7.rrm overview

Company Confidential

1 © Nokia Siemens Networks Radio Resource Manager / June2007

Radio Resource Manager Overview



Contents

• RRM Introduction• Power Control• Load Control• Admission Control• Packet Scheduler• Resource Manager



RRM Introduction



Radio Resource Management

Target for RRM is to ensure the RAN offers:

• The planned coverage for each targeted service• High capacity i.e. low blocking (new calls, handovers)• The required Quality of Service (QoS)• Optimize the use of available capacity (priorities)

By continuously monitoring/adjusting how the available resources are used in accordance with user requests

Radio Resource Management (RRM) is responsible for optimalutilisation of the air interface resources

RRM

Link Quality

Cell CoverageCell Capacity



RRM must be able to:

• Predict the impact on interference (power) of admitting a new user for UL & DL

• Perform appropriate actions (e.g. new call admissions, bitrate increase/decrease etc.) in accordance with prevailing load conditions

• Provide different quality of service for real time (RT) and non-real time (NRT) users

• Take appropriate corrective action when the different cell load thresholds are exceeded in order to maintain cell stability (i.e. load control)

Overload

Load TargetOverload Margin

Po

we

r

Time

Estimated capacity for NRT trafficMeasured load causedby non-controllable load (RT)

RT services must have higher quality assurance than NRT

Radio Resource Management tasks



RRM Module

• Power Control• Load Control• Admission Control• Packet Scheduling• Handover Control• Resource Manager



• RRM is made up of a number of closely interdependent functions (i.e. algorithms)

• These functions can be divided into;

• Cell Based• Load control (LC)

• Admission control (AC)

• Packet scheduling (PS)

• Resource manager (RM)

• Connection Based• Handover control (HC)

• Power control (PC)

RRM Functional Split

PC

HC

Connection based functions

LC

AC

Cell based functions

PSRM



Load Control



…

BTS Measurements

RRI Period

Radio Resource Indication



Load Estimation Methods

• Actual prevailing load in the cell depends on multiple factors

• Number of active connections• Properties of the connections

• Bit rate

• Eb/No requirement

• BLER requirement

• Propagation channel conditions (speed, multipath profile etc.)

• SHO condition

• Activity

• Different methods can be applied to measure/estimate prevailing load conditions, e.g.

• Throughput Based Load Estimation• Power Based Load Estimation



Throughput Based Load Estimation

• The downlink load of the cell can be estimated by using the sum of the downlink allocated bit rates as follows

• The uplink load of the cell can be estimated by using the sum of the load factors of the users connected to this cell.

• Definition of Rmax and loadj requires estimates of Eb/No, little i,

activity, SHO overhead etc.

max

1

R

RN

jj

DL

∑==η

∑=

⋅+=N

jjUL loadi

1

)1(η

Max. allowed throughtput of the cell



Power Based Load Estimation

• The downlink load of the cell can be estimated by dividing the total downlink transmission power, Ptx_total by the maximum Node B transmission

power Ptx_BTS,max.

• The uplink load can be estimated with Prx_noise, the background and

receiver noise and Prx_total, the total received power, according to this

formula:

• Measures the load in actual conditions

max,_

_

BTStx

totaltxDL P

P=η

NRP

P

totalrx

noiserxUL

111

_

_ −=−=η



• The BTS measures the total received power (PrxTotal) and the total transmitted power (PtxTotal) on cell basis

• The BTS reports PrxTo ta l and PtxTo ta l of each cell to the RNC by sending RADIO RESOURCE INDICATION message periodically (RRIndicationPeriod)

• LC updates cell load status for each cell based on RADIO RESOURCE INDICATION

• LC updates non-controllable UL (PrxNC) and DL (PtxNC) load in cell

• AC and PS algorithms work on the current cell load status provided by LC

• Denying call admission (AC) and throttling back NRT traffic (PS) are the overload actions

• After scheduling PS provides LC with PrxNRT, PtxNRT & LNRT estimates

• After admitting RT RAB, AC provides LC with NC load increase estimate

Radio interface load



UL power based load measurement

PRx_Own

PRxTotal = PRx_Own + PRx_Other + PNoise

= PRx_NRT + PRx_RT + PRx_Other + PNoise

= PRx_NRT + PRx_NC

PRx_Other

iUL = PRx_Other / PRx_Own



DL power based load measurement

PTx_Own

PTxTotal = PTx_Own

= PTx_NRT + PTx_RT + PTx_Common

= PTx_NRT + PTx_NC

PTx_Common

Common Channels

PRx_Own_DL

PRx_Other_DL

iDL = PRx_Other_DL / PRx_Own_DL



UL Preventive & Overload Thresholds

Prx Target [dB]

PrxTarget [dB] + PrxOffset [dB]

load factor η

Range [0..1]

tota

l rec

eive

d po

wer

Prx T

otal

[dB

m] Overloaded Area

Marginal Load Area

Feasible Load Area

Range Prx Total [Prx noise ... inf]



Uplink preventive threshold

Preventive threshold = PrxTargetPrx Target is relative to the system noise, it gives an upper threshold for the noise rise

Target threshold defines the optimal operating point of the cell load, up to which PS & AC can operate normally

If cell load exceeds these limits then AC & PS move to PREVENTIVE STATE function

New RT RABs are blocked, and PS can't schedule more NRT bit rates in the cell

PrxTarget range: 0...30 dB, step 0.1 dBdefault: 4 dB



Overload threshold = PrxTarget + PrxOffset

• Overload Threshold defines the limit when the cell is considered to be overloaded• If load in the cell exceeds these limits then AC & PS move to OVERLOAD STATE

function• New calls are blocked, and PS starts to decrease NRT bit rates in the cell

PrxOffset range: 0...6 dB, step 0.1 dBdefault: 1 dB

Uplink overload threshold



DL Preventive & Overload Thresholds

Ptx Target [dBm]

Ptx Target [dBm] + PtxOffset [dB]

load factorη

tota

l tra

nsm

itte

d po

wer

Ptx T

otal

[dB

m]

Overloaded Area

Marginal Load Area

Feasible Load Area

Cell Maximum

Range Ptx Total [0… Ptx BS total]



Preventive threshold = PtxTarget

• Target threshold defines the optimal operating point of the cell load, up to which PS & AC can operate normally

• If cell load exceeds these limits then AC & PS move to PREVENTIVE STATE function

• New RT RABs are blocked, and PS can't schedule more NRT bit rates in the cell

PtxTargetrange: -10...50 dBm, step 0.1 dBdefault: 40 dBm

• Default value depends on the cell max TX power: in case the cell max power is 43 dBm, the PtxTarget should be 40 dBm (3 dB below max)

Downlink Preventive threshold



Overload threshold = PtxTarget + PtxOffset

• Overload Threshold defines the limit when the cell is considered to be overloaded• If load in the cell exceeds these limits then AC & PS move to OVERLOAD STATE

function• New calls are blocked, and PS starts to decrease NRT bit rates in the cell

PtxOffset range: 0...6 dB, step 0.1 dBdefault: 1 dB

Downlink overload threshold



Admission Control



• Maximises capacity whilst maintaining stability

• Decides if new RAB request is admitted into RAN

• AC decision procedure set according to whether;

• Request is for RRC connection and RT or NRT RAB Setup• RAB setup can be for call setup or handover

• Admission control for RAB setup is different for RT and NRT

• For RT RAB admission requests AC;

• estimates the non-controllable power (load) increase that would result from admitting the new RAB

• checks if the new non-controllable load is below a certain threshold

• Bearer is not admitted if the predicted load exceeds defined thresholds in UL or DL

• AC is also responsible for determining quality requirements of the RB including;

• setting RLC and TrCH parameters• BLER & Eb/No targets

• initial SIR target (used in Outer Loop PC) & upper and lower limits for the uplink SIR target

• AC determines the power allocation for the requesting UEs (initial, minimum and maximum transmission power)

LCLC

AC

Network based functions

PSPSRMRM

RNCAdmission Control Functional Overview



RAB Establishment

6

RAB attributes (HLR);• SDU error ratio• traffic class• max bit rate

NokiaRNC

Core NetworkRAB ASSIGNMENT

REQUESTNRT PS call

ACRAB admitted

RAB request denied

Queue RAB

Radio Access Bearer Service Request

1Quality Requirements

of Radio Bearer2

RB attributes (RNC);• target BLER• target Eb/No• initial SIRtarget

Admission Decision

4

RT CS call

Power Increase Estimates

3

UL/DL Load Change Report to LC

5

RRC Connection Establishment

0

NRT Admission Decision Process

(PS)

Admission Control Functional Overview



Load Based Admission Decision Process

• To maintain stability, UL and DL loads at each cell must be maintained below defined thresholds.

• Admission decision takes into account 3 main issues;

• The measured power quantities (current load status of the cell)

• Average total wideband UL received power

• Average total DL transmit power

• Non-controllable UL power

• Non-controllable DL power

• Non-controllable power increase estimation associated with new admissions

• Comparison against admission criteria thresholds

totalrxP _

NRTrxtotalrxNCrx PPP ___ −=

totaltxP _

NRTtxtotaltxNCtx PPP ___ −=



UL Admission Procedure Summary

BTS sends periodically the received UL power to the RNCBTS sends periodically the received UL power to the RNC

Fractional load[0..1]

OVERLOADAREA

MARGINALLOADAREA

FEASIBLELOADAREA

0

Load curve in UL

PrxTotal[dBm]

PrxNoise [dBm]

PrxTarget [dB]

PrxTarget [dB]+PrxOffset [dB]

1

Noise RiseNR [dB]

XX [dB]

RNC compares the measured received power levels against the set thresholdsRNC compares the measured received power levels against the set thresholds

If measured UL (PrxTotal) load exceeds target thresholds (PrxTarget) The NRT RAB bitrate can not be increased and remains at the same level as after previous scheduling period

If measured UL (PrxTotal) load exceeds overload thresholds (PrxTarget+PrxOffset) no RABs can be admitted and NRT bitrates are reduced until PrxTotal reaches again PrxTarget

Over Load

MarginalLoad

FeasibleLoad

In feasible load area the admission decision is based on the power rise estimate of the new RT bearer

If the resulting power is still below PrxTraget the RAB is admitted

rx_targetrx_ncrx_nc PPP ≥∆+In case the RAB can not be admitted it is put into the queueIn case the RAB can not be admitted it is put into the queue



BTS sends periodically the total transmitted DL power to the RNCBTS sends periodically the total transmitted DL power to the RNC

RNC compares the measured transmitted power levels against the thresholdsRNC compares the measured transmitted power levels against the thresholds

If measured DL (PtxTotal) transmitted power exceeds target thresholds (PtxTarget) ,The NRT RAB bitrates can not be increased and they remain at the same level as after previous scheduling period

If measured DL (PTxTotal) transmitted power exceeds overload thresholds (PtxTarget+PtxOffset) no RABs can be admitted and NRT bitrates are reduced until PtxTotal reaches again PtxTarget

Over Load

MarginalLoad

FeasibleLoad

In feasible load area the admission decision is based on the power rise estimate of the new RT bearer

If the resulting power is still below PtxTraget the RAB is admitted

In case the RAB can not be admitted it is putinto the queueIn case the RAB can not be admitted it is putinto the queue

OVERLOAD AREA

MARGINALLOADAREA

FEASIBLELOADAREA

Load curve in DL

PtxTotal[dBm]

PtxTarget [dBm]

PtxTarget [dBm]+PtxOffset [dB]

Cell maximum [dBm]

Load[0...1]

0 1

tx_targettxtx_total PPP ≥∆+

DL Admission Procedure Summary



NRTNRT

RRC connection setupRRC connection setup RAB setupRAB setup

RRC connection request is not rejected either for received wide band power or transmitted power reasons

RT over NRT and pre-emption procedure can be applied

RRC connection request is not rejected either for received wide band power or transmitted power reasons

RT over NRT and pre-emption procedure can be applied

Admitted if non-controllable load is below target threshold

RT over NRT pre-emption procedure can be applied


RT over NRT pre-emption procedure can be applied

Emergency call

Emergency call

RTRT



Admitted if non-controllable load added by estimated change is below target

If non-controllable load added by estimated change is above target, RT RAB pre-emption procedure can be applied

In case of congestion, RT over NRT procedure can be applied

Admitted if non-controllable load added by estimated change is below target

If non-controllable load added by estimated change is above target, RT RAB pre-emption procedure can be applied

In case of congestion, RT over NRT procedure can be applied

Admitted always at 0 bit rate, capacity requests scheduled by PS

Admitted always at 0 bit rate, capacity requests scheduled by PS

Admission Control for RT and NRT



Packet Scheduler



• It is characteristic for RT traffic that it’s load cannot be controlled in efficient way. Load caused by RT traffic, interference from other cell users and noise together is called non-controllable load.

• The available capacity, which is not used by non-controllable load, can be used for NRT radio bearers on best effort basis. To fill the whole load budget and achieve the maximum capacity, the allocation of non-GBR traffic needs to be fast.

• The Packet scheduler is a general feature, which takes care of scheduling radio resources for NRT radio bearers for both uplink and downlink directions; Packet scheduling happens periodically and is implemented for both dedicated (DCH) and common control transport channels (RACH/FACH).

• Scheduled capacity depends on the UE capabilities, Node B capabilities, current load of the cell as well as the availability of the physical radio resources.

Why Packet Scheduling ?



• Packet Scheduler switches between common channels (FACH/RACH = low capacity) and dedicated channels (DCH = higher bit rates)

• Packet Scheduler allocates to RABs temporarily dedicated channels with a set of maximum bit rates

• For instance within an allocation for 384kbit/s, the instantaneous bit rate can be {0, 8, 16, 32, 64, 128, 384} kbit/s

• Packet Scheduler allocates DCH based on Capacity Requests• A Ca pa c ity Re q ue s t (Nokia term) is triggered based on traffic volume

measurement info: the sender (UE or RNC) has data in buffer and no sufficient dedicated channel

• Packet Scheduler releases DCH upon inactivity• Packet Scheduler re-schedules continuously DCH to serve all requests

equally, and take into account changes in non-controllable load

Packet Scheduling Principles



Power/ Load

time

non-controllable loadi.e. RT Traffic

controllable loadi.e. non RT Traffic

PtxTotal (variable) PrxTotal

PtxNrt PrxNrt

PtxTarget PrxTarget

PtxOffset PrxOffset

PtxTargetBTS PrxTargetBTS

overload

De s iredPrx /PtxVa lue sPtxRt PrxRt

Packet Scheduling



Packet Scheduler as part of RRM

• The packet scheduler (PS) co-operates with other radio resource management functions like handover control (HC), load control (LC), admission control (AC) and the resource manager (RM)

• HC provides active set information• LC provides periodical load information to PS on a cell basis• PS informs AC and LC of the load caused by non-real time radio bearers

• AC informs PS when new non-real time radio bearers are admitted, reconfigured or released

• RM allocates the RNC internal resources, downlink spreading codes and takes care of allocating radio links using the base station application protocol (NBAP)



Packet Scheduler functions

• Packet Scheduler consists of multiple different functions which can be categorised based on the scope of the function

• UE-specific part• Functions working based on single radio link/bearer status, measurements and conditions

• Cell-specific part• Functions working based on cell level measurements and conditions



Minimum bit rate

Packet Scheduler actions during call – Loaded cell

Max. bit rate

Initial bit rate

Allocated bit rate

Actual throughput

Load Margin

Normal load

Overload

AC

PS1

PS2

PS3

PS4

PS5

PBS FLXU EOLC FLXU



URA_PCH

CELL_DCH CELL_FACH

CELL_PCH

UTRA RRC Connected Mode

Idle Mode

URA_PCH

CELL_DCH CELL_FACH

CELL_PCH

Cell selectionCell re-selection

Dedicated resourcesallocated (DCH)Tx and Rx mode

Common resourcesallocated (RACH-FACH)

Tx and Rx mode

UE in DRX modediscontinous receiption

UE in DRX modediscontinous reception

RRC State Machine

Cell Update, UL Tx

UL/DL activation timer

Traffic volumeRACH load

Inactivity TimerOverload

UL Tx

Frequent cell updates



Power Control



UL Outer LoopPower Control

Open Loop Power Control(Initial Access)

(Fast) Closed Loop Power Control

RNC

BS

MS

DL Outer LoopPower Control

Power Control types

BLER target



• Purpose: To set the initial transmitted power of PRACH & DPCCH in the UL.• UE determines the uplink preamble power of PRACH

• UE PRACH First Preamble Power =

Transmission power of CPICH (Broadcast on BCH, SIB 5)) - Downlink RSCP measurement from active cell on CPICH (Measured by UE) +Total received wideband interference power at WCDMA BTS (Broadcast on BCH, SIB 7) +Required received C/I at the WCDMA BTS (Broadcast on BCH, SIB 5)

• Open loop PC is a part of the random access procedure for PRACH channel

Path loss calculationsPath loss

calculations

Additional Power

required

Additional Power

required

Example:PtxCPICH=33dBm (Parameter per Node-B)

DL RSCP = -80dBm (Measured by UE)UL_IF = –100 dBm

UL_Required_C/I = -25 dB (Parameter per Node-B)

UE PRACH First Preamble Power = 33 dBm – (-80 dBm) + (-100 dBm) + (-25 dB) = -12 dBm

Open Loop Power Control



tp-mtp-p

Message part

PRACHaccessslots TX atUE

One access slot

tp-a

Acq.Ind.

AICHaccessslots RX atUE

P0

Pp-m

Pre-amblePre-

amble

DL

UL

Random Access Procedure



• UE transmits the first preamble with the power determined by UL open loop PC• If the UE does not detect any acquisition indicator in AICH, it increases the preamble Tx

power by a specified offset Po

• If the UE detects the positive indicator in AICH, it transmits the random access message, 3 or 4 access slots after the UL access slot of the last transmitted preamble

• The Tx power of the control part of random access message should be Pp-m higher than the last transmitted preamble power

• The required power offset values for random access procedure• PowerOffsetLastPreamblePRACHmessage in PRACH(Pp-m)

• PowerRampStepPRACHpreamble (Power Ramp Step)(Po)

Random Access Procedure (1/3)



• The power ramp-up process will continue until 1) A positive AI is received from the network

2) A negative AI is received from the network

3) RACH_preamble_retrans value is exhausted

4) UE reaches UEtxPowerMaxPRACH value

• When the RACH_preamble_retrans value is exhausted, PRACH preamble power will be re-set to the initial value of the cycle and a new power ramp-up cycle initiated. The preamble power ramp-up cycle will be repeated RACH_tx_Max times. At this stage the UE will send a RACH failure message to the network.

• The maximum allowed UE transmit power for the PRACH procedure is defined by UEtxPowerMaxPRACH. Layer 1 of the UE controls the UE transmit power during the PRACH procedure using the ‘commanded transmit power’. If the commanded transmit power exceeds the maximum allowed transmit power then the UE transmits the maximum allowed transmit power.

• If the commanded transmit power exceeds the maximum allowed transmit power by 6 dB then layer 1 of the UE is able to inform higher layers and exit the PRACH procedure. If the step size is 1 dB then this corresponds to transmitting 6 preambles at maximum power.




Downlink / BSDownlink / BS

Uplink / UEUplink / UEPreamble 1 Message part

…. ….

UEtxPo we rMaxPRACH

Preamble n

PRACH_ p re am ble _ re tra ns : The maximum number of

preambles allowed in one preamble ramping cycle

RACH_ tx _ Max : # of preamble power ramping cycles that can be done before RACH transmission

failure is reported,




• PowerOffsetLastPreamblePRACHmessage in PRACH(Pp-m)

• The power offset between the last transmitted preamble and the control part of the PRACH message (added to the preamble power to receive the power of the message control part)

• range: -5 ... 10 dB, step 1 dB default: 2 dB

• The power offset between last preamble and and message part should ensure decoding the RACH message at BS with high probability. Still, it should be mimised to reduce UL interference

• PowerRampStepPRACHpreamble (Power Ramp Step)(Po)

• The power ramp step on PRACH preamble when no acquisition indicator (AI) is detected by the UE

• range: 1 ... 8 dB, step 1 dB default: 2 dB

• If the "power ramp step" is too low then the RACH preamble ramping takes a too long time. If it is too high, then it may cause high noise rise at BS

Random Access Procedure Parameters



• RACH_preamble_retrans

• The parameter describes the number of PRACH preamble retransmissions in a preamble power ramping-up cycle

• range: -1 ... 64, step 1 default: 8

• Note: As the 3GPP requires almost certain detection of PRACH preamble at -19.5dB. The default of 8 retransmissions with 2dB power step, starting from -25dB should be sufficient

• RACH_tx_Max• Maximum number of RACH preamble cycles defines how many times the PRACH pre-amble

power ramping-up procedure can be repeated before UE MAC reports a failure on RACH transmission to higher layers

• range: 1,2...32 default: 8

Random Access Procedure Parameters



Outer Loop Power Control



4

BS RNC


• Outer PC loop is performed to adjust the TARGET SIR in BS/UE, according to the needs of individual radio link:

• UE speed• Changes in the propagation conditions• Available multipath diversity• UE power control dynamics (close to peak power)• SHO branches (Macro Diversity Combining)

• SIR is constantly adjusted in order to maintain a constant QUALITY, usually defined as a certain target BLER




UL OuterLoop PC

Entity #N

UL OuterLoop PC

Entity #1

UL Outer Loop PCController

RNC

BTS 1

UL Fast Closed

Loop PC

BTS 2

UL Fast Closed

Loop PC

UL Outer Loop PC

Uplink Outer Loop Power Control Entities

• In the RNC the functionality of the UL outer loop PC is divided into two parts:

⇒UL outer loop PC Controller, one for each RRC connection

⇒UL outer loop PC Entities, one for each transport channel



• An UL Outer Loop PC Controller controls all UL OLPC Entities under the same RRC connection.

• The UL OLPC Controller sets the parameters for each UL OL PC Entities at the RAB Setup/Modification.

• The UL OLPC Controller also combines SIR Target changes from the UL OLPC Entities and sends the result to the UL OLPC Entity, which is selected to transmit it to the WCDMA BTS.

UL OuterLoop PC

Entity #N

UL OuterLoop PC

Entity #1

UL Outer Loop PCController

RNC

BTS 1

UL Fast Closed

Loop PC

BTS 2

UL Fast Closed

Loop PC

UL Outer Loop PC

Uplink Outer Loop Power Control Controller

• There is one UL outer loop PC Entity for each transport channel in the RNC.

• This UL OLPC Entity calculates the required change in SIR Target according to UL quality estimates (CRC).

• One of UL OLPC Entities under the same radio link is selected to transmit the New SIR Target to the WCDMA BTS.



1. RAB Setup:Initial SIR Target

UL Outer Loop PC Entity #n

UL Outer Loop PC Controller

2. - PC Parameters - Initial SIR target

8. Collection of the SIR target changes and calculation of new SIR Target3. Setting of the UL Outer Loop PC Entities

2. PC Parameters at RAB setup

9. New SIR Target

BTS1. SIR Target

1. UL fast closed loop PC

Admission Control

- Entity selected to transmit new SIR target- Activity reporting period

4. - PC parameters

6. Calculation of SIR Target change

Overload info Load Control

(not used in RAN1)

4. Parameters 7. SIR Target modification command

5. Quality info: BER, BLER

5. L1 FP: UL quality info

10. New SIR Target

10. L1 FP: SIR Target

10. Transmission of new SIR Target value to MDC

MDC

Uplink Outer Loop Power Control Algorithm



• This function is implemented in the UE in order to set the SIR target on each CCTrCH used for the DL closed loop PC.

• This SIR value is adjusted according to an autonomous function of the UE in order to achieve the same measured quality as the quality target set by the RNC.

• In order to control the downlink outer loop PC quality target in UE, Admission Control (AC) determines the value of the DL BLER target for each DCH mapped on a DPCH.

• After Admission Control functionality has determined the DL BLER target for each transport channel, the RNC sends these values to the UE.

Downlink Outer Loop Power Control



Close Loop Power Control



UL Outer LoopPower Control

(Fast) Closed Loop Power Control

RNC

BS

MS

DL Outer LoopPower Control

Closed Loop Power Control

BLER target



MS sets the power on UL DPCCHand UL DPDCH on following way:TPC = '1' --> increase power by 1 dBTPC = '0' --> decrease power by 1 dB

UL DPCCH

MS

Measure received SIR on UL DPCCH Pilot

Compare measured SIR withSIR target value received from

UL outer loop PC

Measured SIR < SIR target --> TPC bit = '1'Measured SIR => SIR target --> TPC bit = '0'

BS

Send TPC bit on DL DPCCH

Changed power on UL DPCCH

UL Closed loop power control

• UL fast closed loop PC shall be active as soon as the frame synchronization has been established in the dedicated physical channels.

• PC frequency 1500 Hz• PC step 1dB• PC delay approx. one slot



DL Fast Loop PC: UTRAN behaviourMS

Measured SIR < SIR target --> TPC command is "1"Measured SIR => SIR target --> TPC command is "0"

Compare measured SIR with SIR target value received from DL outer loop PC

Measure received SIR on DL DPCCH

WCDMA BTS

BS sets the power on DL DPCCH andDL DPDCH following way:

TPC command = "1" --> increase power by 1 dBTPC command = "0" --> decrease power by 1 dB

DL DPCCH + DPDCHs

Send TPC command on UL DPCCH

Changed power on DL DPCCH + DPDCHs

• Upon receiving the TPC commands BS adjusts its downlink DPCCH/DPDCH power accordingly.

• UTRAN shall estimate the transmitted TPC command TPCest

to be 0 or 1, and shall update the power every slot.

• After estimating the k:th TPC command, UTRAN shall adjust the current downlink power P(k-1) [dB] to a new power P(k) [dB] according to the following formula:

P(k) = P(k - 1) + PTPC(k)

where PTPC(k) is the k:th

power adjustment due to the inner loop power control

DownlinkInnerLoopPCStepSize



Resource Manager



Resource Manager

• The main function of RM is to allocate logical radio resources of BS according to the channel request by the RRC layer for each radio connection

• The RM is located in the RNC and it works in close co-operation with AC and PS

• The actual input for resource allocation comes from AC/PS and RM informs the PS about the resource situation

• The RM is able to switch codes and code types for different reasons such as soft handover and defragmentation of code tree

• Manages the BS logical resources• BS reports the available logical HW resources

• Requests for other resources such as ATM• Transport resource manager

• RNC HW manager (L1/L2)

• Maintains the code tree• Allocates the DL spreading (=channelization) codes, UL scrambling code, UL spreading (=channelisation) code type

• Prevents fragmentation, may cause extra HO's



Spreading Code Allocation

The codes are layered from 0 to 11 according to the Spreading Factor (SF)

• Cm(n) : The code order, m, and the code number, n, designates each and every code in the layered orthogonal code sequences

• In DL code order 2 to 8 (SF 4 to 512) are available (Nokia RAN does not support SF = 512)

• In UL code order 2 to 7 (SF 4 to 256) are available

C 0 (0)=(1)

C 1 (0)=(1,1)

C1 (1)=(1,-1)

C 2 (0)=(1,1,1,1)

C 2 (1)=(1,1,-1,-1)

C 2 (2)=(1,-1,1,-1)

C 2 (3)=(1,-1,-1,1)

C 3 (0)=(…)

C 3 (1)=(…)

C 3 (2)=(…)

C 3 (3)=(…)

C 3 (4)=(…)

C 3 (5)=(…)

C 3 (6)=(…)

C 3 (7)=(…)

Code Order 0 (SF 1)

Code Order 1 (SF 2)

Code Order 2 (SF 4)

Code Order 3 (SF 8)

• Code Allocation Algorithm chooses the correct spreading code depending on the TFC type



Spreading Code Allocation

• A code is always allocated from the optimum location in the code tree. It makes the allocated code and the codes in the branches below and above the allocated code unavailable

• Code tree will fragment quickly if releases is not re-arranged

• Re-arrangements in the code tree is done by reallocating the codes in better locations

• The above code tree has 4 codes of equal order. The best locations are in the same branch and very close to one another. The badly located codes are released and optimally reallocated allowing the use of upper layer codes

• Codes are only reallocated when there is a benefit at two code tree layers above the code being reallocated

7.rrm overview

Technology

Transcript of 7.rrm overview