Dr. Stefan BrückQualcomm Corporate R&D Center Germany

3G/4G Mobile Communications Systems

Chapter VII: Fundamental Radio Resource Management

2

Resource Management

Slide 2

Fundamental Radio Resource Management

� Scope of Radio Resource Management

� Radio Resource Management in UMTS� Cell Breathing

� Capacity – Coverage Trade-off

� Power Control� Near Far Problem

� Inner and outer loop power control

� Load Control

3

� Load Control

� Spreading Code Management in HSDPA

� Examples of HSDPA and HSUPA Performance

� Radio Resource Management in LTE� LTE UL Power Control

� Inter Cell Interference Coordination in LTE� Flexible Frequency Reuse

� Heterogeneous Deployments

� Enhanced Inter Cell Interference Coordination in LTE� Almost Blank Subframes (ABS)

Slide 3

� Efficient use of limited radio resources (power, code space, spectrum, time)

� Minimizing interference

� Flexibility regarding services (Quality of Service, user behaviour)

� Simple algorithms requiring small signalling overhead only

� Stability and overload protection

� Self adaptive in varying environments

RRM – High-Level Requirements

4

� Allow interoperability in multi-vendor environments

Radio Resource Management algorithms control the efficient use of resources with respect to interdependent objectives:

� cell coverage� cell capacity� quality of service

Slide 4

System evaluation and standardisation activities on three levels:

• Access Network (Architecture)• Mobility Management

(E)-UTRANInterfaces Architecture

Context of Radio Resource Management

5

• Radio Link (Physical Layer)

• Cellular Network Aspects (Layer 2&3) • Radio Resource Management

Cellular Network Protocols

Slide 5

Radio Resource Management

Packet DataControl

LoadControl

HandoverControl

Non-Access Stratum

Radio Resource Management Components

6

MediumAccessControl

PowerControl

Physical layer

Slide 6

Example of Coverage and Best Server Mapco

vera

ge

map

bes

t se

rver

map

7

Application: RF engineering (cell layout)Legend: red indicates high signal level, yellow indicates low level

cove

rag

e m

ap

Application: HO decisionLegend: color indicates cell with best CPICH in area

bes

t se

rver

map

Slide 7

Interference in CDMA Networks

Interference Problem

Multiple Access Interference Different users interfere dependent on the access scheme (T/F/CDMA)

Intra-Cell Interference Interference caused by users belonging to the same cell

Inter-Cell Interference Interference caused by users belonging to other cells

8

� Frequency reuse factor = 1

� CDMA is subject to high Multiple Access Interference (MAI)

� MAI can be separated in intra-cell and inter-cell interference

� Soft capacity: CDMA capacity (max. number of users) determined by interference is soft� TDMA capacity is given by available number of time slots

� Handling of interference is the main challenge in designing CDMA networks

Slide 8

Interference in CDMA - Uplink

9 Slide 9

Interference in CDMA - Downlink

10 Slide 10

Cell Breathing in CDMA

� Specified service specific C/I needs to be maintained at receiver to guarantee QoS

� Growth of traffic leads to an increase in interference power� Both inter-cell and intra-cell power

� If transmitter power cannot be further increased (max. link power), maximum cell size decreases since specific C/I at far receiver is required

� Load dependent maximum cell is referred to as cell breathing

11

� Load dependent maximum cell is referred to as cell breathing

� As network load changes over day, also the maximum cell size changes over day� Especially during busy hours the interference rises ⇒ small cell size

� Breathing is determined by intra-cell interference

⇒⇒⇒⇒ Active Load Control is needed

Slide 11

Cell Breathing

� Coverage depending on load: Load causes interference which reduces the area where a SIR sufficient for communication can be provided

Coverage low load Coverage medium load Coverage high load

12 Slide 12

Yellow area: Connection may drop or be blocked

Coverage vs. Capacity

� Capacity depends on:� QoS of the users (data rate, error performance (bit-error-rate))

� User behaviour (activity)

� Interference (intra- & inter-/noise)

� Number of carriers/ sectors

� Coverage (service area) depends on:

13

� Coverage (service area) depends on:� Interference (intra- & inter-cell) + noise

� Pathloss (propagation conditions)

� QoS of the users (data rate, error performance (bit-error-rate))

� Thus, trade-off between capacity and coverage

Slide 13

Coverage vs. Capacity

1

1.5

2

2.5

3

3.5

←Downlink

Max

imum

cel

l rad

ius

(km

)

13kbps circuit switched service capacity versus maximum cell radius

14

� Downlink limits capacity while uplink limits coverage

� Downlink depends more on the load (user share total transmit BS power)

0 20 40 60 80 100 120 140 160 180 2000

0.5

1

Uplink→

Erlangs (2% GOS)

Max

imum

cel

l rad

ius

(km

)

Slide 14

� Controls the setting of the transmit power in order to:� Keep the QoS within the required limits, e.g. data rate, delay and BER� Minimise interference, i.e. the overall power consumption

� Power control handles:� Path Loss (Near-Far-Problem), Shadowing (Log-Normal-Fading) and Fast

Fading (Rayleigh-, Ricean-Fading) � Environment (delay spread, UE speed, …) which implies different performance

of the deinterleaver and decoder

CDMA Power Control: Basics

15

� Three types of power control:� Inner loop power control� Outer loop power control (SIR-target adjustment)� Open loop power control (Power allocation)

� Downlink power overload control to protect amplifier� Gain Clipping (GC)� Aggregated Overload Control (AOC)

Slide 15

� Near-Far Problem:� Spreading sequences are not orthogonal

(multi-user interference)� Near mobile dominates� Signal to interference ratio (SIR) is lower for far

mobiles and performance degrades

� Problem can be resolved through dynamic NodeBNodeB

UE 1UE 1

Near-Far Problem in CDMA

16

� Problem can be resolved through dynamic power control to equalize all received power levels

� AND/OR: By means of joint multi-user detection (MUD)

NodeBNodeB

UE 2UE 2

Slide 16

Impact of Power Control

17

Source: H. Holma, A. Toskala (Ed.), “WCDMA for UMTS”,

Slide 17

6

6.5

7

7.5

Req

uir

ed U

L S

IR [

dB

]

Power Control Results

18

5

5.5

0 20 40 60 80 100 120

Velocity [km/h]

Req

uir

ed U

L S

IR [

dB

]

PedA

VehA

SIR requirement strongly depends on the environment (due to different fast fading conditions – Jakes models) ⇒ Outer loop power control needed to adapt SIR

Slide 18

� Closed loop power control is used on channels which are established in both directions, such as DCH

� The receiver generates transmit power commands (TPC) based on the estimated received quality; the TPC are send back to the transmitter in the opposite direction

� The transmit power is adjusted according to the received TPC

ReceiverReceiver DecoderDecoderPAPAUser dataUser dataUser dataUser data

Closed Loop Power Control in UMTS

19

ReceiverReceiver DecoderDecoder

SIRSIR--estimateestimate

BLERBLER--estimateestimate

MUXMUXDeMUXDeMUX

PAPA

TPCTPCcommandscommands

TPCTPCcommandscommands

Adjust SIRAdjust SIRtargettarget

Inner loopInner loop

OuterOuterlooploop

Slide 19

Mobile Station Base Station/RNC

� Inner Loop Power Control (1500 Hz)� Objective: Adjust transmit power to keep the received SIR at a given SIR target

� Realisation using the SIR-estimate after RAKE combining:� If SIRest > SIRtarget, then generate TPC “DOWN”

� If SIRest ≤ SIRtarget, then generate TPC “UP”

� Realisation in base station (Node B)

� Similar mechanisms for up- and downlink

� Outer Loop Power Control (100 Hz max)

Inner/Outer Loop Power Control

20

� Outer Loop Power Control (100 Hz max)� Objective: Meet the required reception quality, e.g. average BLER

� Realisation using the CRC attachment� If CRC ok, then decrease SIRtarget = SIRtarget – ∆down

� If CRC fails, then increase SIRtarget = SIRtarget + ∆up

∆down = ∆up × BLERtarget / (1–BLERtarget)

� Realisation in radio network controller (RNC)

Slide 20

� General limitations in 3rd generation mobile systems:

� Limited resources: spectrum, power, (code space)

� Not all service requests can be granted

� Different types of services in the same cellular environment

� Service parameters (user behaviour and QoS) and environmental conditions (propagation) vary over time

� CDMA-specific impacts:

Load Control: Basics

21

� CDMA-specific impacts:� Coverage in CDMA systems depends on the cell loading: Cell breathing

� CDMA systems may become unstable in highly loaded situationsdue to fast inner-loop power control

Slide 21

Load Control

� Main objective:� Avoid overload situations by controlling system load

� Monitor and controls radio resources of users

� Call Admission Control (CAC)� Admit or deny new users, new radio access bearers or new radio links

� Avoid overload situations, e.g. by means of blocking a new call

� Decisions are based on interference and resource measurements

22

� Congestion Control (ConC)� Monitor, detect and handle overload situations with the already connected users

� Bring the system back to a stable state, e.g. by means of dropping an existing call

� CAC and ConC decisions are based on network measurements averaged over hundreds of ms� No fast fading impacts in these measurments

Slide 22

Basic Resource Equations (CDMA)Rbi: data rate of user i hi: channel coefficient of user i

Ppilot: Pilot power Fi: Orthogonality factor of user i (multipath)

Ith: thermal noise

Downlink

Quality (RAKE):

Resource of user i:

thinterpilotj jii

iibi

t

bi

IIPPhF

PhRW

N

E

+++⋅⋅⋅⋅=

∑ )(

Power dTransmitte Total:Poo

ii P

P=α

23

Uplink

Quality (RAKE):

Resource of user i:

� Resource Consumption strongly depends on: data rate, quality (Eb/Nt), receiver structure (RAKE etc., channel estimation, path tracking, …)

� Non-linear relation between resource, data rate and required Eb/Nt

thinterij j

ibi

t

bi

IIP

PRW

N

E

++⋅=

∑ ≠ˆ

ˆ

ceInterferen Total:Iˆ

otbb

tb

o

ii NERW

NE

I

P

+==α

Slide 23

� Service/BLER-dependent resource consumption

� Uplink example:� Service I: Voice

Rb = 12.2kbps, Eb/Nt = 5dBαI = 0.99%

� Service II: DataRb = 144kbps, Eb/Nt = 3.1dBαII = 7.11%3

4

5

6

7

8

9

10

Eb

/Nt [

dB

]

Voice (12.2 kbps)BLER = 1%

Resource Consumption

24

II

with

W is the channel bandwidth and Rb is the data rate

0

1

2

10 100 1000

Data Rate Rb [kbps]

0,5% 1% 2% 5%

10% 20% 35% 50%

BLER = 1%Data (144 kbps)

BLER = 10%

αααα

)(

)(

tbb

tb

NERW

NE

+=α

Slide 24

8

10

12

14

16

18

20

No

ise

Ris

e/ P

ow

er R

ise

[dB

]

thr_CAC = 75%

thr_ConC = 90%

� UL Measurement

� Load Estimate

�

thintercelli i IIPI ++=∑ ∈ˆ

0

th

thcurrent

IINR

NRI

II

0

0

0 11

=

−=−=η

UL/DL Load Measures

25

0

2

4

6

8

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

System Load

No

ise

Ris

e/ P

ow

er R

ise

[dB

]

� DL Measurement

� Load Estimate

Pcelli i PPP +=∑ ∈0

p

Pcurrent

PPPR

PRP

PP

0

0

0 11

=

−=−=η

Slide 25

Call Admission Control

� Admitting new call always increases the cell loading

� CAC avoids overload by limiting this increase

� CAC load check:

� Threshold setting thrCAC

trade-off between

?CACnewcurrent thr≤α+η

8

10

12

14

16

18

20

No

ise

Ris

e/ P

ow

er R

ise

[dB

]

η_current + α_new

thr_CAC

26

CAC

trade-off between� Maximize capacity: prevent

excessive blocking � Avoid overload: thrCAC < thrConC

0

2

4

6

8

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

System Load

No

ise

Ris

e/ P

ow

er R

ise

[dB

]

η_current

η_current + α_new

Slide 26

� Admitting a new call always increases the cell load

� In order to avoid overload situations, the admission control will limit this load increase

� The principle is to check the current system load plus the expected resource consumption of the new call against the call admission threshold:

load + consumption ≤ thrCAC ?

Call Admission Control (CAC)

27

� In case the admission check fails, the basic strategy is to protect ongoing calls by denying the new user access to the system since dropping is assumed to be more annoying than blocking

� Admission control is required for uplink and downlink

� Arrivals of high-data-rate users that require a large amount of resources (especially in the downlink) may demand global information

Slide 27

Congestion Control

8

10

12

14

16

18

20

No

ise

Ris

e/ P

ow

er R

ise

[dB

]

η_current

NR/PR_max

� Even with efficient CAC overload situations still occur due to� Mobility (esp. downlink)� Activity

� During overload quality of allusers is deteriorated !

� Triggered by measurement

� Action: reduce offered traffic by

ConCcurrent thr≥η

28

0

2

4

6

8

0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

System Load

No

ise

Ris

e/ P

ow

er R

ise

[dB

]

η_reducedthr_ConC

� Action: reduce offered traffic by downgrading one user

� Threshold setting thrConC to preserve the coverage:� NRmax: limitation of MS single

transmit power Pimax� PRmax: limitation of BTS total

transmit power P0max

Slide 28

� Due to the mobility (especially of high data rate users) overload situations occur even if an efficient admission control algorithm is used!

� The congestion control is activated once the system load exceeds the congestion threshold:

load ≥ thrConC

� In order to overcome the overload situation the load must be lowered until

Congestion Control (ConC)

29

� In order to overcome the overload situation the load must be lowered until load < thrConC

� This is done by reducing the offered traffic� Lowering the data rate of one or several services that are insensitive to increased

delays – this might be the most preferred method

� Performing inter-frequency (inter-system) handover

� Removing one or several connections

� Global information may be required to minimize the number of altered connections

Slide 29

30%

35%

40%

45%

50%

Blo

ckin

g P

rob

abili

ty

thr_CAC = 50%thr_CAC = 75%thr_CAC = 90%

12%

14%

16%

18%

20%

Dro

pp

ing

Pro

bab

ility

thr_CAC = 50%thr_CAC = 75%thr_CAC = 90%

Tradeoff between blocking and droppingExample: 64k per user, urban

Call Admission Control: Simulation Results I

30

0%

5%

10%

15%

20%

25%

5 15 25 35 45 55

Offered Traffic [Erlang per s ite ]

Blo

ckin

g P

rob

abili

ty

0%

2%

4%

6%

8%

10%

5 15 25 35 45 55

Offered Traffic [Erlang per s ite ]

Dro

pp

ing

Pro

bab

ility

Slide 30

50%

60%

70%

80%

90%

Cel

l L

oad

ing

Cell load depending on CAC thresholdExample: 64k per user, urban

Call Admission Control: Simulation Results II

31

0%

10%

20%

30%

40%

5 15 25 35 45 55

Offered Traffic [Erlang per site]

Cel

l L

oad

ing

thr_CAC = 50%thr_CAC = 75%thr_CAC = 90%

Slide 31

� Flexibility� Asymmetrical data rates

� Very low to very high data rates

� Control information/user information

� Efficient transmission making good use of CDMA characteristics� Dedicated channel (DCH)

� Minimise transmission power by closed-loop power control

Packet Data Control: Channel Switching

32

� Minimise transmission power by closed-loop power control

� Independence between uplink and downlink capacity

� Common channel� Random access in the uplink (RACH)

� Dynamic scheduling in the downlink (FACH)

� Adaptive channel usage depending on traffic characteristics� Infrequent or short packets ⇒ Common channel (Cell_FACH)

� Frequent or large packets ⇒ Dedicated channel (Cell_DCH)

Slide 32

CELL_DCH CELL_FACH CELL_DCH

DCH Active Time

Channel Switching – Example

33

Page Download Time Reading Time

DCH Active Time“Chatty Applications”

� Example: Web service

� Chatty apps.: keep alive message, stock tickers, etc.(e.g. 100 bytes every 15 sec)

� Second stage: when no activity in CELL_FACH then switch to URA_PCH

Slide 33

� Burst Admission Control� Decision on starting packet data transmission

� Similar principle like call admission control, i.e. check the system load against a threshold that is usually different from thrCAC

� Time horizon: 100msec … 10sec

� Rate Adaptation� Choose data rate according to transmit power

Packet Data Control

34

� Choose data rate according to transmit power� UE nearby NodeB ⇒ high data rate

� UE at cell edges ⇒ low data rate

� Decrease data rate in case of overload, cf. congestion control

Slide 34

� Power Control:� Balances user received quality (BLER, SIR)� Users at cell center get less share of BTS

transmit power assigned than at cell edge� Occurrence of power overload

� Rate Adaptation:� Transmit power ~ data rate

UE 1UE 1

Power Control vs. Rate Adaptation

35

� Transmit power ~ data rate� Users at cell edge get lower data rate

assigned than at cell center� Reduces also power overload

� In UMTS combination of power control and rate adaptation on DCH

NodeBNodeB

UE 2UE 2low data ratelow data rateareaarea

high data ratehigh data rateareaarea

Slide 35

UMTS_urban, 50 kByte

20%

25%

30%

35%

40%

Ou

tag

e P

rob

abili

ty

384k64kadaptive

UMTS_urban, 50 kByte

4

5

6

7

8

Mea

n D

elay

[se

c]

384k

64k

adaptive

Rate Adaptation Performance

36

Rate adaptation significantly improves the RRM performance.

0%

5%

10%

15%

200 300 400 500 600

Cell Throughput [kBit/sec]

Ou

tag

e P

rob

abili

ty

0

1

2

3

200 300 400 500 600

Cell Throughput [kBit/sec]M

ean

Del

ay [

sec]

Slide 36

Spreading Code Management in HSDPA

a) OVSF Code Tree

SF=8

SF=4

SF=2

C16,0 C16,15

Border adjusted by CRNC

37

b) Transmit Power

SF=16

Codes reserved for HS-PDSCH/ HS-SCCH

C16,0 C16,15

Codes available for DCH/ common channels

Tx power available for HS-PDSCH/ HS-SCCH Tx power available for DCH/ common channels

Border adjusted by CRNC

� Note: CRNC assigns resources to Node B on a cell basis

Slide 37

� 36 cells network

� UMTS composite channel model

� FTP traffic model (2 Mbytedownload, 30 sec thinking time)

� The user throughput is decreased when increasing load

Load Impact

1000

1500

2000

2500

Throughput [kbit/sec]

Mean User Throughput

Aggregated Cell Throughput

Cell and User Throughput versus Load

38

decreased when increasing load due to the reduced service time

� The cell throughput increases with the load because overall more bytes are transferred in the same time

0

500

1000

4 6 8 10 12 14 16 18

Number of Users/ Cell

Throughput [kbit/sec]

Slide 38

� 36 cells network

� UMTS composite channel model

� FTP traffic model (2 Mbyte download, 30 sec thinking time)

� Higher category offers higher max. throughput limit

Cat 6 - Cat 8 Comparison

1500

2000

2500

thro

ug

hp

ut

(kb

ps)

Mean User ThroughputPeak User ThroughputAggregated Cell Throughput

HSDPA performance per UE Category

39

max. throughput limit� Cat.6: 3.6 MBit/sec

� Cat.8: 7.2 MBit/sec

� Max. user perceived performance increased at low loading

� Cell performance slightly better

0

500

1000

Cat 6/ 10 users Cat 8/ 10 users Cat 6/ 20 users Cat 8/ 20 users

thro

ug

hp

ut

(kb

ps)

Slide 39

� Example Scenario� 15 users/cell

� Pedestrian A channel model

� Plot generated with field

HSDPA Coverage Prediction

40

� Plot generated with field prediction tool

HSDPA Throughput depends on location

Slide 40

800

1000

1200

Use

r Th

roug

hput

[kbp

s]

10ms TTI, unlimited CE dec. rate 2ms TTI, next release

User vs. Aggregate Cell Throughput for HSUPA

� 36 cells network

� UMTS composite channel model

� FTP traffic model (2 Mbyte upload, 30 seconds thinking time)

#UEs/cell1

2

3

4

41

0

200

400

600

200 400 600 800 1000 1200 1400 1600

Aggregated Cell Throughput [kbps]

Use

r Th

roug

hput

[kbp

s]

� Maximum cell throughput reached for about 7…8 UEs per cell

� Cell throughput drops if #UEs increases further since the associated signaling channel consume UL resources too

5

6

7

8

910

Slide 41

Single User Performance in HSUPA

� Average user throughput (RLC layer) for different channel profiles

� 1 UE in the network

� 1 target HARQ transmission

� For AWGN channel conditions:

� 10ms TTI: up to 1.7 Mbps 2000

2500

3000

3500

Ave

rag

e U

ser

Th

rou

gh

pu

t [k

bp

s]

2ms, 1Tx 10ms, 1Tx

42

� 10ms TTI: up to 1.7 Mbps (near theoretical limit of 1.88 Mbps)

� 2ms TTI: up to 3 Mbps (below theoretical limit 5.44 Mbps)

� E.g. due to restrictions from RLC layer (window size, PDU size)

0

500

1000

1500

AWGN PedA3 PedA30 VehA30 VehA120

Scenario

Ave

rag

e U

ser

Th

rou

gh

pu

t [k

bp

s]

Slide 42

LTE Uplink Power Control

� Open-loop power control is the baseline uplink power control method in LTE (compensation for path loss and fading)� Open-loop PC is needed to constrain the dynamic

range between signals received from different UEs

� Unlike CDMA, there is no intra cell interference to combat; rather, fading is exploited by rate control

� Transmit power per PRB

TxPSD = α•PL + P0Target SINR on PUSCH is now a function of the UE’s path loss:

43

TxPSD(dBm) = α•PL(dB) + P0nominal (dBm)

� PLdB: pathloss, estimated from DL reference signal

� P0nominal (dBm) = Γnominal (dB) + Itot (dBm)

Sum of SINR target Γnominal and total interference Itot sent on BCH

� Fractional compensation factor α ≤ 1 (PUSCH) → only a fraction of the path loss is compensated

� Additionally, (slow) closed loop PC can be used

Target SINR

of the UE’s path loss:

SINR(dBm) = Γnominal (dB) + (1–α)•PL(dB)

Slide 43

Interference Coordination - Flexible Frequency ReuseCell edge

Reuse > 1

Cell centre

Reuse = 1

44

� Cell edge users with frequency reuse > 1,

� eNB transmits with higher power

� Improved SINR conditions

� Cell centre users can use whole frequency band

� eNB transmits with reduced power

� Less interference to other cells

� Flexible frequency reuse realized through intelligent scheduling and power allocation

� Scheduler can place restriction on which PRBs can be used in which sectors

� Achieves frequency reuse > 1

� Reduced inter-cell interference leads to improved SINR, especially at cell-edge

� Reduction in available transmission bandwidth leads to poor overall spectral efficiency

Slide 44

Need for Flexible and Low-Cost Network Deployment Using Mix of Macro, Pico, Relay, RRH and Home eNBs

Heterogeneous Networks

MacroHeNB

Core Network

Internet

Relay

Pico

Backhaul

Relay Backhaul

Pico Pico

45

� Network expansion due to varying traffic demand & RF environment� Cell-splitting of traditional macro deployments is complex and iterative� Indoor coverage and need for site acquisition add to the challenge

� Future network deployments based on Heterogeneous Networks (HeNBs)� Deployment of Macro eNBs for initial coverage only� Addition of Pico, HeNBs and Relays for capacity growth & better user experience

� Improved in-building coverage and flexible site acquisition with low power base stations� Relays provide coverage extension with no incremental backhaul expense

Slide 45

HetNet: Macro-Pico (open access hot-spot)

� Macro-pico deployments with UEs operating in range expansion

� Large bias to compensate the power difference between macro and picofor traffic offloading

� Nominally a UE associates with a base station with strong DL SINR

� With range expansion, a UE can associate with a low power node based on smaller path loss, thus offloading the macro station

� With range expansion the serving cell is not necessarily the strongest one

46

MacroPicoPico

� With range expansion the serving cell is not necessarily the strongest one

Slide 46

HetNet: Macro-Femto (CSG node, HeNB)

� Macro-femto deployments with UEs (or femto cells) operating under strong co-channel interference

� UEs in close proximity of a CSG femto cell they are not allowed to connect to (i.e. are still served by macro cell) may be:

� strongly interfered in DL by the CSG cell

� causing severe interference in UL to towards the femto cell

� Interference management helps the UE to survive

47

� Interference management helps the UE to survive

MacroCSG CSG

Slide 47

Extending X2 to HeNB – 3GPP Status Rel. 10

� During Rel. 10 work, several contributions have been discussed related to introduction of X2 interface for HeNB for mobility enhancement

� HeNB X2 scenarios� HeNB-HeNB when the target cell is an open access HeNB

� HeNB-HeNB for closed/hybrid access with same CSG ID

� The existing X2 functionality is available for eNB and HeNB

48

� The existing X2 functionality is available for eNB and HeNB� There is no separate X2 specification for HeNB and eNB

� The X2 functionality is optional

� X2 interface between eNB and HeNB has been also discussed but no agreement could be reached so far� Scenario is more complex, as there may be a need for X2-GW which requires

additional standardization work

� Postponed to Rel. 11

Slide 48

Enhanced ICIC in Rel. 10

� ICIC includes frequency and time domain components� Frequency domain ICIC was already available in Rel. 8 and Rel. 9

� See lecture chapter 4 for details

� For the time domain ICIC, Almost Blank Subframes (ABSs) are used to protect resources receiving strong inter-cell interference (R3-103775)� Extensions to X2AP are needed� Time domain updates in Rel. 10 are called enhanced ICIC

49

� Time domain updates in Rel. 10 are called enhanced ICIC

� Both scenarios are handled differently, since there is no eNB – HeNB X2� Semi-static eICIC in macro eNB – pico eNB

� Static eICIC in eNB – HeNB scenario� ABS are defined by OAM for time domain eICIC (R3-103775)

� References: � R3-103775, “X2 procedure and OAM requirements to support eICIC”, TSG-RAN WG3 Meeting #70,

November 2010

Slide 49

Updates of the X2AP to support eICIC in Rel10

� Information of ABS pattern is added to the ‘Load Indication’ procedure� The ABS pattern (40ms period for FDD) is sent from the aggressor to the victim

� The ABS patterns are semi-statically updated

� The ‘Load indication’ also contains an invoke indication� This indicator is sent from the victim to the aggressor to ask for eICIC activation

IE/Group Name Presence Range IE type and reference

Semantics description

Invoke Indication M ENUMERATED (ABS

Information, …)

–

50

� Information about the status of the ABS is added to the ‘Resource Status Reporting’ procedure� Its initialization is enabled in the ‘Resource Status Reporting Initiation’ procedure

� This information is sent from the victim to the aggressor

� References: � R3-103667, “Introduction of X2 signaling support for eICIC”, TSG-RAN WG3 Meeting #70, November 2010

� R3-103776, “Enabling reporting of ABS resource status for eICIC purposes”, TSG-RAN WG3 Meeting #70, November 2010

IE/Group Name Presence Range IE type and reference Semantics descriptionDL ABS status M INTEGER (0..100) Percentage of ABS

resource allocated for UEs protected by ABS from strong inter-cell interference.

Slide 50