Funems2013 Regular Presentation Orange Fronthaul

Optical fiber solution for mobile fronthaul to achieve Cloud Radio Access Network

Embedded R esonant esonant

and ModulablE Self-tuning laser cavity for next generation access network transmitter

Philippe CHANCLOU, Anna PIZZINAT, Fabien LE CLECH, To-Linh REEDEKER, Yannick LAGADEC, Fabienne SALIOU, Bertrand LE GUYADER, Laurent GUILLO, Qian DENIEL, Stephane GOSSELIN, Sy Dat LE, Thierno DIALLO, Romain BRENOT, Francois LELARGE, Lucia MARAZZI, Paola PAROLARI, Mario MARTINELLI, Sean O’DULL, Simon Arega GEBREWOLD, David HILLERKUSS, Juerg LEUTHOLD, Giancarlo GAVIOLI, Paola GALLI

Session 9e, 04 July 2013

contact: philippe.chanclou [email protected] @orange.com

Future Network & MobileSummit 2013

Copyright 2013 ORANGE Labs

Summary:



Context of Cloud Radio Access Network



Which are the main constraints of fronthaul?



Discussion on optical network for the fronthaul



Self-seeded WDM solution



Conclusion




Summary:















Conclusion




Optical architecture for Mobile fronthaul: Step 2: Distributed base station with « traditional » backhaul

Step 1: Macro M acro base station

RRH ~15kg

RRH

RRH RRH

coax

D-RoF RRU RRU

D-RoF BBU

RRU

RRU: Remote Radio Unit RRH: Remote Radio Head BBU: BaseBand Unit CSG: Cell-Site Gateway D-RoF: Digital Radio oe! Fi"e! #C$R% o! &BS'% standa!d( Session 9e, 04 July 2013

e l m e u t s d y o S m

BBU


CSG

 



RRH close to the antenna Energy savings

Space constraints in cell site cabinet



Step 3: BBU hostelling with stacking (or BBU centralisation) •

BBU colocalised in secured CO (no need for IPSec)

•

X2 optimisation,

•

Future proof with respect to LTE-A evolutions (CoMP support)

•

Energy and deployment savings (expected 20%), site engineering

•

Feasible today

RRH

Cent!al &..ice D-RoF

RRH

RRH RRH

optical distribution network

BBU


BBU


BBU

D-RoF

F!ont/a0l Session 9e, 04 July 2013



%$)*$+S netwo!k

,

S1

*'SG

Back/a0l Copyright 2013 ORANGE Labs

Step 4: C-RAN (or BBU hostelling with resource pooling) 4 Cs of C-RAN: Centralization, Cloud, Cooperation, Clean









RRH

RRH RRH



Same advantages as step 3, plus Less interfaces to core network (S1 and X2) Simplification of mobility management CAPEX savings due to reduced BBU number Trials ongoing in other countries

RRH

RRH RRH

-

D-RoF

RRH

RRH RRH

optical distribution network

F!ont/a0l

Cent!al &..ice g n d i a c o n a L l a b

,

BBU


%$)*$+S netwo!k

Back/a0l

RRH might be replaced by Active Antenna Arrays (AAA): new antennas with integrated RRH Session 9e, 04 July 2013



S1

Definition of fronthaul and backhaul

RRH

RRH RRH

Central Office g ta - o BBU RRH

RRH

BBU RRH

X2

IP/MPLS network

EPC


S1 MASG EN BBU

fronthaul Session 9e, 04 July 2013



CE BIOM


backhaul Copyright 2013 ORANGE Labs

Summary:















Conclusion




Main constraints of fronthaul for optical network 1/2

typical fronthaul requirements Data rate

Symmetric bit rate Maximum BER 10-12 Multi-rate: CPRI1 = 614.4 Mbit/s OBSAI1 = 768 Mbit/s CPRI2 = 1228.8 Mbit/s OBSAI2 = 1536 Mbit/s CPRI3 = 2457.6 Mbit/s CPRI4 & OBSAI3 = 3072.0 Mbit/s CPRI5 = 4915.2 Mbit/s CPRI6 &OBSAI4= 6144.0 Mbit/s CPRI7 = 9830.4 Mbit/s

Frequency accuracy

+/- 2 ppb (NB. Time & phase synchronization is required for LTEAdvanced; A phase accuracy requirement budget will be allocated to the CPRI link )

Jitter

Jitter values are specified according to CPRI requirements (v5.0 – 2011-09-21)




Main constraints of fronthaul for optical network 2/2

typical fronthaul requirements Latency vs. reach

LTE-A: 190µs round trip time (about 20km round trip)

Environment condition

RRH is an outdoor equipment (typ. -40 to +85°C)

Network operation

Scope of responsibility (fiber and radio network)




Fronthaul : scope of responsibility

RRH

RRH RRH

demarcation point demarcation point

Central Office e l m e u s o y S m

RRH

RRH RRH

Mobile operator Session 9e, 04 July 2013

Fiber provider Future Network & MobileSummit 2013

BBU


BBU

e l m e u t d s o y S m

Mobile operator Copyright 2013 ORANGE Labs

Summary:















Conclusion




Dedicated Fiber RRH

RRH

RRH

B&W* SFP

B&W* SFP

B&W* SFP Fiber Monitoring passive part

PRO’S CPRI

• no extra equipment cost for transmission

When PtP fibres is available Optional fiber monitoring

CON’S • requires many fibers • extra equipment for fiber infrastructure monitoring

Fiber Monitoring Active part part B&W* SFPs

BBU BBU Session 9e, 04 July 2013



CWDM : passive solution RRH

PRO’S • few fibers • No active components on passive mux • High MTBF • Suited for outdoor deployment • Low cost point for CWDM technology (cf. • No introduction of transport latency • up to 16 channels per fiber

CWDM Fixed color SFP

CWDM Fixed color SFP RRH

RRH

CWDM Fiber Monitoring passive part

passive devices CWDM

Fiber Monitoring Active part

CON’S • inventory management required to align optic color with RRH-BBU link • not bidirectional (2 fibers per link) • no native OAM

CWDM CWDM Fixed color SFPs

BBU BBU




RRH

DWDM : passive solution

PRO’S • 44/88/96 channels per fiber • Bidirectional • No active components (passive mux)

CPRI colorless DWDM RRH

Un-mapped traffic Fiber Monitoring passive part

Fiber Monitoring Active part

CPRI colorless DWDM

BBU BBU

CON’S • No native OAM




Fronthaul : Infrastructure monitoring principle

« Slave » passive box for demarcation point : - no touch of the traffic - optical attenuation 1 à 2 dB - monitor loopback - optical bandwidth : 1260 up to 1610 nm ∼

« master » box offering: - add and drop of supervision wavelength (1630nm) - optical attenuation 1 to 3 dB - optical bandwidth : 1260 up to 1610 nm - deliver alarms and supervision indicators -Power monitoring : per line or per wavelength ∼

1630nm Px1Px2 Tx Rx loop back for only 1630 nm

passive


active



Central Office

Summary:












Self-seeded DWDM solution



Conclusion




Fronthaul : CWDM solution

RRH

Monitoring wavelength Tx

D e M U X

M U X


Px2

Passive device offering filtering loop back


Central Office

Rx

Px1

Active Equipment

X U M

X U M e D


BBU

e l m e u t d s y o S m

BBU


BBU


Fronthaul : DWDM fronthaul



RRH



M U X / D e M U X


Automatic and passive assignment of the wavelength (colorless) Single fiber (bi-directional) architecture

Infrastrusture monitoring wavelength & channels monitor

? Passive device offering filtering loop back


Active Equipment

Central Office

X M e D / X U M


BBU


BBU


BBU


Fronthaul : DWDM fronthaul by self seeded (1/3) 



RRH

) x T ( A O S R


) x R ( D P A / N I P

SFP transceiver

M

X

X / D e M U X

M e D / X U M

i RSOA (Tx) d i B

self seeded source Session 9e, 04 July 2013



Central Office

BBU


BBU


BBU


Fronthaul : DWDM fronthaul by self seeded (2/3) 



RRH

) x T ( A O S R


) x R ( D P A / N I P

SFP transceiver

M

X

X / D e M U X

M e D / X U M

i RSOA (Tx) d i B

ASE source Amplified self-seeded source Session 9e, 04 July 2013

Sliced ASE by AWG



Central Office

BBU


BBU


BBU


Fronthaul : DWDM fronthaul by self seeded (3/3) Amplified Self Seeded 



RRH 

) x T ( A O S R

Automatic and passive assignment of the wavelength (colorless) Single fiber (bi-directional) architecture standard WDM ODN

) x R ( D P A / N I P

SFP transceiver

M

X

X / D e M U X

M e D / X U M

i RSOA (Tx) d i B

Amplified self seeded source Session 9e, 04 July 2013



Central Office

BBU


BBU


BBU


Fronthaul : DWDM fronthaul by self seeded 2,5Gbit/s results 0

ASE

-10

-2

Amplified Self-Seeded

BER@10-3 with FEC

-3

Sliced

-4

-20

-5

22dB

-30

) m B-40 d ( r e -50 w P

Sliced ASE

) R-6 E B ( -7 g o L-8

-60

-9

-70

-10

-80

-11

-90 1531,5

mp e Self-Seeded

ER=8dB

-12 1532,5

1533,5

1534,5

0

1535,5

Wavelength (nm)

•

Optical spectrum centered at 1533.7nm

(CH9 of

•

2

4 6 8 10 12 14 16 18 20 Feeder Optical budget (dB)

19dB of feeder optical budget

AWG)

•

22dB of optical improvement




Impact on cavity length log(BER)= f(Received power)@2.5Gb/s -2 -3 -4

SS-5km

-5

) R E-6 B ( g-7 o l

-

-8

m

-9 -10 -11 -12 -35

-30

-25 -20 -15 -10 Received power (dBm)

Self-seeded_10m-Cavity


-5

Self-seeded_5km-Cavity



Impact on cavity length

log(BER)= f(Received power)@2.5Gb/s

log(BER)= (cavity length)@2.5Gb/s

-2

-2

-3

-3

-4

-4

SS-5km

-5

) R E-6 B ( g-7 o l

Amp-SS-5km

-

-8

m

-9

Amp-SS-10m

-10

SS

-5 ) R-6 E B-7 ( g o L-8

-9

-11

-10

-12 -35

-30


-5



Amplified SS_10m-cavity

Amplified SS_5km-cavity


-11 -12


0

5

10 15 20 25 30 35 40 45 Cavity length (km)





-2

ER=5.6dB

-2

-3

-3

-4

-4

SS-5km

-5

) R E-6 B ( g-7 o l

Amp-SS-5km

-

-8

m

-9

Amp-SS-10m

-10

SS

-5 Amp-SS

) R-6 E B-7 ( g o L-8

-9

-11

ER=8dB

-10

-12 -35

-30


-5






-11 -12


0

5

10 15 20 25 30 35 40 45 Cavity length (km)





-2

-2

-3

BER@10-3 with FEC

-3

-4

-4

SS-5km

-5

) R E-6 B ( g-7 o l

Amp-SS-5km -

-8 -9

Amp-SS-10m

-10

SS

-5 ) R-6 E B-7 ( g o L-8

Amp-SS

-9

-11

45km-long external cavity

-10

-12 -35

-30


-5






-11 -12


0

5

10 15 20 25 30 35 40 45 Cavity length (km)


Conclusions

Expected gains from C-RAN (most come from BBU Hostelling)

- Radio Site engineering improvements (footprint reduced, energy efficiency, less operations on site..) - Radio performance improvements to be challenged with alternative solutions - Transport: leverage future-proof infrastructure and concentrate complexity at central office level

Technology short term possibilities

- CWDM: good, simple, cost effective option with additional “passive” fiber monitoring for first mile

Optical architecture perspective

- Self seeded source allows to achieve a colorless DWDM source with - automatic and passive wavelength assignment - single fiber (bidirectional) link - athermal wavelength allocation question : do we have solution up to 10Gbit/s?




Acknowlegment

FP7-ICT-2011-7


ERMES m edded R esonant and ModulablE Self-tuning laser cavity for next generation access network transmitter



Fronthaul : Architecture of « passive » solution Automatic fibre protection one link = 2 fibres

RRH

RRH RRH

path 1

M U X

path 2

e l m e u s o y S m

passive

RRH

RRH

X U M

Central Office

RRH

X U M

M U X




BBU


BBU

e l m e u t d s o y S m

Fronthaul : short term vs mid/long term solutions #Fibers on last mile

Synchro & latency

Integrated monitoring solution

i!"

Ok

“Passive” (done by extra equipment)

Short term for CWDM

#$% &2 t$ 4'

Ok

“Passive” (done by extra equipment)

Typ. OTN based (CPRI encapsulation)

Short term

(e)y l$% &1 t$ 2'

Ok (Freq. only)

Part of OTN

Passive + active WDM

Passive CWDM + OTN

Short term

#$% &2 t$ 4'

TBC

Passive + part of OTN

Synergy with FTTH

wdm // to gpon or integration in NGPON2

Long term

FTTH infra reused

TBC

PON OAM (OMCI)

CPRI over Ethernet

CPRI I/Q encapsulated in Ethernet

Long term

*edi+m t$ i!"

TBC

Ethernet OAM (MEF)

Name

Description

Avail.

Dedicated fibre

one fiber per RRH per techno

Short term

Passive WDM

Colored SFP (RRH & BBU) cwdm or dwdm

Active WDM




Funems2013 Regular Presentation Orange Fronthaul

Recommend Documents