lecture4 ee689 channel pulse model

8/10/2019 Lecture4 Ee689 Channel Pulse Model

1/27

Sam Palermo Analog & Mixed-Signal Center

Texas A&M University

ECEN689: Special Topics in High-SpeedLinks Circuits and Systems

Spring 2012

Lecture 4: Channel Pulse Model & Modulation Schemes


2/27

Announcements & Agenda

No class next Monday ISI Channel pulse model

Peak distortion analysis Compare NRZ, PAM-4, and Duobinary modulation Reference material for this lecture

Peak distortion analysis paper by Casper (posted onweb)

Notes from H. Song, Arizona State Papers posted on PAM-4 and duobinary modulation

2


3/27

Inter-Symbol Interference (ISI)

Previous bits residual state can distort the current bit,resulting in inter-symbol interference (ISI) ISI is caused by

Reflections, Channel resonances, Channel loss (dispersion)

Pulse Response

3

( )

( )t c1

( )t h ( )

( )t c1

( )( )t y 1

( )( ) ( )( ) ( )t ht ct y = 11


4/27

NRZ Data Modeling

An NRZ data stream can be modeled as asuperposition of isolated 1s and 0s

4

Data = 1000101

1 Symbol

0 Symbol

( )( ) ( ) ( )( )T k t ukT t ut ck 11 +

( )( ) ( )( )t ct c k k 10 =

( )

=

=


12/27

Peak Distortion Analysis Example 1

12

( )( )

( ) ( )

( )( )( )

( ) ( ) 288.0389.0007.0540.02

389.0

007.0

540.0

0

0

1

0

0

1

)1(0

==

=

=

=

= >

=


13/27

Worst-Case Bit Pattern

Pulse response can be used to find the worst-case bit pattern

13

[ ]... ...a Pulse 654321012 aaaaaaaaa =

( )[ ]... )()(1)()()()()(- ... 21123456 asignasignasignasignasignasignasignasign

Flip pulse matrix about cursor a 0 and the bits are the inverted sign ofthe pulse ISI

Worst-Case Bit Pattern Eye 10kbits Eye


14/27

Peak Distortion Analysis Example 2

14

( )( )( )

( )

( )( )( )

( ) ( ) 338.0542.0053.0426.02

542.0

053.0

426.0

0

0

1

0

0

1

)1(0

==

=

=

=

= >

=


15/27

Modulation Schemes

15

Binary, NRZ, PAM-2 Simplest, most common modulation format

PAM-4 Transmit 2 bits/symbol Less channel equalization and circuits run speed

Duobinary Allows for controlled ISI, symbol at RX is current bit plus preceding bit Results in less channel equalization

[ ] [ ] [ ]1+= n xn xnw

0

1

00

01

11

10

0, if x[n-1]=0

1, if x[n-1]=0 OR0, if x[n-1]=1

1, if x[n-1]=1

No Pre-Coding Case


16/27

Modulation Frequency Spectrum

16

Majority of signal power

in 1GHz bandwidth


in 0.5GHz bandwidth


in 0.5GHz bandwidth


17/27

Nyquist Frequency

Nyquist bandwidth constraint: The theoretical minimum required system bandwidth to detect R S

(symbols/s) without ISI is R S /2 (Hz) Thus, a system with bandwidth W=1/2T=R S /2 (Hz) can support a

maximum transmission rate of 2W=1/T=R S (symbols/s) without ISI

17

/Hz)(symbols/s 222

1 =W RW R

T S S

For ideal Nyquist pulses (sinc), the required bandwidth is onlyR S /2 to support an R S symbol rate

Modulation Bits/Symbol Nyquist Frequency

NRZ 1 R s /2=1/2T bPAM-4 2 R s /2=1/4T b

Duobinary is not Nyquist signaling, as it employscontrolled ISI for reduced signal bandwidth


18/27

NRZ vs PAM-4

PAM-4 should be considered when Slope of channel insertion loss (S 21) exceeds reduction in PAM-4

eye height Insertion loss over an octave is greater than 20*log10(1/3)=-9.54dB

On-chip clock speed limitations 18


19/27

PAM-4 Receiver

3x the comparators of NRZ RX

19

[Stojanovic JSSC 2005 ]


20/27

NRZ vs PAM-4 Desktop Channel

Eyes are produced with 4-tapTX FIR equalization Loss in the octave between 2.5

and 5GHz is only 2.7dB NRZ has better voltage margin

20

Loss at 2.5GHz = -4.8dB

Loss at 5GHz = -7.5dB


21/27

NRZ vs PAM-4 T20 Server Channel

Eyes are produced with 4-tapTX FIR equalization Loss in the octave between 2.5

and 5GHz is 15.8dB PAM-4 might be a better choice

21

Loss at 2.5GHz = -11.1dB

Loss at 5GHz = -26.9dB


22/27

Multi-Level PAM Challenges

Receiver complexity increases considerably 3x input comparators (2-bit ADC) Input signal is no longer self-referenced at 0V differential

Need to generate reference threshold levels, which will be dependenton channel loss and TX equalization

CDR can display extra jitter due to multiple zerocrossing times

Smaller eyes are more sensitive to cross-talk due to

maximum transitions Advanced equalization (DFE) can allow NRZ signaling to

have comparable (or better) performance even with>9.5dB loss per octave

22


23/27

Duobinary Signaling

23

[NEC ISSCC 2005 & 2009 ]

Binary(1, -1)

Duobinary(2, 0, -2)Channel

TX

EQ

RX

EQ

[ ] [ ] [ ]1+= n xn xnw

[ ]nw[ ]n x


24/27

Duobinary Signaling w/ Precoder

24

[ ]n x

[ ]nw1 [ ]nw2

[ ]n y

[Lee JSSC 2008 ]

[NEC ISSCC 2005 ]

With precoder, middle signal at the receiver maps to a 1 and high and low signal maps to a 0

Precoder allows for binary signal out of transmitter resulting in a

power gain Channel can be leveraged to aid in duobinary pulse shaping

Eliminates error propagation at receiver

Similar performance to using a 1-tap loop-unrolled DFE at RX


25/27


26/27

Modulation Take-Away Points

Loss-slope guidelines are a good place to start inconsideration of alternate modulation schemes

More advanced modulation trades-off receiver complexityversus equalization complexity

Advanced modulation challenges Peak TX power limitations Setting RX comparator thresholds and controlling offsets CDR complexity Crosstalk sensitivity (PAM-4)

Need link analysis tools that consider voltage, timing, andcrosstalk noise to choose best modulation scheme for agiven channel

26


27/27

Next Time

Link Circuits Termination structures Drivers

Receivers

27

lecture4 ee689 channel pulse model

Documents