IPA Light call management Introduction

IPA light call management architecture

Ding Junhua
2009/09/14

For internal use

1
Nokia Siemens Networks

Presentation / Author / Date

Contents
Background knowledge
Call Management Concept
Call Basic Concept
Call Management SW architecture
Call Resource Handling
User Plane resource Management
Call example
Q&A

For internal use

2
Nokia Siemens Networks

Presentation / Author / Date

Background knowledge

For internal use

3
Nokia Siemens Networks

Presentation / Author / Date

Wireless communication evolution

Frequency 1
Frequency 2
Frequency 3
Frequency Division Multiple Access (FDMA)

M
S
1
100

101
0

M
S
2

M
S
3

M
S
1

M
S
2

M
S
3

M
S
1

M
S
2

M
S
3

M
S
1

M
S
2

M
S
3

M
S
1

M
S
2

Time Division Multiple Access (TDMA)

100
1001010 1010
100101
10010
0100010101110
1010101
101
0010001
110
01011110
1001010
00
1
1010010
1001010
010010000100100
0
0
1
10100101010010101000101011101010101001000101011110101001001010001001000010
0
0
1
0
1
0
1
0
0
1001010100101010001010111010101010010001010111101010010010100010010000100
1001001 011000000110000100001
01010100 00101011101010 01010
00
1010
101001001 00010101111010
01
0010
10 00
0000011
100110011101010101 00011001111010011101010101
1001010
1110
10010101
0001
0010010
1010
1001010100101010001010111010101010010001010111101010010010100010010000100
0 11
0010101
1
1
0
0
1
000100
0
0
1
1
1
0
1
0
0
10
1
1
0
10 0010
101
0
1
0
1
0
0101
1001010100101010001010111010101010010001010111101010010010100010010
001
0000100
011
0111010
01010001001
10
00101
00
0
1
0
1
11
1
0
0
11
0 101 1 0
10001010
0101
001010
01001
110101010100
01
1
1
0
10
00
0100011
1
0
0
01
0
1
00
Code
Division
Multiple
Access
(
CMDA)
1
01
1
000001
10010
100
000010000
100 100

For internal use

4
Nokia Siemens Networks

Presentation / Author / Date

UMTS Network Architecture

GSM/EDGE Access Network
BSC/PCU

Circuit Switched Core Network

3G MSC
GMSC

External Networks

GSM/MAP
Core

PSTN

MGW

SMS
WAP

HLR

NMS

Packet Switched
Core Network
IP
Backbone

Iu
Interfaces

3G SGSN
UMTS Radio Access Network
For internal use
5
Nokia Siemens Networks

Presentation / Author / Date

GGSN
Firewall

Internet

How to make a call in UMTS network

BTS service is ok.
RNC>MSC common connection is ok, RNC>BTS common connection is ok.(signalling
level)
CELL has been setup. (including common channel has been setup and SI broadcasting is
ok).
UE Power on
Cell selection/reselection: select frequency.
UE reads random access parameters from BTS and initiate the Radom access attempt.
System Registration (authentication)
Enter IDLE state
UE make a call
RRC connection setup. (SRB setup,radio resource setup)
NAS procedure (CR, CC) to setup the SCCP connection between CN>RNC.
RAB setup
Iu data path setup.
Radio resource setup.
Iub data path setup (including FP sync).
Call connect.

For internal use

6
Nokia Siemens Networks

Presentation / Author / Date

Call setup

For internal use

7
Nokia Siemens Networks

Presentation / Author / Date

UTRAN
UTRAN handles the all
radio-related
functionality.

Core Network
Iu

Iu

UTRAN
Iur

RNC
Iub
Node B

RNC

Iub

Iub
Node B

Node B

Iub
Node B

Uu (the radio interface)

UTRAN: UMTS Terrestrial Radio Access
Network
RNC:
Radio Network Controller
UE:
User Equipment

For internal use

8
Nokia Siemens Networks

Presentation / Author / Date

UE

IP based UTRAN Overview

For internal use

9
Nokia Siemens Networks

Presentation / Author / Date

Model

UTRAN OSI model

Control plan

CTRL
L3

3GPP TS 25.201 V3.1.0

3GPP TS 25.201 V3.1.0

User plane

User plane

Control plane

USER
DATA

USER
DATA

CTRL

RRC

RRC

Signaling
Radio Bearer

L3

Radio Bearer
L2

RLC

RLC

L2

MAC

Logical channels

- grouped by information content

- User Data
- Control and signaling

Transport channels

RLC

RLC

L2

MAC

L2

PHY

L1

- grouped by method of transport

PHY

L1

Physical channels

Physical Channels Distinguished by:

- RF Frequency
- Channelization Code
- Spreading Code
- Modulation (I/Q) Phase (uplink)
- Timeslot (TDD mode)

UE

For internal use

10
Nokia Siemens Networks

Presentation / Author / Date

UTRAN

UTRAN Elements and Interfaces

UTRAN

WCDM
A BTS

Uu

User Equipment
(UE)

RNC
WCDM
A BTS

Iu
b

Iur

WCDM
A BTS

RNC
WCDM
A BTS

For internal use

11
Nokia Siemens Networks

Presentation / Author / Date

Iu

Core
Network
(CN)

RNC (radio network controller)

The RNC (Radio Network Controller) is the

network element responsible for the control of

the radio resources of UTRAN.
It interfaces the CN (normally to one MSC and
one SGSN) and also terminates the RRC
(Radio Resource Control) protocol.
RNC role:
CRNC. The RNC controlling one Node B,
it is the role for Node B.
Serving RNC. The SRNC for one mobile
is the RNC that terminates both the Iu
link for the transport of user data and the
corresponding RANAP signalling to/from
the core network.
Drift RNC. The DRNC is any RNC, other
than the SRNC, that controls cells used
by the mobile.

For internal use

12
Nokia Siemens Networks

Presentation / Author / Date

RNC functionality
Management of terrestrial channels

Allocation of traffic channels in Iu and Iub interfaces

Management of radio channel configurations in RAN
Radio resource management
Admission Control
Resource Manager
Packet Scheduler
Load Control
Power Control
Hand-over control
Handovers

Controlled by RNC
Can be initiated also by MS
Different types of handover: soft, softer, hard handover with intra & inter RNC handovers

Ciphering
Maintenance
Fault localisation

Reconfiguration of RNC and reconfiguration support for BTS

Software updates in RNC and BTS
Operation
Modification of parameters of RNC and BTS
Modification of the radio access network
Configuration of RNC HW
Administration of RNC equipment

For internal use

13
Nokia Siemens Networks

Presentation / Author / Date

mcRNC Network architecture

Node B:s

Uu

OMS

mcRNC

BTS O&M

Iu-CS

IP

IP

Uu
Iu-PS

Iub

IP

mcRNC
For internal use
14
Nokia Siemens Networks

Presentation / Author / Date

Iub
Iur
Iu-CS
Iu-PS
etc

Iu-BC

ADA Network architecture

MSC-S+MGW+NVS
WCDMA BTS

RNC

Iub

GGSN

SGSN

IMS:

IM, Presence, PoC

Video sharing
VoIP, IP Centrex

Content &
Connectivity

Iu

Internet+
Intranets

Iur

Iu

ADA

Iu
Iu
Iu

Iur

ADA
HSPA
device
For internal use
15
Nokia Siemens Networks

Presentation / Author / Date

Standardized
Standardized with vendor
specific extensions

CM functionality for RNC

The radio resource management is managed by RRM,RRC,NBAP in application.
Call management is responsible for data path setup/modify/release of Iu, Iur, Iub.
Following is example of AMR call data path:

For internal use

16
Nokia Siemens Networks

Presentation / Author / Date

Call Management Concept

For internal use

17
Nokia Siemens Networks

Presentation / Author / Date

What is call management

CM (Call Management) is part of IPA Light SW which provides an interface to switching resources

and user plane resources to L3 application. In IPA Light, CM only provide services to mcRNC and
ADA application software.
As ADA requirement is not clear currently, so this introduction will focus on mcRNC part.
CM provides a logical resource model, which gives to the clients a simplified view to the physical
resources.
Call Management deals with call which maps to the user service, following services are supported:
Real Time speech
Signaling link data
Real Time data
Non Real Time data
RAB in DRNC
HSPA data

For internal use

18
Nokia Siemens Networks

Presentation / Author / Date

CM functionality
In IPA Light, CM provided following functionalities:
Call Management. (Such as Call Create, Call Release)
Generic Connection Resource Management (Such as IP
resource)
Generic Service Management (Such as cell service)
Adaptation of different resources (IP connection, VCP
connection)
User Plane Resource Management
Unit management (Such as USUP restart handling)
Owner ID management.
Test Call support (leg troubleshooting).

For internal use

19
Nokia Siemens Networks

Presentation / Author / Date

Call Resources in mcRNC

Iu,Iur

EIPU

(IP)

USPU

Iub

Internal
Processor
Switch
EIPU

(IP)

CSPU

For internal use

20
Nokia Siemens Networks

Presentation / Author / Date

Call Resource in mcRNC

External transport resources.
In mcRNC, only IP network is supported, so the external resource are :
IP Iub (FP over IP + UDP)
IP Iur (FP over IP + UDP)
IP IuCS(RTP/RTCP over IP + UDP)
IP IuPS(data over IP + UDP +GTPU)
Internal transport resources
VCP is used to manage the internal unit connection.
User Plane resources
In mcRNC, CSUP provide the UP service for common user (CELL, CCH) and USUP provide
the UP service for dedicated user (SRB, RAB)

For internal use

21
Nokia Siemens Networks

Presentation / Author / Date

Call Basic concept

For internal use

22
Nokia Siemens Networks

Presentation / Author / Date

What is call
Call means all the resources between physical endpoints at the external interfaces.
Call is used for specific 3G service, such as SRB, common channel, RAB.
Call can be two-sided or multisided and it consists of legs and a leg connector and can contain up
to16 legs.
The logical module of two-sided call:

For internal use

23
Nokia Siemens Networks

Presentation / Author / Date

Relations and concepts

The CM handles the resources of a call (e.g. legs), but the

resources seen in application layer are the logical

resources.
The call can be two-sided call and multi-sided call, in RNC
only two-sided call is supported.
The through switch connection means all the connections
and recourses between physical endpoints at the external
interfaces of the NE. The through switch connection is
composed of legs and one leg connector. Typically, one
through switch connection (one call) contains two legs
one for incoming side of a call and an other one for
outgoing side of a call.
MDC (Macro Diversity Combining) service is logical
service which can be added to a logical leg.

For internal use

24
Nokia Siemens Networks

Presentation / Author / Date

Call Concept
.

1 call can have 0-32 connections, and 1 connection have 2 connection points which can be service or
leg.
1 call can have 0-15 legs, and 1 leg can contain 0-1 dsp service.
For internal use
25
Nokia Siemens Networks

Presentation / Author / Date

Call Create
.

For internal use

26
Nokia Siemens Networks

Presentation / Author / Date

Resource model : two-side call

up to 6 "sides" in a call (incoming, outgoing,

termination, pre-allocated IN and OUT, and

secondary IN(IN_SEC))
these "sides" can be branchable: multiple legs
are supported (up to 6 per side)
only IN and OUT can be connected together
(through connection)
pre-allocated legs are used for fast switching
situations
Detailed description of the logical leg type:
in_c normal incoming logical leg, which can be
used in through connection
out_c normal outgoing logical leg, which can
be used in through connection
in_pre_c pre-allocated incoming logical leg,
which is seized, but must be swapped before it
can be used in through connection
out_pre_c pre-allocated outgoing logical leg,
which is seized, but must be swapped before it
can be used in through connection
term_c the termination of a multisided call.
in_sec_c establish the HSDPA DL connection
For internal use
27
Nokia Siemens Networks

Presentation / Author / Date

What is leg
Leg is an abstraction of the real resources.
A leg is composed of services and sub connections between these services.
A leg has an unambiguous start point that is identified by TPI. These start points are used to

identify legs. Start points of legs are visible to the client, but the detailed structure of legs is hidden
from them.
A leg can contain 2 services max.
All information of leg is stored to leg packet.
In IPA Light, following legs are supported:
UDP leg
RTP leg
GTP leg

For internal use

28
Nokia Siemens Networks

Presentation / Author / Date

What is service
Services are divided in two category, leg's services and other services (cell etc.).
A service is realized in a functional unit and has one (terminating service) or two termination points.
A service is added to a leg by making subconnections to the termination points of the service.

The supported services in IPA Light are:

Macro Diversity Combining (MDC)
Other services: cell_c, hsdpa_c.

For internal use

29
Nokia Siemens Networks

Presentation / Author / Date

Connecting legs and services

Connection principle is flexible: any type of leg or service can be connected to any type of leg or
service
Legs and services have special connection points which can be either
termination points, OR
UP services
Connection is called leg connector
Leg connector can be VCP connection (e.g. if UP-services are in different units),or dynamic link
inside UPUP/CSUP.
connection
points

Leg_1

leg
connector

For internal use

30
Nokia Siemens Networks

Leg_2

Presentation / Author / Date

Call Management SW Architecture

For internal use

31
Nokia Siemens Networks

Presentation / Author / Date

SW Architecture

For internal use

32
Nokia Siemens Networks

Presentation / Author / Date

SW Architecture
(CRH)
CM Domain
Call Resource Handling (CRH)
User Plane resource Management (UPM)
CRH
The Generic Connection Resource Manager Program Block (RM2PRB) provides services for the call and

resource management. It provides a common interface for the Layer 3application programs to access
the switching resources of the system.RM2PRB uses adaptation instance (LEMANA) to allocate the
resources for the calls.
The Refresher of Hand's Time Supervision (RFRPRB) is a supervision program block. It provides service
for hand process supervision. Each call in the system has an own Refresh Group, which consists of hand
processes belonging to the same call. RM2PRB starts this Refresh Group, when the call is requested. All
Layer 3applications must join to this Refresh Group, so that they can be supervised as well.
The Distributed Leg Control Program Block (LEMANA) is adaptation instance, which provides selection of
the IP-based resources (GTP or UDP legs). LEMANA also provides connection of given resources (e.g.
connect two legs together).
The Resource Unit Handling Program Block (UHAPRB) takes care of the owner id handling, restart
handling, unit state change handling
The Resource Management Utilities Program Block (RMUTIL) takes care of simulating real application for
RM2PRB.
The Leg Utilities Program Block (LGUTIL) is the service terminal counterpart for RMUTIL. LGUTIL
commands RMUTIL and also inquires data from RM2PRB.

For internal use

33
Nokia Siemens Networks

Presentation / Author / Date

SW Architecture
(UPM)
UPM
The Local User Plane Resource Manager Program Block (LRMPRB) takes care of local user

plane reservation, modification, release. LRMPRB is located in the Linux node of each
processing unit, which contains a SE node for user plane processing. LRMPRB receives user
plane resource allocation requests from LEMANA, and it always manage the logical user plane
resources only from the SE node located to the same processing unit with LRM itself.
The centralized User Plane Resource Manager (RM3PBR) takes care of USUP/CSUP resource
state management, and provide Diagnostics interface to UP resources. It keeps record on the
user plane resource situation of each user plane processing unit on the basis of the information
received from LRMPRB.
The user plane Utilities Program Block (UMUTIL) is the service terminal provided to configure
and check the RM3PRB resources.

For internal use

34
Nokia Siemens Networks

Presentation / Author / Date

Call Resource Handling

For internal use

35
Nokia Siemens Networks

Presentation / Author / Date

Functionalities

Call Resource Handling (CRH) is a domain in IPA Light, which contains the needed software to
setup (release) switching resources for calls.
Following functionalities are provided by CRH:
Call Management Services
Generic Connection Resource Management
Generic Service Management
Adaptation
Resource supervision services

For internal use

36
Nokia Siemens Networks

Presentation / Author / Date

Call management services

Call Management handles the request related to one call. These services are used to start new

calls, supervise ongoing calls and release calls.

This functionality provides a simplified call model for the clients, which helps to keep track of the
resources.
The Call Management provides following functionalities:
Call identification provides for each call a unambiguous call identifier (call_id). The Call
Management functionality allocates an new call id for each new call. The identifier is released
after the call has ended.
Call setup allows clients to start new calls: the call gets a call id and the call supervision is
started.
Call release functionality releases the call, when the last resource of that call is released. Call
Management provides also mechanism to release the whole call (e.g. in error situations).
Call inquiries allows clients to examine the state of calls.
Call supervision ensures that the ongoing calls are working properly. It also supervises the
client hand process related to the call. The call supervision releases call, when it detects
problems with it.

For internal use

37
Nokia Siemens Networks

Presentation / Author / Date

Generic Connection Resource Management

services
The Generic Connection Resource Management services

provides an interface to the switching resources for the

client.
The basic resource is logical leg. The logical leg is
identified by Physical Resource identifier (PRid). The
following services provides seizuring and releasing of
logical legs, connecting and disconnecting the logical legs,
and several special services (e.g. adoption and swapping).
rm_resource_req_as, rm_resource_release_req_as,
rm_resource_modification_req_as,
rm_resource_swap_req_as, rm_resource_inquiry_req_as
are CM provided services to the client, and following is the
description:
Resource seizure allocates a new switching resource.
Resource release handles releasing of the seized
resources.
Resource connections is used to connect two Logical
Legs together to form a Through Connection.
Resource swapping provides fast switching when the
client wants to rapidly changes the resource
connections with minimum break in through
connection.
Resource inquiries allows clients to examine the state
of resource in a call.
Resource supervision ensures that the resources are
eventually released, even in fault situations.

For internal use

38
Nokia Siemens Networks

Presentation / Author / Date

Connecting legs and services

The logical model of resources has a special connection point, which is used when

connecting resources. This connection point can be either:

UDP, the leg has an IP address and UDP port allocated when it was created. This IP
address and UDP port must be used when the leg is connected to other resources.
GTP, the leg has a GTP tunnel ID allocated when it was created.The IP address and GTP
tunnel ID must be used when the leg is connected to other resources.
Service, the leg/service has a special SPRM-service, which is used for the
connection. This service can be connected inside the same unit to another
service, or it can be connected to termination point (which is part of vcp
connection to other unit).
There are four different ways to connect two resources:
connection inside UP-unit, direct connect.
connection inside EIPU-unit, direct connect.
connection with DMXMSG connection, take care by UP itself, not visible by CM.
connection with VCP connection.

For internal use

39
Nokia Siemens Networks

Presentation / Author / Date

Generic Service Management services

The logical service is an abstraction of a

physical service. The abstract service has

following characteristics:
type: the type of service (CELL
/HSDPA/ MDC)
res_char: the resource parameters
(e.g. bitrates and packetsizes)
conn_points: one service can have one
or two connection points. They are
named 'in' and 'out

The Generic Service Management allows

following requests to the resources:

Service allocation reserves needed
resource for the service.
Service release handles releasing of
the service's resources.
Service modification allows the client
to modify the characteristics of the
allocated resource.
Service connection handles the
connections to other resource.

For internal use

40
Nokia Siemens Networks

Presentation / Author / Date

Adaptation services

The Adaptation services provide adaptation of

different hardware technologies. All resources
in different technologies are abstracted to a
common logical leg-model.
The Adaptation services implements these
logical concepts. The implementation is
different in different technologies. Only
Adaption of IP is supported in IPA Light.
In IPA Light the only adaptation instance in
LEMANA. LEMANA is light weight TNSDL
family.

For internal use

41
Nokia Siemens Networks

Presentation / Author / Date

UDP leg example

For internal use

42
Nokia Siemens Networks

Presentation / Author / Date

Test Call Support

Test call functionality enables better fault traceability in our own labs as well as in operators
network environments. Test call can be used in situations like:
When there is detected some problems for example in certain type of calls.
Before adding/replacing new processing units to the mcRNC, they can be first tested to detect data path
quality inside mcRNC network element.

There are two types of test calls: Originated by the application (RNC/MGW) or originated by the
LGUTIL service terminal command. The LGUTIL originated calls can be called also as internal test
calls.
Test calls are presented in the test_call test_call_ind_t; of the message from application.

For internal use

43
Nokia Siemens Networks

Presentation / Author / Date

Resource supervision services

The resource supervision services provide release procedures for error cases like unit restarts and
state changes. Resource supervision also provides mechanism to release hanging resources from
the system.

For internal use

44
Nokia Siemens Networks

Presentation / Author / Date

Owner ID concept

Owner ID is allocated for each USCP/CSCP after startup, and the value will changed continuously.
When call is setup, ownerID is also transferred to resource owners which is related with this call.
Change Owner ID has release the hanging resource which happened in CP side.
Release Owner ID can release the hanging resource which happened in transport, and UP side.

For internal use

45
Nokia Siemens Networks

Presentation / Author / Date

Owner ID change and release

The resources of RU internal, such as between RM2,LEMANA and TAC will not be considered in this

release.
For traffic and transport resource only rm2 call has leg, this resource supervision is needed, for
standalone GTP tunnel, it is not in the own ID management concept.
Own ID only managed in RM2, LRM and TAE which represent CP, UP and TT resources.
When receive own ID change, rm2 will loop the leg in the call, and get the start point, and send
message to TAE to change the owner ID.
When release the old owner ID, UHAPRB will broadcast the release message to LRM and TAE, then
to release LRM will release the UP resource and TAE will release the traffic resource.
Impact to current design:
CM will configure the TAE if there is leg created even it is DRNC leg.
TAE will create the table for one side routing.
Change owner ID will not go through the LEMANA, RM2 will directly contact TAE, LRM
Release owner ID will not go through the LEMANA also.

For internal use

46
Nokia Siemens Networks

Presentation / Author / Date

UP restart

For internal use

47
Nokia Siemens Networks

Presentation / Author / Date

CP restart

For internal use

48
Nokia Siemens Networks

Presentation / Author / Date

UP Resource Management

For internal use

49
Nokia Siemens Networks

Presentation / Author / Date

User plane resource manager

The user plane management functionalities are for user plane services management, such as

USUP service start/modify/stop, USUP selection, USUP failure indication and the real user plane
functionalities are data channels of voice announcements, voice packets, common channels, and
dedicated channels.
The whole UP resource management feature consists of 4 entities Physical UP resource
manager (RMDPRB), local UP resource manager( LRMPRB) ,centralized logical UP resource
manager (RM3PRB) and SPRM user interface (UMUTIL).
The centralized DSP Resource Manager (RM3PBR) and local UP resource manager (LRMPRB)
take care of the logical DSP resource reservation during the leg creation. The RMDPRB takes care
of the physical user plane resource allocation during the leg creation.
The managed resources of UP are:

For internal use

50
Nokia Siemens Networks

Presentation / Author / Date

Allocating Resources

For internal use

51
Nokia Siemens Networks

Presentation / Author / Date

Work principle
CSCP only allocate the UP resource from CSUP in own unit.
The resource selected for the control applications are always from the USUP which locates at the

same processing unit unless there is no enough resource in own processing unit.
The user plane resource management is transparently to the UP application.
LRM only manage the UP resources of own processing unit and RM3 manage all USUP resource
state.
Interface between LRM and RM3 are:
Allocating: RM3PRB provide service to LRMPRB to select user plane resources from other
processing unit also LRMPRB provide interface to RM3PRB to allocating user plane resource of
same unit.
Notify: RM3PRB provide interface to LRMPRB to notify the state change of user plane resources.
For each unit, there are 2 thresholds (a: full b: free) to manage the own unit user plane resource
status.
When own unit resources is > a and current status is free, LRMPRB will notify RM3PRB to
change the own unit status to full
When own unit resources is < b and current status is full, LRMPRB will notify RM3PRB to change
the own unit status to free
Normally a is >= b and both value can be configured with UMUTIL.
List: LRMPRB provide interface to RM3PRB to list the detailed resource information of own unit.
Release: When USUP/CSUP restart, LRMPRB provide interface to RM3PRB to release the
resource related with the failure USUP/CSUP (owner ID release).

For internal use

52
Nokia Siemens Networks

Presentation / Author / Date

Call example

For internal use

53
Nokia Siemens Networks

Presentation / Author / Date

SRB concept

SRB is signaling radio bearer which is responsible for signaling data transfer in RNC, it is RRC
message.
SRB is setup during RRC connection setup.
The RRC data is over RLC/MAC/FP/UDP/IP.

For internal use

54
Nokia Siemens Networks

Presentation / Author / Date

SRB setup on DCH Data Path

For internal use

55
Nokia Siemens Networks

Presentation / Author / Date

SRB setup on DCH message flow

For internal use

56
Nokia Siemens Networks

Presentation / Author / Date

SRB setup on FACH Data Path

For internal use

57
Nokia Siemens Networks

Presentation / Author / Date

SRB setup on FACH message flow

For internal use

58
Nokia Siemens Networks

Presentation / Author / Date

UE state change

UE state means UE RRC connection state which is the signaling connection between UE<-->RNC.
UE RRC connection state also means the RRC message transport over which channel.
When RRC connection is on DCH, there is dedicated IUB link for signaling data.
When RRC connection is on FACH, the signaling data is transferred on common channel.
RRC connection can be released only when RRC is on FACH or DCH state. so if RRC connection is
on PCH we must first change the rrc state to FACH or DCH then to release this RRC connection.
Possible state changes and affect to cm:
IDLEDCH, IDLEFACH
-- SRB setup needed.
PCHDCH,PCHFACH
-- SRB setup needed.
FACHIDLE,FACHPCH.
-- SRB release needed
FACHDCH
-- SRB reconfiguration needed,1 leg created and connect the MDC service.
DCHIDLE,DCHPCH
-- SRB release needed.
DCHFACH
-- SRB reconfiguration needed, leg will be disconnect from MDC service then released.

For internal use

59
Nokia Siemens Networks

Presentation / Author / Date

CS call concept

CS call is the basic service for user, it is for UE to make to voice call and cs data service.
AMR call is type of CS call which use the IuCS interface.
CS call can over DCH or HSPA.

For internal use

60
Nokia Siemens Networks

Presentation / Author / Date

CS Call on DCH Data Path

For internal use

61
Nokia Siemens Networks

Presentation / Author / Date

Message Flow
Step 1:Iu resource setup

For internal use

62
Nokia Siemens Networks

Presentation / Author / Date

Step 2:Iub resource setup

PS Call concept

PS call is the data service for user, such as download, Browsing, internet access
It used IuPS interface which connect to SGSN.
PS call can over RACH/FACH,DCH or HSPA.

For internal use

63
Nokia Siemens Networks

Presentation / Author / Date

PS call setup on DCH Data Path

For internal use

64
Nokia Siemens Networks

Presentation / Author / Date

PS call setup on DCH message flow (inactive GTP)

Inactive GTP means that first GTP tunnel created without any call, then some data received to
triggered the real resource creation.
Iub resource creation is same as CS call setup on DCH.

For internal use

65
Nokia Siemens Networks

Presentation / Author / Date

PS setup on FACH/RACH data path

The Iu resource is same as the PS on DCH, just without the Iub resource setup which already
done when common channel setup.
The Iu and Iub connect is done by application itself.

For internal use

66
Nokia Siemens Networks

Presentation / Author / Date

PS on HSPA data path

The Iu resource is same as the PS on DCH, just there are two Iub resources dedicated for uplink
and downlink.
From call management point of view, 2 Iub leg are some especially the MAC protocol are different
and the handover concept is different.

For internal use

67
Nokia Siemens Networks

Presentation / Author / Date

handover

Soft Handover is one of the most interesting feature of the WCDMA operation
handover means transfer of user connection from one radio channel to other.
Hard Handover:

Hard handover means that all the old radio links in the UE are removed before the new radio links
are established. Hard handover can be seamless or non-seamless. Seamless hard handover
means that the handover is not perceptible to the user. In practice a handover that requires a
change of the carrier frequency (inter-frequency handover) is always performed as hard handover.
Soft Handover

Soft handover means that the radio links are added and removed in a way that the UE always
keeps at least one radio link to the UTRAN. Soft handover is performed by means of macro
diversity, which refers to the condition that several radio links are active at the same time. Normally
soft handover can be used when cells operated on the same frequency are changed.

In a typical UMTS system, the UE is in soft/softer handover around 50% of the time. One of the
very important requirements for the soft/softer handover is that the frames from different cells
should be within 50ms of each other or this would not work.
Softer handover

Softer handover is a special case of soft handover where the radio links that are added and
removed belong to the same Node B (i.e. the site of co-located base stations from which several
sector-cells are served. In softer handover, macro diversity with maximum ratio combining can be
performed in the Node B, whereas generally in soft handover on the downlink, macro diversity with
selection combining is applied.
For internal use
68
Nokia Siemens Networks

Presentation / Author / Date

Soft handover introduction

Soft handover is initiated from the RNC and the core network is not involved in this procedure
When UE is in soft handover, at least UE will occupy 2 radio links

For internal use

69
Nokia Siemens Networks

Presentation / Author / Date

Soft handover intra-RNC

For internal use

70
Nokia Siemens Networks

Presentation / Author / Date

Soft handover inter-RNC SRNC side

For internal use

71
Nokia Siemens Networks

Presentation / Author / Date

Soft handover inter-RNC DRNC side,same EIPU

For internal use

72
Nokia Siemens Networks

Presentation / Author / Date

hard handover
Radio access network (RAN) supports three types of HHO:

Intra-Frequency Hard Handover:

Intra-frequency hard handover is lossless for NRT radio bearer but it causes short disconnection of RT
radio bearer. Intra-frequency hard handover is required to ensure handover between cells controlled
by separate RNCs in situations when inter-RNC soft handover is not possible.

Inter-Frequency Hard Handover

Inter-frequency handover is lossless for NRT radio bearer but it causes short disconnection of RT radio
bearer. Inter-frequency handover is required to support multiple carrier frequencies in the radio
network.

Inter-System Hard Handover

RAN supports the following types of inter-system handover for Circuit switched (CS) services:
Handover from WCDMA to GSM
Handover from GSM to WCDMA

Impact to Call management

From call management point of view, the HHO for call management is:
Release the old leg.
Setup the new leg.

For internal use

73
Nokia Siemens Networks

Presentation / Author / Date

SRNC relocation introduction

The SRNS relocation is used for moving the SRNC functionality from one RNC to another RNC
closer to the User Equipment (UE) if the UE moves during the communication. Both the radio
access network (RAN) and the core network are involved.

Regarding the UEs mobility it is really the handovers that count. The SRNS relocation procedures
can be seen as a subset for handover procedures: there are handovers without SRNS relocation
but no SRNS relocations without handovers.

UE-not-involved relocation VS UE-involved relocation.

UE-not-involved: Iur interface exist, SHO
UE-involved: Iur interface not exist, HHO : 1.inter-frequncy 2.Iur not available 3.Iur not configured.

UE not involved SRNC relocation has not affect CM as for source SRNC, it is normal call release
procedure, and for dest SRNC, it is normal call setup procedure, we only need to care one thing
that is PS NRT relocation downlink data buffering.

For internal use

74
Nokia Siemens Networks

Presentation / Author / Date

SRNC relocation introduction

There are 2 kinds of SRNC relocation in 3GPP as following, in NSN implementation, we
only support the scenario 2, scenario 1 supported only in IHSPA.

For internal use

75
Nokia Siemens Networks

Presentation / Author / Date

SRNC relocation handling for SRB

SRNC:
Before relocation, there is 1 in Iur leg and 1 service for srb.
After relocation, the SRB leg will be removed.
DRNC
Before relocation, there are 2 legs for SRB, one 1 in leg, another is out leg and 1 connection
between in and out leg.
Relocation prepare phase: setup 1 standalone MDC service (in_pre leg setup).
Relocation commit: swap the service with the connect the in leg to MDC. release the iur leg.
After relocation, there is 1 leg, 1 service and 1 connection for SRB.
Relocation cancellation:
No action to SRNC.
Release the standalone service in DRNC.

For internal use

76
Nokia Siemens Networks

Presentation / Author / Date

SRNC relocation handling for CS RAB

SRNC:
Before relocation, there is 1 in Iur leg, 1 out Iu leg and 1 service for CS RAB.
After relocation, all resources will be removed.
DRNC
Before relocation, there are 2 legs for CS RAB, one 1 in leg, another is out leg and 1 connection
between in and out leg.
Relocation prepare phase: setup 1 iu leg which include 1 rtp leg, 1 MDC service and 1
connection between leg and service(out_pre leg setup).
Relocation commit: swap the out_pre leg with the out leg(Iur) . release the original Iur leg.
After relocation, there is 2 leg, 1 service and 2 connection for CS RAB.
Relocation cancellation:
No action to SRNC.
Release the out_pre leg in DRNC.

For internal use

77
Nokia Siemens Networks

Presentation / Author / Date

SRNC relocation handling for PS RAB

SRNC:
Before relocation, there is 1 in Iur leg, 1 out Iu leg and 1 service for PS RAB.
After relocation, all resources will be removed.
DRNC
Before relocation, there are 2 legs for PS RAB, one 1 in leg, another is out leg and 1

connection between in and out leg.

Relocation prepare phase: setup 1 iu leg which include 1 gtp leg, 1 MDC service and
1 connection between leg and service(out_pre leg setup).
Relocation commit: swap the out_pre leg with the out leg(Iur) which means
reconnect the in leg to MDC service and release the original Iur leg.
After relocation, there is 2 leg, 1 service and 2 connection for PS RAB.
Relocation cancellation:
No action to SRNC.
Release the out_pre leg in DRNC.

For internal use

78
Nokia Siemens Networks

Presentation / Author / Date

SRNC relocation of PS NRT message flow

For internal use

79
Nokia Siemens Networks

Presentation / Author / Date

PS DL Data buffering during relocation(1)

Before relocation, there is no Iub connection in source RNC.

There is one Iur link

For internal use

80
Nokia Siemens Networks

Presentation / Author / Date

PS DL Data buffering during relocation(2)

Prepare the resources in DRNC side, pre Iu leg (GTP tunnel created and USUP service in
created) is created in DRNC.
The data path is not changed in both source RNC and DRNC

For internal use

81
Nokia Siemens Networks

Presentation / Author / Date

PS DL Data buffering during relocation(3)

Before DRNC switch to SRNC, the preparation is needed in source RNC side.
The downlink data will pre-block by GTP forward tunnel (means the DL data from SGSN will not
goto TAE but directly go to Target RNC TEID 3.)
Forward tunnel is only from source to destination, it is not by-direction.
From target RNC point of view, the source RNC is SGSN, and as the GTP is buffered, so these data are
buffered in target RNC.
From uplink side, there is no change ,the packet is still go through the TAE in source RNC.
For internal use
82
Nokia Siemens Networks

Presentation / Author / Date

PS DL Data buffering during relocation(4)

The target RNC begins to act as SRNC.

Buffering data can be sent to UE now.
UL data can be sent to SGSN TEID. (If SGSN use the same TEID for original tunnel and new tunnel,

the UL data is not lost, else do not know whether SGSN can handle the new UL data, it is assumed
that the SGSN has already allocated the new SGSN TEID for the new tunnel).
As SGSN has not been switched to new tunnel, so the DL packet still go through the source RNC GTP.
For internal use
83
Nokia Siemens Networks

Presentation / Author / Date

PS DL Data buffering during relocation(5)

After SGSN change the routing, the DL data will not go through the source RNC.

For internal use

84
Nokia Siemens Networks

Presentation / Author / Date

PS DL Data buffering during relocation(6)

All resources in original SRNC will be released.

The Iur resource in target SRNC will be released.
For internal use
85
Nokia Siemens Networks

Presentation / Author / Date

HSPA serving cell change

As UE can only receive the

HSDPA traffic from one HS-DSCH
channel so the handover concept
is different for HSDPA users.
An example of inter-BTS SCC:
1.reconfigure the original BTS to
remove the UE from the HSDSCH.
2. reconfigure the target BTS to
add the UE from the HSDSCH.
3.setup the data path for downlink.
from cm pov, it is in_sec leg
branch add.
4.reconfigure the source BTS to
remove the UE from HS-DSCH.
5.reconfigure the target BTS to
add the UE to HS-DSCH.
6.reconfigure the UE to update the
active set to receive downlink
packet from new HS-DSCH.
7.remove the original downlink
data path, from cm pov, it is
original in_sec leg branch remove

For internal use

86
Nokia Siemens Networks

Presentation / Author / Date

UE

Trg Node B

Src Node B

RNC

UE in CELL_DCH state, NRT RB mapped to HS-DSCH,

RRC: MEASUREMENT REPORT (CPICH Ec/No,CPICH RSCP)

RNC initiates inter-BTS serving

HS-DSCH cell change

NBAP: RADIO LINK

RECONFIGURATION PREPARE
NBAP: RADIO LINK
RECONFIGURATION READY
NBAP: RADIO LINK RECONFIGURATION PREPARE
NBAP: RADIO LINK RECONFIGURATION READY

User Plane setup

NBAP: RADIO LINK

RECONFIGURATION COMMIT
(CFN)
NBAP: RADIO LINK RECONFIGURATION COMMIT (CFN)
RRC: RADIO BEARER RECONFIGURATION (CFN)

UE Starts Tx/Rx for

HS-DSCH in target
BTS and stops Tx/Rx
for HS-DSCH in source
BTS at the given
activation time

At a time CFN PreCFNSwitchTime

RNC stops the MAC -d flow to old
BTS and makes the capacity request
to new BTS and starts the the
MAC d flow.

User Plane release

RRC: RADIO BEARER RECONFIGURATION COMPLETE

Q&A

For internal use

87
Nokia Siemens Networks

Presentation / Author / Date

Q&A
1. ?
2. ?
3. ?

For internal use

88
Nokia Siemens Networks

Presentation / Author / Date

Thanks

For internal use

89
Nokia Siemens Networks

Presentation / Author / Date