Point-to-point digital link design



Σχετικά έγγραφα
Main source: "Discrete-time systems and computer control" by Α. ΣΚΟΔΡΑΣ ΨΗΦΙΑΚΟΣ ΕΛΕΓΧΟΣ ΔΙΑΛΕΞΗ 4 ΔΙΑΦΑΝΕΙΑ 1

2 Composition. Invertible Mappings

HOMEWORK 4 = G. In order to plot the stress versus the stretch we define a normalized stretch:

EE512: Error Control Coding

Approximation of distance between locations on earth given by latitude and longitude

Monolithic Crystal Filters (M.C.F.)

Calculating the propagation delay of coaxial cable

Instruction Execution Times

Phys460.nb Solution for the t-dependent Schrodinger s equation How did we find the solution? (not required)

the total number of electrons passing through the lamp.

6.003: Signals and Systems. Modulation

Block Ciphers Modes. Ramki Thurimella

derivation of the Laplacian from rectangular to spherical coordinates

Potential Dividers. 46 minutes. 46 marks. Page 1 of 11

The Simply Typed Lambda Calculus

Homework 3 Solutions

ΚΥΠΡΙΑΚΗ ΕΤΑΙΡΕΙΑ ΠΛΗΡΟΦΟΡΙΚΗΣ CYPRUS COMPUTER SOCIETY ΠΑΓΚΥΠΡΙΟΣ ΜΑΘΗΤΙΚΟΣ ΔΙΑΓΩΝΙΣΜΟΣ ΠΛΗΡΟΦΟΡΙΚΗΣ 19/5/2007

Strain gauge and rosettes

CMOS Technology for Computer Architects

Capacitors - Capacitance, Charge and Potential Difference

[1] P Q. Fig. 3.1

(1) Describe the process by which mercury atoms become excited in a fluorescent tube (3)

Modbus basic setup notes for IO-Link AL1xxx Master Block

High Performance Voltage Controlled Amplifiers Typical and Guaranteed Specifications 50 Ω System

Assalamu `alaikum wr. wb.

Example Sheet 3 Solutions

Other Test Constructions: Likelihood Ratio & Bayes Tests

First Sensor Quad APD Data Sheet Part Description QA TO Order #

Statistical Inference I Locally most powerful tests

Second Order RLC Filters

Υλοποίηση Δικτυακών Υποδομών και Υπηρεσιών: OSPF Cost

k A = [k, k]( )[a 1, a 2 ] = [ka 1,ka 2 ] 4For the division of two intervals of confidence in R +

Math 6 SL Probability Distributions Practice Test Mark Scheme

Section 8.3 Trigonometric Equations

Aluminum Electrolytic Capacitors (Large Can Type)

Concrete Mathematics Exercises from 30 September 2016

The challenges of non-stable predicates

ΚΥΠΡΙΑΚΗ ΕΤΑΙΡΕΙΑ ΠΛΗΡΟΦΟΡΙΚΗΣ CYPRUS COMPUTER SOCIETY ΠΑΓΚΥΠΡΙΟΣ ΜΑΘΗΤΙΚΟΣ ΔΙΑΓΩΝΙΣΜΟΣ ΠΛΗΡΟΦΟΡΙΚΗΣ 6/5/2006

Physical DB Design. B-Trees Index files can become quite large for large main files Indices on index files are possible.

UNIVERSITY OF CALIFORNIA. EECS 150 Fall ) You are implementing an 4:1 Multiplexer that has the following specifications:

Απόκριση σε Μοναδιαία Ωστική Δύναμη (Unit Impulse) Απόκριση σε Δυνάμεις Αυθαίρετα Μεταβαλλόμενες με το Χρόνο. Απόστολος Σ.

6.1. Dirac Equation. Hamiltonian. Dirac Eq.

Electrical Specifications at T AMB =25 C DC VOLTS (V) MAXIMUM POWER (dbm) DYNAMIC RANGE IP3 (dbm) (db) Output (1 db Comp.) at 2 f U. Typ.

Problem Set 3: Solutions

Smaller. 6.3 to 100 After 1 minute's application of rated voltage at 20 C, leakage current is. not more than 0.03CV or 4 (µa), whichever is greater.

Démographie spatiale/spatial Demography

CHAPTER 25 SOLVING EQUATIONS BY ITERATIVE METHODS

Tunable Diode Lasers. Turning Laser Diodes into Diode Lasers. Mode selection. Laser diodes

Multilayer Ceramic Chip Capacitors

5.4 The Poisson Distribution.

TaxiCounter Android App. Περδίκης Ανδρέας ME10069

Multilayer Ceramic Chip Capacitors

C.S. 430 Assignment 6, Sample Solutions

UDZ Swirl diffuser. Product facts. Quick-selection. Swirl diffuser UDZ. Product code example:

Graded Refractive-Index

ST5224: Advanced Statistical Theory II

Elements of Information Theory

ANSWERSHEET (TOPIC = DIFFERENTIAL CALCULUS) COLLECTION #2. h 0 h h 0 h h 0 ( ) g k = g 0 + g 1 + g g 2009 =?

Fractional Colorings and Zykov Products of graphs

Aluminum Electrolytic Capacitors

Inverse trigonometric functions & General Solution of Trigonometric Equations

TMA4115 Matematikk 3

Econ 2110: Fall 2008 Suggested Solutions to Problem Set 8 questions or comments to Dan Fetter 1

1) Formulation of the Problem as a Linear Programming Model

ΕΙΣΑΓΩΓΗ ΣΤΗ ΣΤΑΤΙΣΤΙΚΗ ΑΝΑΛΥΣΗ

1000 VDC 1250 VDC 125 VAC 250 VAC J K 125 VAC, 250 VAC

Second Order Partial Differential Equations

Bounding Nonsplitting Enumeration Degrees

Type 947D Polypropylene, High Energy Density, DC Link Capacitors

ω ω ω ω ω ω+2 ω ω+2 + ω ω ω ω+2 + ω ω+1 ω ω+2 2 ω ω ω ω ω ω ω ω+1 ω ω2 ω ω2 + ω ω ω2 + ω ω ω ω2 + ω ω+1 ω ω2 + ω ω+1 + ω ω ω ω2 + ω

DESIGN OF MACHINERY SOLUTION MANUAL h in h 4 0.

Ordinal Arithmetic: Addition, Multiplication, Exponentiation and Limit

Δίκτυα Καθοριζόμενα από Λογισμικό

CHAPTER 48 APPLICATIONS OF MATRICES AND DETERMINANTS

Areas and Lengths in Polar Coordinates

Areas and Lengths in Polar Coordinates

Every set of first-order formulas is equivalent to an independent set

3.4 SUM AND DIFFERENCE FORMULAS. NOTE: cos(α+β) cos α + cos β cos(α-β) cos α -cos β

ΙΑΤΜΗΜΑΤΙΚΟ ΠΡΟΓΡΑΜΜΑ ΜΕΤΑΠΤΥΧΙΑΚΩΝ ΣΠΟΥ ΩΝ ΣΤΑ ΠΛΗΡΟΦΟΡΙΑΚΑ ΣΥΣΤΗΜΑΤΑ "VIDEO ΚΑΤΟΠΙΝ ΖΗΤΗΣΗΣ" ΑΝΝΑ ΜΟΣΧΑ Μ 11 / 99

4.6 Autoregressive Moving Average Model ARMA(1,1)

Code Breaker. TEACHER s NOTES

ΤΕΧΝΟΛΟΓΙΚΟ ΠΑΝΕΠΙΣΤΗΜΙΟ ΚΥΠΡΟΥ ΤΜΗΜΑ ΝΟΣΗΛΕΥΤΙΚΗΣ

Matrices and Determinants

( ) 2 and compare to M.

Precision Metal Film Fixed Resistor Axial Leaded

Section 1: Listening and responding. Presenter: Niki Farfara MGTAV VCE Seminar 7 August 2016

PARTIAL NOTES for 6.1 Trigonometric Identities

department listing department name αχχουντσ ϕανε βαλικτ δδσϕηασδδη σδηφγ ασκϕηλκ τεχηνιχαλ αλαν ϕουν διξ τεχηνιχαλ ϕοην µαριανι

Πανεπιστήμιο Πατρών Πολυτεχνική Σχολή Τμήμα Μηχανικών Ηλεκτρονικών Υπολογιστών και Πληροφορικής

Δίκτυα Επικοινωνιών ΙΙ: OSPF Configuration

ΚΥΠΡΙΑΚΗ ΕΤΑΙΡΕΙΑ ΠΛΗΡΟΦΟΡΙΚΗΣ CYPRUS COMPUTER SOCIETY ΠΑΓΚΥΠΡΙΟΣ ΜΑΘΗΤΙΚΟΣ ΔΙΑΓΩΝΙΣΜΟΣ ΠΛΗΡΟΦΟΡΙΚΗΣ 24/3/2007

ΓΡΑΜΜΙΚΟΣ & ΔΙΚΤΥΑΚΟΣ ΠΡΟΓΡΑΜΜΑΤΙΣΜΟΣ

Lecture 2: Dirac notation and a review of linear algebra Read Sakurai chapter 1, Baym chatper 3

Διπλωματική Εργασία του φοιτητή του Τμήματος Ηλεκτρολόγων Μηχανικών και Τεχνολογίας Υπολογιστών της Πολυτεχνικής Σχολής του Πανεπιστημίου Πατρών

65W PWM Output LED Driver. IDLV-65 series. File Name:IDLV-65-SPEC

Ερευνητικά πεδία στη φωτονική τεχνολογία. Φωτονική ολοκλήρωση Optical Computing Βιοϊατρική Αισθητήρες Τηλεπικοινωνίες

Προσομοίωση BP με το Bizagi Modeler

ΤΕΧΝΟΛΟΓΙΚΟ ΠΑΝΕΠΙΣΤΗΜΙΟ ΚΥΠΡΟΥ ΣΧΟΛΗ ΓΕΩΤΕΧΝΙΚΩΝ ΕΠΙΣΤΗΜΩΝ ΚΑΙ ΔΙΑΧΕΙΡΗΣΗΣ ΠΕΡΙΒΑΛΛΟΝΤΟΣ. Πτυχιακή εργασία ΑΝΑΠΤΥΞΗ ΔΕΙΚΤΩΝ ΠΟΙΟΤΗΤΑΣ ΕΔΑΦΟΥΣ

ΕΛΛΗΝΙΚΗ ΔΗΜΟΚΡΑΤΙΑ ΠΑΝΕΠΙΣΤΗΜΙΟ ΚΡΗΤΗΣ. Ψηφιακή Οικονομία. Διάλεξη 8η: Producer Behavior Mαρίνα Μπιτσάκη Τμήμα Επιστήμης Υπολογιστών

CHAPTER 101 FOURIER SERIES FOR PERIODIC FUNCTIONS OF PERIOD

Transcript:

15-16η Διάλεξη Point-to-point digital link design Γ. Έλληνας, Διάλεξη 15-16, σελ. 1 Outline Link Design Rise-Time Intro to WDM WDM-LANs Γ. Έλληνας, Διάλεξη 15-16, σελ. 2 Page 1

Point-to-point digital link design The simplest optical fiber communications system is a point-to-point link in which an optical transmitter and receiver are connected to one another via an optical fiber. This simple architecture is typical of those used in trans-oceanic links for example. Information source Optical transmitter Optical receiver Information recipient Optical fiber Γ. Έλληνας, Διάλεξη 15-16, σελ. 3 Point-to-point digital link design In digital communications, three key specifications are: 1. The length of the link (in km) 2. The bit rate (in Mb/s or Gb/s) 3.The bit error rate (BER). In addition, such considerations as component cost (per subscriber) and reliability also have to be taken care of. In satisfying the link specifications, a designer has a number of decisions to take, because... Γ. Έλληνας, Διάλεξη 15-16, σελ. 4 Page 2

Bandwidth (bit rate) and repeater spacing are determined by: Transmitter (e.g. laser) Power Modulation bandwidth Transmission Medium (fiber) λ Attenuation λ Δλ Dispersion Receiver (photodiode) λ Sensitivity Modulation bandwidth Γ. Έλληνας, Διάλεξη 15-16, σελ. 5 A. Choice of operating wavelength Short haul links (e.g. LANs) :- use short wavelengths (e.g. 0.85 μm). Moderate fiber losses can be tolerated and the technology is cheap. By using multimode fiber, connectors are more rugged than for single mode. Long haul links (e.g. transatlantic) :- use long wavelengths where attenuation and dispersion are low. (e.g. 1.3 μm - gives dispersion minimum, or 1.55 μm - has attenuation minimum and is compatible with optical amplifiers). Γ. Έλληνας, Διάλεξη 15-16, σελ. 6 Page 3

A. Choice of operating wavelength Loss of modern fibers Γ. Έλληνας, Διάλεξη 15-16, σελ. 7 A. Choice of operating wavelength 20 10 Dispersion (ps/(nm.km)) 0-10 -20 Dispersion for a silica single-mode fiber Γ. Έλληνας, Διάλεξη 15-16, σελ. 8 Page 4

B. Choice of source Power :- laser couples more power into single mode fiber than LED, but high-bit rate versions can be expensive and require temperature and optical power control. This makes them unsuitable for short links. Spectral width :- at short wavelengths (high material dispersion) LEDs with large spectral widths might cause problems with intersymbol interference. At 1.3 μm, we have very low dispersion fiber, which combined with low spectral width lasers allows high bit rates (e.g. 10 Gb/s and above). Γ. Έλληνας, Διάλεξη 15-16, σελ. 9 B. Choice of source Comparison of spectral widths Single-Mode Laser Diode Relative Power Density LED 3 to 6 nm Fabry-Perot Laser Diode < 1 pm 50 to 100 nm Wavelength Γ. Έλληνας, Διάλεξη 15-16, σελ. 10 Page 5

C. Choice of fiber Multimode :- modal dispersion limited can be used with LEDs and laser diodes graded index multimode fiber can achieve reasonable reduction in modal dispersion. Single-mode : no modal dispersion problems can only be used with laser diodes (high tolerance coupling) can support > 1 Tb/s (using WDM) small core diameter (8μm) leads to high tolerance (high price) connectors. Γ. Έλληνας, Διάλεξη 15-16, σελ. 11 D. Choice of photodetctor PIN :- simpler construction than APD relatively low sensitivity available for short and long wavelengths higher bandwidths achievable compared to APDs (up to 100 GHz) APD :- better receiver sensitivity temperature sensitive high bias voltages Γ. Έλληνας, Διάλεξη 15-16, σελ. 12 Page 6

Link Power Budget Put simply, the link power budget is an "accounting" procedure in which one calculates how much power can be lost between the transmitter and the receiver for a given receiver sensitivity (which depends on the bit rate) and transmitter power output. The resulting budget is allocated to connector losses, splice losses, fiber losses and a safety margin (system margin). db and dbm units are used in the link power budget. Γ. Έλληνας, Διάλεξη 15-16, σελ. 13 The db and dbm units See corresponding lecture for a discussion of this topic. Advantage of this approach is that it replaces multiplication & division with addition/subtraction in calculation of link gain/link loss. P in (dbm) G (db) P out (dbm) = P in (dbm) + G (db) P in (dbm) αl (db) P out (dbm) = P in (dbm) - αl (db) Γ. Έλληνας, Διάλεξη 15-16, σελ. 14 Page 7

power link budget LASER fiber PHOTODIODE P S (dbm) αl (db) P R (dbm) αl max = P S - P R Γ. Έλληνας, Διάλεξη 15-16, σελ. 15 LASER fiber OPTICAL AMPLIFIER PHOTODIODE fiber loss α db/km Loss due to fiber connector fiber pigtail; v. short, negligible loss fiber splice (permanent connection) Introduces splice loss Laser-to-fiber coupling loss; can be minimised using lenses Γ. Έλληνας, Διάλεξη 15-16, σελ. 16 Page 8

LASER fiber OPTICAL AMPLIFIER PHOTODIODE Power level (dbm) Total link loss M a Γ. Έλληνας, Διάλεξη 15-16, σελ. 17 Distance along link (km) Power link budget In practical applications, we often use components that have connectors attached (i.e. they are connectorized). fiber with one connector is known as a fiber pigtail. A length of fiber with connectors on both ends is called a patchcord (or jumper cable). In many link budgets, the splice loss is often lumped together with the fiber loss. We also include a safety factor known as the system margin (M a ) to account for component degradation. A typical value for M a is 6 db. Γ. Έλληνας, Διάλεξη 15-16, σελ. 18 Page 9

Example Calculate maximum link length for a system with: a connectorized laser transmitter (P S = 3 dbm) a connectorized receiver with sensitivity P R = -40 dbm a fiber patchcord (α F = 0.5 db/km, including splice losses) connector losses of α C = 1 db and system margin of 6 db Laser (P S ) α C fiber α F L α C Receiver (P R ) Total link loss (db) = P S - P R = α F L + 2 α C + M a Γ. Έλληνας, Διάλεξη 15-16, σελ. 19 α F L = P S - P R -2 α C -M a = 35 db Hence L max = 35 / 0.5 = 70 km Laser (P S ) P S = 3 2 Power level (dbm) α C α F L α C Receiver (P R ) α C α F L max N.B. Not to scale! -33-34 P R = -40 0 70 α C M a Distance (km) Γ. Έλληνας, Διάλεξη 15-16, σελ. 20 Page 10

Digital link design: Rise time budget In the previous example, we saw how the maximum link distance is affected by the fiber attenuation and also the source power and the photoreceiver sensitivity for a given bit rate; this gave us the link power budget. Information source Optical transmitter Optical receiver Information recipient Optical fiber Γ. Έλληνας, Διάλεξη 15-16, σελ. 21 power link budget LASER fiber PHOTODIODE P S (dbm) αl (db) P R (dbm) αl max = P S - P R - M a Γ. Έλληνας, Διάλεξη 15-16, σελ. 22 Page 11

Rise-time budget However, bit rate and repeater spacing are also determined by rise-time considerations: Transmitt er (e.g. laser) Power Transmission Medium (fiber) λ Attenuation Receiver (photodiode) λ Sensitivity Modulation bandwidth λ Δλ Dispersion Modulation bandwidth Γ. Έλληνας, Διάλεξη 15-16, σελ. 23 Concept of rise-time Any real-life system with an input/output will have a finite bandwidth. For example, consider typical modulation response of a laser diode: Γ. Έλληνας, Διάλεξη 15-16, σελ. 24 Page 12

i in (t) 0 Previous diagram relates to sinusoidal (jω) response. The corresponding step-response shows that it takes a finite time to reach the steady-state, and that in some cases there may even be relaxation oscillations: 0 Concept of rise-time t i in (t) p LD (t) p LD (t) 0 0 Typical step-response of a laser diode, showing turn-on delay and relaxation oscillations (due to low damping factor) Γ. Έλληνας, Διάλεξη 15-16, σελ. 25 t Similarly, a photodiode will take a finite time to respond to step-changes in the incidient optical power, as shown below for the case of a pulse input: p PD (t) Concept of rise-time p PD (t) i out (t) i out (t) 0 0 t 0 0 Note: Output current pulse shape depends on the device capacitance and also the width of the depletion region. The above response is quite poor due to large junction capcitance. Γ. Έλληνας, Διάλεξη 15-16, σελ. 26 t Page 13

Concept of rise-time Finally, the optical fiber itself will exhibit its own rise time due to the effects of dispersion. Γ. Έλληνας, Διάλεξη 15-16, σελ. 27 Concept of rise-time In the case of single-mode fibers, this is entirely due to intramodal dispersion, with the main contribution to this being being material dispersion. In multimode fibers, the dominant effect is intermodal dispersion. (Although material dispersion also exists, it is negligible in comparison). Although attenuation is important, it does not have an impact on rise-time. It affects the link power budget instead. Γ. Έλληνας, Διάλεξη 15-16, σελ. 28 Page 14

So, not surprisingly, the optical fiber link as a whole will have a rise-time (and fall-time) in response to a rectangular pulse input: i in (t) i in (t) Concept of rise-time p LD (t) fiber p PD (t) i out (t) i out (t) 0 0 t 0 0 Γ. Έλληνας, Διάλεξη 15-16, σελ. 29 t Definition of rise-time and fall-time N.B. y-axis is voltage, current or optical power as appropriate Rise-time: time taken to rise from 10% to 90% of the steady-state value of the pulse. Fall-time: time taken to fall from 90% to 10% of the steady-state value of the pulse. Γ. Έλληνας, Διάλεξη 15-16, σελ. 30 Page 15

Rise-time Budget Put simply, the rise-time budget is an "accounting" procedure in which one calculates how much pulse spreading can be tolerated between the transmitter and the receiver for a given transmitter rise-time, photoreceiver risetime and dispersion due to the fiber (both modal and chromatic, as appropriate). Γ. Έλληνας, Διάλεξη 15-16, σελ. 31 Rise-time Budget The total rise-time of the fiber-optic link is known as the system rise time t sys. It depends on the rise-times of the individual systems components, and assuming these are independent of one another, they affect t sys as follows: t = t + t + t + sys 2 TX 2 mat 2 mod t 2 RX Γ. Έλληνας, Διάλεξη 15-16, σελ. 32 Page 16

Rise-time Budget t TX = optical transmitter rise-time t RX = optical receiver rise-time t mat = material dispersion rise-time t mod = modal dispersion rise-time (for MM fiber only) The usual requirement on t sys is: t sys < 0.7 τ where τ is the pulse duration. Γ. Έλληνας, Διάλεξη 15-16, σελ. 33 LASER fiber rise-time budget PHOTODIODE t TX t mat t mod t = t + t + t + sys 2 TX 2 mat 2 mod t 2 RX t RX Γ. Έλληνας, Διάλεξη 15-16, σελ. 34 Page 17

Rise-time Budget The pulse duration depends on the data format. Two main data formats are used: NRZ and RZ Γ. Έλληνας, Διάλεξη 15-16, σελ. 35 Consider an NRZ stream composed of alternating 0 s and 1 s, and an RZ stream composed entirely of 1 s: τ NRZ NRZ RZ τ RZ T = 101010... 1/B T 111111... For NRZ signalling, τ NRZ = 1/B T, hence: t sys < 0.7 B T For RZ signalling, τ RZ = 1/2B T, hence: t sys 0.35 < B Γ. Έλληνας, Διάλεξη 15-16, σελ. 36 T Page 18

t mat : material dispersion rise-time due to material dispersion significant in single-mode fibers t mat = D mat σ λ L D mat = material dispersion parameter (ps/nm.km) σ λ = spectral width of optical source (in nm or μm) t mat can usually be neglected if the spectral width is narrow (e.g. in DFB lasers) or if operation is at 1.3 μm. Γ. Έλληνας, Διάλεξη 15-16, σελ. 37 t mod : modal dispersion rise-time due to (inter)modal dispersion dominant in multimode fibers In theory t mod is proportional to fiber length. In a real system, pulse distortion increases less rapidly after a certain initial length because of mode coupling. Γ. Έλληνας, Διάλεξη 15-16, σελ. 38 Page 19

t mod : modal dispersion rise-time Empirically, it can be shown that t mod in ns is given by: t mod = 0.44 L B O q where q is between 0.5 and 1.0 (depends on amount of mode coupling), L is the fiber length (km) and B O is the 3dB electrical bandwidth (in GHz) of 1 km of fiber. Γ. Έλληνας, Διάλεξη 15-16, σελ. 39 t RX : photoreceiver rise-time Assuming a simple low-pass RC characteristic for the frequency response of a photoreceiver, then we can relate t RX (in ns) to the 3 db receiver bandwidth (B RX in units of GHz) as follows: t RX = 0.35 B RX i signal i noise Cd R d A simple photodiode model; high frequency versions also include parasitics due to the packaging Γ. Έλληνας, Διάλεξη 15-16, σελ. 40 Page 20

t TX : optical transmitter rise-time This is a function of both the intrinsic frequency response (of either the LED or the laser diode) along with any drive electronics. LED and laser diode data sheets usually specify the device rise time. Γ. Έλληνας, Διάλεξη 15-16, σελ. 41 Finally... Although the above equations/analysis may appear to be straightforward, be VERY careful in using units. Bandwidths of laser diodes, for example, tend to be in the GHz range, so rise-times tend to be quoted in ns. However, it is possible to encounter a mix of units when performing rise-time calculations. If in any doubt, convert all quantities to SI units, and perform calculations in SI units, before converting to ns at the end. Γ. Έλληνας, Διάλεξη 15-16, σελ. 42 Page 21

Wavelength Division Multiplexing Limitations with earlier systems: Only a single wavelength is used, i.e. a minute fraction of the available optical bandwidth in fiber Only two end points are connected together, not allowing for a multi-user environment. Γ. Έλληνας, Διάλεξη 15-16, σελ. 43 Five generations of optical communications: Γ. Έλληνας, Διάλεξη 15-16, σελ. 44 Page 22

Wavelength Division Multiplexing Two developments have allowed us to start to access the massive bandwidth potential of optical fiber: Optical amplifiers: Provide amplification in the low loss window of optical fiber, and over a broad range of wavelengths (typically 40 nm centred on 1550 nm). Provide transparent operation, i.e. photons in, photons out. Wavelength division multiplexing: WDM components allow us to multiplex together different wavelength channels. Γ. Έλληνας, Διάλεξη 15-16, σελ. 45 How much bandwidth is there? Assume that our system will operate around 1550 nm and use optical amplifiers. The BW is given by the spectral range over which we get adequate amplification, say 10 db for 50 km spacing (50 km 0.2 db/km). Γ. Έλληνας, Διάλεξη 15-16, σελ. 46 Page 23

How much bandwidth is there? From earlier diagram, the following spectral bandwidth falls well within the 10 db gain requirement: 1530 nm 1550 nm 1570 nm λ 1 = c/f 1 λ 0 λ 2 = c/f 2 Δλ = λ 2 - λ 1 Δf = f 1 - f 2 λ 1 = 1530 nm, λ 2 = 1570 nm, i.e. λ 1 λ 2 λ 0 Γ. Έλληνας, Διάλεξη 15-16, σελ. 47 How much bandwidth is there? c c c 2 1 Δf = f1 f 2 = = λ λ λ λ Δf 1 2 λ 0 2 c Δλ ( λ λ ) 1 2 In this case, Δλ = 40 nm, λ 0 = 1550 nm which gives: Δf = 5 THz (= 5000 GHz) No laser exists that is capable of being modulated over this range. Γ. Έλληνας, Διάλεξη 15-16, σελ. 48 Page 24

How much bandwidth is there? The highest 3 db bandwidth achieved in lasers is about 30 GHz, and although modulators can do better, they are still limited to less than 100 GHz. Most off-the-shelf lasers are specified at no more than a few GHz. If we take a 1 GHz device at 1.55 μm, we can calculate how much spectral BW it occupies: 2 c Δλ λ0 Δf Δf Δλ = 0.008 nm 2 λ0 c In other words, a single laser occupies a very small percentage of the available spectral bandwidth. Γ. Έλληνας, Διάλεξη 15-16, σελ. 49 WDM These limitations can be overcome with wavelength division multiplexing (WDM). Many different wavelengths (from different lasers) share the same optical fiber. In early WDM, wavelengths were separated by 10 nm; current separation is down to 0.8 nm, and is called dense WDM (i.e. DWDM). WDM holds the promise of high bandwidth multi-user networks, for applications such as video-on-demand, telemedicine etc. in local, wide and metropolitan area networks (LAN, WAN, MAN). Γ. Έλληνας, Διάλεξη 15-16, σελ. 50 Page 25

WDM Spectral windows in single-mode fiber offer phenomenal bandwidths: Γ. Έλληνας, Διάλεξη 15-16, σελ. 51 WDM Present-day optical devices have bandwidths well below this (< 100 GHz), so to take advantage of the available fiber bandwidth, we can use many wavelengths in the 1.55 μm window (where we have the advantage of optical amplification): Loss, db/km λ 1 λ 2 λ n 0.2 0.1 1.3μm 1.55μm λ Γ. Έλληνας, Διάλεξη 15-16, σελ. 52 Page 26

WDM ITU (International Telecommunication Union) have set a standard spacing of 100 GHz for WDM systems (originates from FDM technology). For the 1550 nm window, this is 0.8 nm spacing. Therefore in our example of 5 THz bandwidth, we have 50 wavelength channels available. If each channel supports 10 Gb/s bit rate for example, then the aggregate bit rate would be 500 Gb/s or 0.5 Tb/s. It s estimated that with 1 Tb/s we could transmit all the world s TV station output simultaneously. Γ. Έλληνας, Διάλεξη 15-16, σελ. 53 US Fiber-Optic Backbone Γ. Έλληνας, Διάλεξη 15-16, σελ. 54 Page 27

BW Requirements Γ. Έλληνας, Διάλεξη 15-16, σελ. 55 Advantages of WDM: Capacity upgrades. If each wavelength can support a bit rate of a few Gb/s (40 Gb/s in state-of-the-art), then system capacity is increased massively by using many wavelengths. Transparency. Each optical channel (i.e. wavelength) can support any signal format (e.g. digital or analogue, TDM etc.) Wavelength rerouting and switching. Can switch wavelengths and route signals by wavelength, adding an extra dimension to network design. Γ. Έλληνας, Διάλεξη 15-16, σελ. 56 Page 28

WDM Components: MUX WDM multiplexer: used to combine several different wavelengths onto one fiber: λ 1 λ λ λ λ λ 1 2, 3, 4 2 MUX λ 3 λ 4 Should have low insertion loss. Γ. Έλληνας, Διάλεξη 15-16, σελ. 57 WDM Components: DMUX WDM demultiplexer: used to remove several different wavelengths from one fiber: λ 1 λ 1 λ 2, λ 3, λ 4 DEMUX λ 2 λ 3 λ 4 Should have low insertion loss, high selectivity Γ. Έλληνας, Διάλεξη 15-16, σελ. 58 Page 29

WDM Components: EDFAs and Filters Optical amplifier (EDFA): used to provide gain to wavelengths in the 1.55 μm band. Tuneable optical filter: used to filter out a single wavelength for a photodetector to produce a tuneable receiver: Input Passband tuned to third wavelength Output Γ. Έλληνας, Διάλεξη 15-16, σελ. 59 WDM Components: Tunable lasers, add/drop MUX Tuneable laser diodes (figures of merit are large wavelength tuning range, high-speed tunability, high data rate transmission, rigid wavelength stability and repeatability). Add/drop multiplexers for selective wavelength routing/extraction: Add λ add Input λ add Γ. Έλληνας, Διάλεξη 15-16, σελ. 60 Page 30

Components for WDM Tuneable laser isolator N x N star coupler EDFA Photodiode Tuneable optical filter Wavelength router Γ. Έλληνας, Διάλεξη 15-16, σελ. 61 Passive N N star coupler P 1 P 2 P N N N Power split out equally amongst all output fibers; power in any one output is: (P 1 + P 2 +... + P N ) N N/B. All input wavelengths are multiplexed onto each output Γ. Έλληνας, Διάλεξη 15-16, σελ. 62 Page 31

Application: Wavelength-selective WDM λ 1 Provides fixed wavelength links for N transmitter/receiver pairs over a single fiber: λ 1 λ λ λ λ λ 1 2, 3, 4 2 MUX DEMUX λ 3 λ 4 λ 2 λ 3 λ 4 Γ. Έλληνας, Διάλεξη 15-16, σελ. 63 Application: Broadband WDM More flexible: offers broadcast and select: λ 1 Power combiner EDFA Power splitter Filter λ 1 λ 2 N x 1 1 x N λ 3 λ n λ 1 Receivers can tune in to any of the broadcast wavelengths Power splitter/combiner has 10 log N db loss: need EDFA Γ. Έλληνας, Διάλεξη 15-16, σελ. 64 Page 32

WDM Local Area Networks Network hierarchy: LAN e.g. School ECE MAN e.g. Univ. of Cyprus LAN: Local area network MAN: Metropolitan area network WAN: Wide area network WAN e.g. Cyprus Γ. Έλληνας, Διάλεξη 15-16, σελ. 65 WDM LANs: Basic Architectures Dual rail (bus) configuration fiber λ 1 λ 2 λ N TX TX...... TX RX RX RX fiber coupler Γ. Έλληνας, Διάλεξη 15-16, σελ. 66 Page 33

WDM LANs: Basic Architectures Passive star configuration TX RX λ 1 λ 2 TX RX TX RX λ N Star Coupler λ 3 TX RX Γ. Έλληνας, Διάλεξη 15-16, σελ. 67 Experimental WDM Local Area Networks (i) LAMBDANET (developed by Bellcore). Node 1 16 16 Node 2 λ 2 λ 1 λ 16 Node 16 STAR COUPLER λ 3 λ4 λ 5 Node 3 Node 4 Node 5 Γ. Έλληνας, Διάλεξη 15-16, σελ. 68 Page 34

LAMBDANET Individual LAMBDANET node: Electronics interface Laser λ 1 Photoreceivers WDM DMUX Γ. Έλληνας, Διάλεξη 15-16, σελ. 69 LAMBDANET In LAMBDANET, each node is equipped with one fixed transmitter (DFB laser) emitting a unique wavelength and N fixed receivers. (N = no. of nodes in the network). Incoming wavelengths are separated using a wavelength demultiplexer, and each individual wavelength is sent to a photoreceiver. Each node s transmitter is fixed on that node s home wavelength. Γ. Έλληνας, Διάλεξη 15-16, σελ. 70 Page 35

LAMBDANET No tuneable components needed: relatively simple system to build. Other advantage is contention-free broadcast capability and support for one-to-one links as well as multicasting. Disadvantage: not easily scaleable, needs N data wavelengths for N nodes cost per node is high. 16-node system demonstrated with 2 Gb/s per channel. Γ. Έλληνας, Διάλεξη 15-16, σελ. 71 Experimental WDM Local Area Networks (ii) Rainbow (developed by IBM) NODE 1 λ 1 LASER TX. TUNABLE FILTER RX. STAR COUPLER λ 2 NODE 2 λ 1,λ 2. λ N λ 1,λ 2. λ N λ N NODE N λ 1,λ 2. λ N Γ. Έλληνας, Διάλεξη 15-16, σελ. 72 Page 36

Rainbow Broadcast and select architecture. Each node broadcasts a unique wavelength and is able to select any one of the wavelengths present in the network via a tuneable filter. Originally developed to network IBM PS/2 computers. Total capacity: 9.6 Gb/s, 32 nodes. Γ. Έλληνας, Διάλεξη 15-16, σελ. 73 Rainbow Protocol used to set up connections is as follows: (a) Idle receivers continually scan across all wavelengths. (b) If node wishes to transmit to node, then it continually sends a request (using λ 1 ) to for a connection. (c) When detects the request from, it locks its filter onto λ 1. (d) Node then sends a connection accept (using λ 2 ) to node. (e) When detects s acceptance, it locks its filter to λ 2, and a full duplex connection is established. Γ. Έλληνας, Διάλεξη 15-16, σελ. 74 Page 37

Experimental WDM Local Area Networks LAMBDANET and Rainbow are star topologies with N wavelengths assigned to N nodes (i.e. no wavelength re-use). Alternative topologies are possible, e.g. chain and ring. The following diagram shows a four-node ring network where add-drop multiplexers (ADM) are employed to allow wavelength re-use. Γ. Έλληνας, Διάλεξη 15-16, σελ. 75 Ring Network λ 1 λ 2 λ 3 λ 1 λ 2 λ 3 ADM λ 1 λ 4 λ 5 λ1 λ 4 λ 5 λ 4,λ 5,λ 6 NODE 1 λ 2,λ 3,λ λ 1,λ 2,λ 4 6 RING NODE 2 NETWORK NODE 4 NODE 3 λ 1,λ 3,λ 5 λ 3 λ 5 λ 6 λ 3 λ 5 λ 6 λ 2 λ 4 λ 6 λ 2 λ 4 λ 6 Γ. Έλληνας, Διάλεξη 15-16, σελ. 76 Page 38

Wavelength assignment table Alternative assignments are possible, as long as wavelengths are not in contention with one another; e.g. next diagram Γ. Έλληνας, Διάλεξη 15-16, σελ. 77 Wavelength assignment λ 2 λ 4 λ 6 λ 2 λ 4 λ 6 λ 3 λ 5 λ 6 λ3 λ 5 λ 6 λ 1,λ 3,λ 5 NODE 1 λ 1,λ 2,λ λ 2,λ 3,λ 6 4 NODE 2 NODE 4 NODE 3 λ 4,λ 5,λ 6 λ 1 λ 4 λ 5 λ 1 λ 4 λ 5 λ 1 λ 2 λ 3 λ 1 λ 2 λ 3 Γ. Έλληνας, Διάλεξη 15-16, σελ. 78 Page 39

Ring Networks Number of wavelengths added at each node equals the number that are received: all add/drop multiplexers are the same. Advantages: full interconnection between nodes is possible, i.e. any node can talk to any other. One might expect N 2 wavelengths would be needed to achieve this, but by re-using wavelengths as shown above, need far fewer. (e.g. for N = 4, only need 6, not 16). Γ. Έλληνας, Διάλεξη 15-16, σελ. 79 Broadcast & Select Multihop Networks Disadvantage of single-hop networks such as Rainbow is the need for rapidly tuneable lasers or receiver filters. Multi-hop networks overcome this problem by not having a direct connection between all node pairs. This allows each node to have a small number of fixed wavelength transmitters and receivers, i.e. node complexity is reduced. Γ. Έλληνας, Διάλεξη 15-16, σελ. 80 Page 40

Example: Shufflenet Multihop Network Passive star configuration λ 1 λ 2 λ 6 λ 8 Node 1 Node 2 Node 4 λ 5 λ 7 λ 7 λ 8 Star Coupler λ 3 λ 4 λ 3 λ 1 Node 3 λ 2 λ 4 λ 5 λ 6 Γ. Έλληνας, Διάλεξη 15-16, σελ. 81 Example: Shufflenet Multihop Network Dual rail configuration λ 1 λ 2 λ 3 λ 4 λ 5 λ 6 λ 7 λ 8 TX RX Node TX RX TX RX TX RX λ 5 λ 7 λ 6 λ 8 λ 1 λ 3 λ 2 λ 4 Γ. Έλληνας, Διάλεξη 15-16, σελ. 82 Page 41

Shufflenet Logical interconnection pattern and wavelength assignment: λ 1 1 3 1 λ 2 λ 5 λ 6 2 λ 3 λ 7 4 2 λ 4 λ 8 Γ. Έλληνας, Διάλεξη 15-16, σελ. 83 Shufflenet In general: k columns each have p k nodes, where p is the number of fixed transceiver pairs per node. The total number of nodes is then N = kp k The total number of wavelengths is N λ = pn = kp k + 1 The maximum number of hops needed to reach a given node is H max = 2k -1 Γ. Έλληνας, Διάλεξη 15-16, σελ. 84 Page 42

Shufflenet For example, for p = 2, k =2, we have: 1 λ 1 λ 2 5 λ 9 λ 10 1 2 λ 3 λ 4 6 λ 11 λ 12 2 3 λ 5 λ6 λ7 7 λ 13 λ 14 λ15 3 4 8 4 λ 8 λ 16 Γ. Έλληνας, Διάλεξη 15-16, σελ. 85 Shufflenet Total number of nodes is N = kp k = 2.2 2 = 8 Total number of wavelengths is N λ = pn = kp k + 1 = 16 The maximum number of hops needed to reach a given node is H max = 2k - 1 = 2.2-1 = 3 Γ. Έλληνας, Διάλεξη 15-16, σελ. 86 Page 43