lecture 12

27
EE 230: Optical Fiber Communication Lecture 12 From the movie Warriors of the Net Receivers

Upload: mumtaz-ali

Post on 26-Jan-2016

213 views

Category:

Documents


0 download

DESCRIPTION

For 34 optical communication.

TRANSCRIPT

Page 1: Lecture 12

EE 230: Optical Fiber Communication Lecture 12

From the movieWarriors of the Net

Receivers

Page 2: Lecture 12

Receiver Functional Block Diagram

Fiber-Optic Communications Technology-Mynbaev & Scheiner

Page 3: Lecture 12

Receiver Types+Bias

Is

RL 50 Amplifier

Output

+Bias

Is

Amplifier

Output

Ct

Rf+Bias

Is

RL

Amplifier

Output

EqualizerCt

Low Impedance

Low SensitivityEasily MadeWide Band

High Impedance

Requires Equalizer for high BWHigh SensitivityLow Dynamic RangeCareful Equalizer Placement Required

Transimpedance

High Dynamic RangeHigh SensitivityStability ProblemsDifficult to equalize

Page 4: Lecture 12

Equivalent Circuits of an Optical Receiver

High Impedance Design Transimpedance Design

Transimpedance with Automatic Gain Control

Fiber-Optic Communications Technology-Mynbaev & Scheiner

Page 5: Lecture 12

Receiver Noise Sources

•Photon NoiseAlso called shot noise or Quantum noise, described by poisson statistics

•Photoelectron NoiseRandomness of photodetection process leads to noise

•Gain Noiseeg. gain process in APDs or EDFAs is noisy

•Receiver Circuit noiseResistors and transistors in the the electrical amplifier contribute to circuit noise

Photodetector without gain Photodetector with gain (APD)

Page 6: Lecture 12

Noise

2

2Noise Power=4

4 4

nn

rms rms

VkTB i R

R

kTBi V kTRBR

2

m

spectral density= V /Hz

for FETs4kTK=

gwhere is the FET corner frequency and is the channel noise factor

c

c

Kf

f

f

Frequency

Nois

e P

ow

er

Frequency

Nois

e P

ow

er

Frequency

Nois

e P

ow

er 1/f noise

Fc

Johnson noise (Gaussian and white)

1/2 1/22rms noise current 2ni qIB

Shot noise (Gaussian and white)

“1/f” noise

Page 7: Lecture 12

Johnson (thermal) Noise

Noise in a resistor can be modeled as due to a noiseless resistor in parallel with a noise current source

2 2

The variance of the noise current source is given by:

4

Where is Boltzman's constant

T is the Temperature in Kelvins

B is the bandwidth in Hz (not bits/sec)

Bi

B

k TBi

R

k

s = »

Page 8: Lecture 12

Photodetection noise

The electric current in a photodetector circuit is composed of a superposition of the electrical pulses associated with each photoelectron

The variation of this current is called shot noise

If the photoelectrons are multiplied by a gain mechanism then variations in the gain mechanism give rise to an additional variation in the current pulses. This variation provides an additional source of noise, gain noise

Noise in photodetector

Noise in APD

Page 9: Lecture 12

Circuit Noise

Page 10: Lecture 12

Signal to Noise RatioSignal to noise Ratio (SNR) as a function of the average number of photo electrons per receiver resolution time for a photo diode receiver at two different values of the circuit noise

Signal to noise Ratio (SNR) as a function of the average number of photoelectrons per receiver resolution time for a photo diode receiver and an APD receiver with mean gain G=100 and an excess noise factor F=2

At low photon fluxes the APD receiver has a better SNR. At high fluxes the photodiode receiver has lower noise

Page 11: Lecture 12

Dependence of SNR on APD Gain

Curves are parameterized by k, the ionization ratio between holes and electrons

Plotted for an average detected photon flux of 1000and constant circuit noise

Page 12: Lecture 12

Receiver SNR vs Bandwidth

Double logarithmic plot showing the receiver bandwidth dependence of the SNR for a number of different amplifier types

Page 13: Lecture 12

Basic Feedback Configuration

+

-RiIs

RoA Vi

Vo

IiIs

If

Parallel Voltage Sense:Voltage Measured and heldConstant => Low Output Impedance

Parallel Current FeedbackLowers Input Impedance

1

s f i

is i

i

i iin

s m

i i i

Vi AV

R

V RZ

i R

1 1

o i i

i s f s o

o i s o

o i mt

s i m

V Ai R

i i i i V

V AR i V

V AR RZ

i AR R

Stabilizes Transimpedance Gain

1 1test o o

test i m

V R RZo

I AR R

+

-Zi

Zo

ZtIi

Ii

Page 14: Lecture 12

Transimpedance Amplifier Design

+

-Z i

i

ZeroInput Impedance

Output Voltage Proportional to Input current

+

-Ri

Vi RoA Vi

Typical amplifier modelWith generalized input impedanceAnd Thevenin equivalent output

o i i i

i ms

V AV AR i

VAR R

i

+

-RiVi

RoA Vi

is+

-

Vo

Calculation ofOpenloop transimpedance gain: Rm

Page 15: Lecture 12

Transimpedance Amplifier Design Example

Rc

Rf

Q1

Q2

Vcc1

Vbias

Vcc2

Photodiode

Out

Transimpedance approximately equals Rflow values increase peaking and bandwidth

Controls open loop gain of amplifier, Reduce to decrease “peaking”

Most Common TopologyHas good bandwidth and dynamic Range

See Das et. al. Journal of Lightwave TechnologyVol. 13, No. 9, Sept.. 1995

For an analytic treatment of the design of maximally flathigh sensitivity transimpedance amplifiers

Page 16: Lecture 12

“Off-the-shelf” Receiver Example

2 17 222 1.8 10dDetector

i qI I B x A

2 2 12 22Re

41.9 10Detectorsistor

s

kTi I B i x A

R

12 2 12 210

2Re 1

410 7.5 10

NF

Detectorsistor Amps

kTi I B i x A

R

2 2 12 210

2Re 1 2

410 7.6 10

TotalNF

Detectorsistor Amp Amps

kTi I B i x A

R

45.22

20.14

16.63

16.59

Sensitivity

dBm

dBm

dBm

dBm

.

+Bias

Is

Amplifier 1Gain1=20dBNF1=7dB

Output

Amplifier 2Gain2=20dBNF2=7dB

50

C=400ffId=10nA=0.7

NFTotal NF1 NF2 1

Gain1}NF 10Log10

4kTRs Vn2

4kTRs

Page 17: Lecture 12

Bit Error Rate

BER is equal to number of errors divided by total number of pulses (ones and zeros). Total number of pulses is bit rate B times time interval. BER is thus not really a rate, but a unitless probability.

Page 18: Lecture 12

Q Factor and BER

on

thon

off

offth VVVVQ

21

2

1 QerfBER

Page 19: Lecture 12

BER vs. Q, continued

When off = on and Voff=0 so that Vth=V/2, then Q=V/2. In this case,

221

2

1

V

erfBER

Page 20: Lecture 12

Sensitivity

The minimum optical power that still gives a bit error rate of 10-9 or below

Page 21: Lecture 12

Receiver Sensitivity

1/22

2 22

Sensitivity= Average detected optical power for a given bit error rate

For pin detectors

2 damplifier

hvP Q iq

i i qI I B

(Sm

ith a

nd P

erso

nick

198

2)

2 /2

-9

Probability of error vs. Q is to good approximation:

1 E 2

eg. for a SNR = Q = 6 Bit Error Rate= P(E)=10

QePQ

Page 22: Lecture 12

Dynamic Range and Sensitivity Measurement

Dynamic range is the Optical power difference in dB over which the BER remains within specified limits (Typically 10-9/sec)

The low power limit is determined by the preamplifier sensitivity

The high power limit is determined by the non-linearity and gain compression

PattenGenerator

Transmitter Adjustable Attenuator

Optical Receiver

Bit ErrorRate Counter

Optional Clock

Input Optical Power

Feedback ResistanceHigh Rf(High Impedance Preamplifier)

Low Rf(Transimpedance Preamplifier

Dynamic Range

Maximum Signal Level

receiver Sensitivity

Experimental Setup

Page 23: Lecture 12

Eye Diagrams

Formation of eye diagram

Eye diagramdegradations

Transmitter“eye” mask

determination

Computer Simulation of a distorted eye diagramFiber-Optic Communications Technology-Mynbaev & Scheiner

Page 24: Lecture 12

Power Penalties

• Extinction ratio

• Intensity noise

• Timing jitter

Page 25: Lecture 12

Extinction ratio penalty

Extinction ratio rex=P0/P1

offonex

ex RP

r

rQ

2

1

1

ex

exex r

r

1

1log10

Page 26: Lecture 12

Intensity noise penalty

rI=inverse of SNR of transmitted light

221log10 QrII

II RPr

Page 27: Lecture 12

Timing jitter penalty

Parameter B=fraction of bit period over which apparent clock time varies

22

83

4 Bb

2/2/1

2/1log10

222 Qbb

bJ