation
 Polynomial deflation (rooting) reliable in floating-point
 Polynomial inflation (expansion) may degrade roots
 Keep native form computed by filter design algorithm
Expected outcomes are to understand
 Speedups from convolution assembly routine vs. C
 Quantization effects on filter stability (IIR)
 FIR vs. IIR: how to decide which one to use
12
Lab 3. Digital Filters


Test Equipment
 Agilent Function Generator
 HP 60 MHz Digital Storage Oscilloscope
 Spectrum Analyzer
Evaluation Procedure
 Sweep filters with sinusoids to construct magnitude and
phase responses
• Manually using test equipment, or
• Automatically by LabVIEW DSP Test Integration Toolkit
 Check filter output for cut-off frequency, roll-off factor…
 FIR: Compare execution times (in Code Composer) of
• C without compiler optimizations
• C with compiler optimizations
• C callable assembly language routine
 IIR: Compute execution times (in Code Composer)
13
Lab 4. Data Scramblers


Aim: Generate pseudo-random bit sequences
 Build data scrambler for given connection polynomial
 Descramble data via descrambler
 Obtain statistics of scrambled binary sequence
Expected outcomes are to understand
 Principles of pseudo-noise (PN) sequence generation
 Identify applications in communication systems
14
Lab 4. Data Scramblers

Evaluation procedure
 Check scrambler output for various deterministic
sequences as input(s)
 Descrambler must recover input sequence from
scrambled one
 Test for sequence period, autocorrelation and other
significant statistical properties

Using DSP Test & Measurement Toolkit instead
 Compute autocorrelation of PN sequence
 Compare this autocorrelation with that of white noise
generated by LabVIEW to measue PN sequence quality
15
Lab 5. Digital PAM Transceiver

an
Aim: Develop PAM transceiver blocks in C
 Amplitude mapping to PAM levels
 Interpolation filter bank for pulse shaping filter
 Clock recovery via phase locked loops
L
gT[n]
D/A
3d
d
-d
-3 d
Transmit
Filter
4-PAM
L samples per symbol
gT,0[n]
an
gT,1[n]
D/A
Transmit
Filter
Filter bank implementation
16
Lab 5. Digital PAM Transceiver



Expected Outcomes are to understand
 Basics of PAM modulation
 Zero inter-symbol interference condition
 Clock synchronization issues
Test Equipment: Same as Lab 3
Evaluation Procedure
 Generate eye diagram to visualize PAM signal quality
 Observe spectrum of modulated signal
 Prepare DSP modules to test symbol clock frequency
recovery subsystem
17
Lab 6. Digital QAM Transmitter

Aim: Develop QAM transmitter blocks in C
 Differential encoding of digital data
 Constellation mapping to QAM levels
 Interpolation filter bank for pulse shaping filter
an
1
Serial/
parallel
converter
Bit
stream
J
L
gT[m]
Map to 2-D
constellation
bn
cos(w0 n)
L
+
D/A
gT[m]
L samples per symbol
sin(w0 n)
18
Lab 6. Digital QAM Transmitter



Expected outcomes are to understand
 In-phase and quadrature modulation principles
 Bandwidth efficiency issues
Test equipment: same as Lab 5
Evaluation procedure
 Verify differential encoding and QAM mapping
 Generate eye diagram to visualize QAM signal quality
 Observe spectrum of modulated signal
19
Lab 7. Digital QAM Receiver – Part 1




Aim: Develop QAM receiver blocks in C
 Carrier recovery
 Coherent demodulation
 Decoding of QAM levels to digital data
Expected outcomes are to understand
 Carrier detection and phase adjustment
 Design of receive filter
 Probability of error analysis to evaluate decoder
Test equipment: Same as Lab 6
Evaluation procedure
 Recover and display carrier on scope
 Regenerate eye diagram and QAM constellation
 Observe signal spectra at each decoding stage
20
Developed Voiceband Transceiver:
Now What? Got Anything Faster?

Multicarrier modulation divides broadband
channel into narrowband subchannels
 No inter-symbol interference if constant subchannel
gain and ideal sampling
 Based on fast Fourier transform (FFT)

ADSL/VDSL and IEEE 802.11a/g & 802.16a
magnitude
channel
carrier
Each ADSL/VDSL subchannel is 4.3 kHz wide (about a
voice channel) and carries QAM encoded subsymbol
subchannel
frequency
21
Multicarrier Modulation by IFFT
e j 2 f1t
X1
g(t)
g(t)
e
X1
e
+
Discrete
time
X2
e
g(t) : pulse shaping filter
X N /2
j 2
2
n
N
x
j 2 f N / 2 t
x
1
n
N
x
I
j 2 f 2 t
x
g(t)
e
X N /2
Xi
x
e
X2
Q
j 2
j 2
+
N /2
n
N
x
Xi : ith symbol from encoder
22
Multicarrier Modulation (ADSL)
Q
Xi
00101 QAM
Mapping
I
X0
N/2
subchannels
(carriers)
X1
X2
XN/2
Mirror complex
data (in blue) and
take conjugates:
e j   e j   2 cos()
XN/2-1*
X2*
N-point
Inverse
Fast
Fourier
Transform
(IFFT)
x0
x1
x2
xN-1
N realvalued
time
samples
forms
ADSL
symbol
X1*
23
Multicarrier Modulation (ADSL)
Inverse FFT
ADSL
CP
N
v samples
CP
downstream upstream
4
32
64
512
N samples
s y m b o l (i)
copy
CP
s y m b o l ( i+1)
copy
CP: Cyclic Prefix
ADSL frame is an ADSL
symbol plus cyclic prefix
D/A + transmit filter
24
Multicarrier Demodulation (ADSL)
S/P
~
X0
N/2
subchannels
(carriers)
~
X N 21
~
XN 2
~*
X N 21
~
X 1*
N-point
Fast
Fourier
Transform
(FFT)
~
x0
~
x1
~
x
2
N time
samples
~
xN 1
25
ADSL Transceiver: Data Xmission
2.208 MHz
N/2 subchannels N real samples
Bits
00110
S/P
quadrature
amplitude
modulation
(QAM)
encoder
TRANSMITTER
N/2 subchannels
QAM
decoder
add
cyclic
prefix
P/S
each block programmed in lab and
covered in one full lecture
each block covered in one full lecture
RECEIVER
P/S
mirror
data
and
N-IFFT
invert
channel
=
frequency
domain
equalizer
P/S parallel-to-serial
D/A +
transmit
filter
channel
N real samples
N-FFT
and
remove
mirrored
data
S/P serial-to-parallel
remove
S/P cyclic
prefix
time
domain
equalizer
(FIR
filter)
receive
filter
+
A/D
FFT fast Fourier transform
26
Telecom and University Tracks
Modem Design, Implementation, and
Testing Using NI’s LabVIEW
Dr. Brian L. Evans
Associate Professor
The University of Texas at Austin
[email protected]
27