Nortel BSC3000 userguide

GSM BSC 3000 and TCU 3000 Description - Technical1075A

Student GuideGuide release: 16.03Guide status: StandardDate: November, 2006

FOR TRAINING PURPOSES ONLY

411-1075A-001.1603

Copyright © 2006 Nortel Networks. All rights reserved.

The information contained in this document is the property of Nortel Networks. Except as specifically authorized in writing by Nortel Networks, the holder of this document shall not copy or otherwise reproduce, or modify, in whole or in part, this document or the information contained herein. The holder of this document shall protect the information contained herein from disclosure and dissemination to third parties and use the information solely for the training of authorized individuals.

THE INFORMATION PROVIDED HEREIN IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND. NORTEL NETWORKS DISCLAIMS ALL WARRANTIES, EITHER EXPRESSED OR IMPLIED, INCLUDING THE WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL NORTEL NETWORKS BE LIABLE FOR ANY DAMAGES WHATSOEVER, INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL, LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES, ARISING OUT OF YOUR USE OR RELIANCE ON THIS MATERIAL, EVEN IF NORTEL NETWORKS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.

Information subject to change without notice.Nortel, Nortel Networks, the Globemark device, and the Nortel Networks logo are trademarks of Nortel Networks.

NORTEL CONFIDENTIAL – FOR TRAINING PURPOSES ONLY

Visit us at: nortel.com/training

DescriptionThis course is a comprehensive technical description of the BSC3000 and TCU3000 products. This course applies to V16 release of the BSS.

Intended audienceThis course is designed for people who need to know the functions and architecture of the BSC3000 and TCU3000.

PrerequisitesThis course has the following prerequisites:

• 1061A: GSM GPRS System Overview - Technical

ObjectivesAfter completing this course, you will be able to:

• Describe the physical and functional architecture of the BSC 3000and TCU 3000,

• Describe module functions and interfaces,• Trace the signaling and traffic paths inside and outside the equipment.

Course introduction

Overview


ReferencesThe following documents provide additional information:

BSC / TCU 3000 Reference ManualNTP 411-9001-126


Contents

1. Introduction

2. BSC 3000 and TCU 3000 Presentation

3. BSC 3000 and TCU 3000 Architecture

4. Data Flow Exercises

5. BSC 3000 and TCU 3000 Operation

6. BSC 3000 and TCU 3000 Maintenance and Enhanced Exploitability

7. BSC 3000 and TCU 3000 Provisioning

8. Exercises Solutions


Publication History

New reference name (formerly PR4)Compliant with V15.1 BSS Release

July, 200515.01/EN

Compliant with V15.1.1(Preliminary) BSS Release

November, 200515.02/EN

Compliant with V15.1.1(Standard) BSS Release

May, 200615.03/EN

Compliant with V16 (Preliminary) BSS Release

June, 200616.01/EN

Compliant with V16 (Standard) BSS Release

October, 200616.02/EN

New templateNovember, 200616.03/EN

CommentsDateVersion


1-1


nortel.com/training

November, 2006411-1075A-001.1603

Section 1

Introduction

1-2

FOR TRAINING PURPOSES ONLYNovember, 20061-2 411-1075A-001.1603

About Knowledge Services

> Knowledge Services offers three programs to help you get the most out of your Nortel solutions.• Training with a focus on eLearning• Certification• Documentation

> Making the global transition to “e”• We are transitioning many of our programs so we can meet the

demands of the 21st century; including a new focus on eLearning, an industry-leading certification program, new opportunities to save, vehicles for electronic communication to keep you in the know, and more.

Knowledge Services programs help you speed your time to proficiency. Through our programs, you can:

• Save time and money on quality, comprehensive training with our new eLearning portfolio

• Build the foundation for skills needed to successfully achieve certification through our training programs

• Gain hands-on experience with Nortel Networks solutions through our advanced lab courses

• Demonstrate and validate your knowledge and hands-on skills by achieving certification through our industry-leading certification program

1-3


Nortel Homepage

www.nortel.com

1-4


Training & Certification Page

www.nortel.com• Select Training

• Select the appropriate product family …

• …Choose a product…

• …And get the content

Select the appropriate geographic region and language - allows you to customize your view

Point of Contacts: • CAMs (Customer Account Managers) – The customer can direct

questions/issues to their internal training prime, who can be in contact with the Nortel CAM.

• CSRs (Customer Service Rep) of regional calling center number

• Instructor – provide business cards/email address/phone number

1-5


Training Page

Page that appears when “Training” is selectedDepending on your selection, you see the training offer in your region (NA, EMEA, ASIAPAC, CALA) or the global offer.

1-6


Curriculum Paths Page

Page that appears when “Curriculum Path” is selected.You can select the appropriate training according to your job function.

1-7


Technical Documentation

www.nortel.comSelect Support & TrainingSelect Technical Documentation

1-8


GSM BSS Nortel Technical Publications

BSS Documentation roadmap

411-9001-000BSS Overview411-9001-001

OMC-R Fundamentals411-9001-006

BTS S8000/S8002/S8003/ S8006 Fundamentals

411-9001-063BTS e-cell Fundamentals

411-9001-092PCUSN Fundamentals

411-9001-091BSC 3000/TCU 3000

Fundamentals 411-9001-126BTS S12000

Fundamentals 411-9001-142

BTS 18000Fundamentals411-9001-160

Concepts

New in this Release411-9001-088

Upgrading

BSS CT2000 Fundamentals 411-9001-137

WPS for PCUSN Configuration Procedures

411-9001-201

BSS CT2000 Configuration -

Procedures 411-9001-148

Configuring

OMC-R Commands Reference -

Security, Administration, SMS-CB, and Help

menus 411-9001-130

Administrationand Security

BSS Fundamentals -Operating Principles

411-9001-007

BSS Configuration -Operating Procedures

411-9001-034

OMC-R Commands Reference –

Objects and Fault menus 411-9001-128

BSS Parameter Reference

411-9001-124RACE Fundamentals and

Commands Reference 411-9001-127

Operations Fault and Performance Management

BTS S12000 Troubleshooting411-9001-144

Fault Management -Maintenance Principles

411-9001-039BTS 18000

Troubleshooting411-9001-162

BTS S8000/S8002/S8003/ S8006 Fault Clearing

411-9001-103

BSS Fault Clearing Advanced Maintenance

Procedures411-9001-105

PCUSN Fault Clearing411-9001-106

BTS S12000 Fault Clearing

411-9001-143BTS 18000

Fault Clearing411-9001-161

BSC 3000/TCU 3000 Fault Clearing 411-9001-131

Call Trace/Call Path Trace Analyzer Performance Management 411-9001-060

OMC-R Commands Reference-

Configuration, Performance, and

Maintenance menus 411-9001-129

BSS Performance Management -

Observation CountersDictionary

411-9001-125

TML (BTS) Commissioning and Fault

Management 411-9001-051

TML (BSC 3000/TCU 3000) Commissioning

and Fault Management 411-9001-139

OMC-R Routine Maintenance and Troubleshooting411-9001-032

BSC 3000/TCU 3000 Troubleshooting 411-9001-132

e-cell Troubleshooting411-9001-090

BTS S8000/S8003 Troubleshooting411-9001-048

BTS S8002 Troubleshooting 411-9001-084

WPS for PCUSN Installation&

Administration411-9001-202

WPS for PCUSN Fundamentals411-9001-802

WQA Fundamentals411-9001-205

WQA Installation and Administration411-9001-206

WQA Configuration Procedures

411-9001-207

BSS Terminology411-9001-803

CT2000 Configuration Reference

411-9001-804

BSS Performance Management -

Observation Counters Fundamentals 411-9001-133

CT2000 Installation& Administration411-9001-149

GSM BSS Nortel Technical PublicationThis suite is sorted by job functions category.

1-9


Course Objectives

> Describe the physical and functional architecture of the BSC3000 and TCU3000

> Describe the modules functions and interfaces

> Trace the signaling and traffic paths inside and outside the equipment

> Explain how to operate and maintain BSC3000 and TCU3000

1-10


Course Contents

> Introduction

> BSC 3000 and TCU 3000 Presentation

> BSC 3000 and TCU 3000 Architecture

> Data Flow Exercises

> BSC 3000 and TCU 3000 Operation

> BSC 3000 and TCU 3000 Maintenance and Enhanced Exploitability

> BSC 3000 and TCU 3000 Provisioning

> Exercises Solutions

1-11

Student notes:

1-12

Student notes:

2-1


nortel.com/training

November, 2006411-1075A-001.1603

BSC 3000 and TCU 3000 Presentation

Section 2

2-2


Objectives

After this module of instruction, you will be able to

> list the BSC 3000 functions

> list the TCU 3000 functions

2-3


Contents

> BSS in GSM Network> BSC Functions > TCU Functions > BSC 3000 and TCU 3000 > BSC 3000 and TCU 3000 Hardware

2-4


BSS in GSM Network

TRAU(TCU)

BSCOMC-R

MSCRadio

InterfaceA Interface

Ater Interface

Abis Interface

NSS

BSS

OMN Interface

Public Switched Telephone Network

MS

MS

S8000Outdoor

BTS

SunStorEdge A5000

RadioInterface

e-CellBTS

PCUSN

GPRS Core NetworkInternet

Gb Interface

Agprs Interface

BTS18020

Combo

S12000Indoor

BTS

BTS18010

The Base Station Subsystem includes the equipment and functions related to the management of the connection on the radio path.It mainly consists of one Base Station Controller (BSC), and several Base Transceiver Stations (BTSs), linked by the Abis interface.An optional equipment, the Transcoder/Rate Adapter Unit (TRAU) so called TransCoder Unit (TCU) within Nortel Networks BSS products, is designed to reduce the number of PCM links.These different units are linked together through specific BSS interfaces:

• Each BTS is linked to the BSC by an Abis interface.

• The TCUs are linked to the BSC by an Ater interface.

• The A interface links the BSC/TCU pair to the MSC.

• The Agprs interface links the BSC to the PCUSN.

2-5


BSC Functions 1 - Basic Functions

Routing

Terrestrial Resources Management

MSCBTS

BTS

BTS

BTS

CAUTION: CRASHON E12 HIGHWAY

Traffic ConcentrationSMS-CB

Management

The basic functions of the BSC are the following:• Terrestrial resource management:

—setup/release of terrestrial channels,

—channel switching between MSC and BTS.

• Radio resource management:

—setup/release of radio channels,

—radio channel monitoring.

• Traffic concentration on the Ater interface.

• Short Message Service - Cell Broadcast management:

—broadcasts short messages defined on OMC-R that are towards target cells.

2-6


BSC Functions 2 - OA&M Functions

Data + Software

OMC-R Interface Management

Ethernet

Observation

BTS and TCU Management

Supervision

Shut down

Startup

The main OA&M functions of the BSC are the following:• BTS and TCU management:

—software downloading,

—initialization,

—supervision,

—configuration and reconfiguration,

—observations.

• OMC-R Interface management which consists of:

—managing links with the OMC-R,

—providing the services requested by the OMC-R,

—storing the BSS configuration data: software storage and distribution among the various entities of the BSS.

2-7


TCU MSC

TCU Functions

BSCBTS

16 kbps16 kbps16 kbps16 kbps

4 speech channels+ signaling

or 4 data channels

4x64 kbit/s

MSC Premises

1 x 64 kbit/s

Converts the GSM speech frames intoPSTN/ISDN A-Law or µ-Law speech.

Abis Ater A

Also called "TRAU" for Transcoder and Rate Adapter Unit.

The concept of remote transcoders is used to convey four multiplexed channels at 16 kbps onto a single 64 kbit/s PCM channel.Multiplexing is implemented within the BTS, thus the number of PCM links needed on the Abis interface is reduced.The TCU enables code conversion of 16 kbit/s channels from the BSC into 64 kbit/s channels for the MSC in both directions.TCU is the product designation of Nortel for the TRAU (Transcoder and Rate Adapter Unit) specified in the GSM recommendations.

2-8


BSC 3000 and TCU 3000 1 - Physical Presentation

BSC 3000 TCU 3000

The BSC 3000 and the TCU 3000 are one-cabinet equipment assemblies, composed of two Nodes and one Service Area Interface. These Nodes are each housed in a sub rack comprising two shelves.The cabinet is designed for indoor applications. The design allows a front access to the equipment. External cabling from below or above is supported.The Service Area Interface or SAI is installed on the left side of the cabinet:

• it provides front access to the PCM cabling,

• it contains electrical equipment used to interface the BSC or the TCU and the customer cables.

The product is EMC compliant. No rack enclosure is required for this reason, as EMC compliance is achieved at the sub rack level (Control Node, Interface Node and Transcoding Node).

2-9


BSC 3000 and TCU 3000 2 - Physical Description

ServiceArea

Interface(PCM cabling)

ControlNode

InterfaceNode

TranscodingNode

TranscodingNode

BSC 3000 TCU 3000

ServiceArea

Interface(PCM cabling)

Power Supplies

Fans

Fans

The BSC 3000 is a one-cabinet equipment, composed of two Nodes and one Service Area Interface.The two BSC 3000 Nodes are the Control Node and the Interface Node. In addition, the Control Node (in charge of Call Processing and OA&M) of the BSC 3000 implements a Fault Tolerant architecture, based on redundancy of processes and a load balancing mechanism on the processors, allowing fast recovery of service (within a few seconds) after a hardware failure. The TCU 3000 is a one-cabinet equipment, composed of up to two Transcoding Nodes and one Service Area Interface.The power supply for both the BSC 3000 and TCU 3000 is –48 V dc. The maximum power consumption of the BSC 3000 or TCU 3000 is 2 kW. Each Node (sub rack) is powered by one rack power distribution tray. Each sub rack is cooled by four fans (replaceable). The fan rack is also referred to as the Cooling Unit assembly.

2-10


BSC 3000 and TCU 30003 - Mixed System Architecture

TCU 2GV14.3

BSC 2GV14.3 BSC 3000

V15BSC 3000

V15

TCU 3000V15

TCU 2GV14.3

BTSsV12.4V14.3

BTSsV15

BTSsV15

X.25

Ethernet

OMC-RV15

PCU SNTCU 2GV14.3

The V15 release will introduce the enhanced capacity on the BSC 3000 for Edge functionalities.The BSC 3000 and TCU 3000 are intended to interwork with current BSC 12000, BTS and OMC-R products.These products will require a software upgrade to deal with the BSC 3000 and TCU 3000. The OMC-R/BSC 3000 link is TCP/IP over Ethernet, instead of native X.25 for BSC 12000.The OMC-R/BSC 3000 link over A/Ater Interface is not available in V15.1.1

2-11


BSC 3000 and TCU 3000 Hardware 1 - Cabinet Structure and Cooling

900

600300

Front View Side View

2200

600

The frame dimensions (ETSI standard) are 60 x 60 x 220 centimeters.The total dimensions of the BSC 3000 or TCU 3000 cabinet (frame + SAI) are as follows:W = 90 cm, D = 60 cm, H = 220 cm.The maximum weight of the BSC 3000 or TCU 3000 equipment is 570 kg (BSC 3000+TCU 3000 = 1100 kg). This yields a maximum floor load of 1000kg/m2. The BSC 3000 and TCU 3000, being totally front access equipment, can be installed back to back or back to wall. The work space required in front of the cabinet is 60 to 90 cm width. The cooling unit supports four fan units and an air filter for the equipment is mounted above an air plenum to direct cooling air to the fans. These four fan units are individually replaceable from the front.For climatic and thermal conditions, the BSC 3000 and TCU 3000 are compliant with:

• temperature: –5 °C to +45 °C,

• relative air humidity: 5% to 90%, in operating conditions.

2-12


BSC 3000 and TCU 3000 Hardware 2 - Generic Module

LED Description Meaning

Triangular shape, red color

Rectangular shape, green colormodule status

Round shape, Yellow color Read/write status

The term “module” refers to a circuit pack enclosed by a metallic housing. Packaging circuit packs in modules provides the following features and benefits:

• a single level of EMC shielding,

• radiated containment across boards within a shelf,

• defined volume control of environmental noise,

• ElectroStatic Discharge protection for circuit packs,

• handling ruggedness,

• minimizes EMC retesting for new designs,

• provides visual indicators (LED) on the front plate.

All BSC 3000 & TCU 3000 modules have the same LEDs on the upper part of the front plate of each module to ease on-site maintenance and reduce the risk of human error.

2-13

Student notes:

2-14

Student notes:

3-1


nortel.com/training

November, 2006411-1075A-001.1603

BSC 3000 and TCU 3000 Architecture

Section 3

3-2


Objectives


> List the external interfaces and associated protocols of the BSC 3000 and TCU 3000

> List the different modules of the Control Node, Interface Node and Transcoding Node

> Describe the role of each module

3-3


Contents

> BSC/TCU 3000: External Links> BSC 3000 and TCU 3000 Generic

Architecture> BSC 3000: Control Node> Control Node: ATM Platform> Control Node Architecture> Operation and Maintenance

Unit > Mass Memory Storage> Traffic Management Unit > ATM Subsystem > ATM Switch Module > BSC 3000: Interface Node

> Interface Node > ATM RM > Switching Unit > Low Speed Access Resource

Complex > TCU 3000: Transcoding Node> Common Equipment Module > Transcoder Resource Module> Internal PCM S-Link Allocation > Maintenance Trunk Module

Bus> Shelf Interface Module > Service Area Interface

3-4


BSC/TCU 3000: External Links

BTS BSC TCU MSC

OMC-R

Ethernet

LAPDOMLLAPDRSL

LAPDOML

SS7

PCUSN

LAPD

OM

L

GPRS

Voice Data

Dat

a

LAPDGSL

LAPD

RSL

LAPD

GSL

Agprs

Abis Ater

Three types of signaling are transported over the Abis interface:• LAPD/OML related to the Operation and Maintenance,

• LAPD/RSL for the Radio Signaling Link,

• LAPD/GSL for the GPRS Radio Signaling Link.

The BSC can be connected to the OMC-R through an Ethernet network or through the A interface.Two types of signaling are transported over the Ater interface:

• LAPD/OML for control of the TCU transcoders by the BSC,

• SS7 going to the MSC.

Three types of GPRS signaling are transported over the Agprs interface: • LAPD/OML for control of the PCUSN by the BSC,

• LAPD/RSL for the Radio Signaling Link,

• LAPD/GSL for the GPRS Radio Signaling Link.

3-5


BSC 3000 and TCU 3000 Generic Architecture

OMUOAM

ATM SW

InterfaceNode

TranscodingNode

ControlNode OMU

OAMMMS

ATM SW TMU

TrafficManagement

TMU

TrafficManagement

MMS

Private

TRM

CEMCEM

LSA RC

TRMLSA RCCEM

CEM

LSA RC

LSA RC

ATM RM

8K RM

8K RM

Shared

ATM RM

The BSC 3000 is composed of the Control Node and the Interface Node. The TCU 3000 is composed of the Transcoding Node.Control Node main functionalities are:

• Management of OAM for the C-Node, I-Node and T-Node,• Traffic management towards the BTSs and MSC,• BTS supervision, Transcoding Node supervision,• OMC-R link management,• Failure detection and processing,• HandOver procedures,• BSS configuration and software management,• BSS performance counter management,• ATM Management.

Interface Node main functionalities are:• I-Node OAM management,• Switch management and Timeswitch control,• PCM interface,• ATM Management.

Transcoding Node main functionalities are:• T-Node OAM management,• Switch management and call processing,• BSC Access,• Carrier Maintenance.

3-6


BSC 3000: Control Node

Shelf 1

Shelf 0

TMU

TMU

TMU

ATM

SW

ATM

SW

TMU

TMU

TMU

TMU

SIM

B

TMU

TMU

MM

S Pr

ivat

eM

MS

Shar

ed-F

iller

--F

iller

-

TMU

TMU

TMU

SIM

A

ControlNode

OM

U

-Fill

er -

TMU

-Fill

er -

OM

U

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 3 4 7 8 11 12 13 14 159 102 5 6

TMU

MM

S Sh

ared

MM

S Pr

ivat

e

The Control Node is composed of the following modules:• the Operation and Maintenance Unit (or OMU), which manages all BSC resources,

ensures BSC survival, BSS interface with the OMC-R and disk management,

• the Mass Memory Storage (or MMS), holds all the data:

— private: managed by one OMU,

— shared: managed by both OMUs, for data that must be secured and still accessible in the event of an OMU or disk failure,

• the ATM Switch (or ATM SW), which implements the ATM network used as the Control Node backplane, and provides ATM on OC-3 connectivity towards the Interface Node,

• the Traffic Management Unit (or TMU), which provides the processing capability required to perform the GSM/GPRS processing and protocol termination required for GSM interfaces,

• the Shelf Interface Module (or SIM), which provides the power (-48 V) and alarm interfaces for the sub-rack.

The maximum configuration for the Control Node is the following:• 2 OMU modules,

• 14 TMU modules,

• 4 MMS modules, (2 private and 2 shared),

• 2 ATM SW modules,

• 2 SIM modules.

3-7


Control Node: ATM Platform

OMUOAMControlNode

S-links

InterfaceNode

CEM

64 kbit/s

ATM RM

ATM/PCMInterface

ATM RM

ATM/PCMInterface

OMUOAM

TMUTMU

TMUTMU

TrafficManagement

ATM Links (25 Mbit/s) ATM Links

(155 Mbit/s)

Plane 1

Plane 2ATM SW

ATM SW

The Control Node is a computing and signaling platform built around an ATM switch. Globally, the Control Node is designed as a fully redundant ATM switch for any inside and outside communications. Internal and external exchanges are carried over ATM through a redundant optical OC3 connection using ATM at 155 Mbps:

• for internal communication between the Control and the Interface Nodes.

The platform is also fully ATM inside; no ATM connection is ended on the access port of the Control Node (ATM switches), but on any computing module inside the shelf. The addressing to/from the Control Node is based on Vpi, Vci.ATM RMs and ATM SW modules are provisioned in pairs to provide redundancy and connection protection:

• both planes are used at the same time,

• all messages exchanged between ATM RMs and ATM SW modules are duplicated.

3-8


Control Node Architecture

MMS

ToInterface

Node

ToInterface

Node

SCSIInterface

OMUOAM OMUOAM

SCSIInterface

MMSMMS

MMS

TMU

TrafficManagement

TMU

TrafficManagement

TMU

TrafficManagement

TMU

TrafficManagement

ATM Link (155 Mbit/s)

ATM Link

(155 Mbit/s)

ATM Links (25 Mbit/s)


ControlNode

OMC-ROMC-R

ActivePassive

ATM SW ATM SW

The Control Node is composed of the following three functional modules:• the Asynchronous Transfer Mode SWitch or ATM SW, which implements the

ATM network used as the Control Node backplane, and provides ATM on OC-3 connectivity towards the Interface Node,

• the Operation and Maintenance Unit or OMU, which manages all BSCresources, ensures BSC survival, BSS interface with the OMC-R and disk management,

• the Traffic Management Unit or TMU, which provides the processing capability required to perform the GSM treatments and protocol termination required for the GSM interfaces. One TMU computes 300 Erl (whatever the subscribers profile).

The Mass Memory Storage or MMS, is only a hard disk.

3-9


Operation and Maintenance Unit 1 - Overview

OMU

OperationAdministration

&Maintenance

ControlNode

InterfaceNode

OMC-R

DiskManagement

OMC-RInterface

TMLInterface

TMU

TrafficManagement

ATM SW

TMU

TrafficManagement

TML

MMS

The Operation and Maintenance Unit module is responsible for the following functions:

• management of all BSC resources (both Control and Interface Nodes),

• BSS interface with the OMC-R (Ethernet),

• OMC link management either by a physical serial link or constant bit rate data sent to the ATM datalink,

• disk management,

• Local Maintenance Terminal (TML).

The OMU is provisioned in a 1+1 redundancy scheme.

3-10


Operation and Maintenance Unit 2 - OA&M Functions

PCUSN

InterfaceNode

OMU

ConfigurationManagement

PerformanceManagement

FaultManagement

ATM SW ATM RM

8K RM LSA RC

OMC-R

BSCOA&M

TranscodingNode

TRM

CEM

CEM

LSA RCTMU

TrafficManagement

BTSOA&M

TCUOA&M

PCUSNOA&M

ControlNode

TMU

TrafficManagement

BTSOA&M

TCUOA&M

PCUSNOA&M

The OA&M function is in charge of management and supervision of:• internal BSC equipment,

• other BSS equipment: BTS, TCU, PCUSN.

The OA&M entity for BTS resources is mapped on the TMU (for direct Abis access), as well as TCU and PCU supervision and OA&M. For each resource, the OA&M ensures classical functions as:

• Configuration Management,

• Fault Management: detection, resolution, notification, correction,

• Performance Management: measurements,

• Upgrade Management,

• Test Management.

The OA&M is not a simple OMC-R agent on the BSC, it has its own decision criteria to involve some actions after orders or observations:

• overload protection,

• switching of activity (swact) on module failure (fault tolerance),

• defense against applicative inconsistencies.

3-11


Operation and Maintenance Unit 3 - BSS Interface with OMC-R

OAM

Association(proprietary)

RFC 1006

TCP

IP

Ethernet

BSC OMC

OAM

Association(proprietary)

RFC 1006

TCP

IP

Ethernet

TCP/IPNetwork

RS232 (Debug)

RJ45

Though the same OMC-R manages both the BSC 2G and the BSC 3000, the interface between the BSC 3000 and the OMC-R is Ethernet TCP/IP, instead of X.25 as for the BSC 2G. Two data paths are available for OMC-R access and/or other purposes:

• PCM: on one or more TS (DS0) via the LSA RC module, (available in V15)

• Ethernet: TCP/IP on Ethernet 10/100 Mbps.

The direct Ethernet connection is provided by the RJ45 connector of the OMU faceplate.A switching device or four-ports LAN Hub, located in the SAI, is required.A small sub layer based on IETF RFC 1006, allows dialog with Association (proprietary) and Application layers. When the BSC 3000 is remote from the OMC-R, they can be interconnected through a network (X.25, Frame relay, etc.) with a minimum throughput of 128 kbps.

3-12


Mass Memory Storage1 - SCSI Bus

ControlNode

SCSI BusOMUOAM

MMS

OMUOAM

MMS MMS MMS

PrivateDisk

PrivateDisk

SharedDisk

SharedDisk

Active Passive

There are four Mass Memory Storage modules (hard disk) in the BSC.They are linked to the OMU modules through four SCSI buses.Two SCSI buses are dedicated to the two private disks storing:

• OS AIX (400 Mb),

• software for OMU boards.

Two of them are for mirrored shared disks storing:• local MIB (BDA),

• observations, notifications,

• Call traces,

• Supervision,

• BTS and TCU softwares.

The pair of shared SCSI buses, and the disks on them are only managed by the active OMU. The shared SCSI buses will only be accessed after “election” of the active OMU.When a switch of activity occurs (fault tolerance mechanism), the newly active OMU gains control of the pair of shared SCSI buses.

3-13


Mass Memory Storage 2 - Disk Sub-system

ControlNode

SCSI BusOMUOAM

MMS

OMUOAM

MMS MMS MMS

PrivateDisk

PrivateDisk

SharedDisk

SharedDisk

Active Passive

1 2

Transactions when:OMU-1 is activeOMU-2 is activeAlways available

Each OMU module controls a private disk which holds all the private data (OS and System data) for the module and a pair of shared disks (BSS database and GSM data) managed in a mirroring way. Each Mass Memory Storage module contains a SCSI-2 hard disk of 9 Gbyteseach.At boot time, each OMU module has access to its private SCSI and so to its private disk.The pair of shared disks holds the data that must be secured and still be accessible in the event of an OMU failure or a disk failure.The protection of the shared disks is independent from the protection of the OMUs: the non active OMU can be extracted from the system without any impact on the disk transactions.In the event of the extraction of the active OMU, a swact of the OMUs occurs, and the disk subsystem is still protected from a single failure.

SWACT = SWitch of ACTivity. This refers to a sparing action where an inactive board takes control over a faulty active board. In the Control Node, this applies to the OMU and the TMU modules, in the Interface Node to the CEM and the LSA RC.

3-14


Mass Memory Storage3 - MMS2 Introduction

ITMBlock

DC/DC converter

-48 V

MTM bus

SCSI bus

LVD SCSITerminator

Remove request

LVD SCSI disk

80 pts SCA

SCSIExpander

LVD SCSITerminator

Disk shut-downExpander isolation

End of SCSI bus

LVD SCSITerminator

Activation is slot dependent

Activation is slot dependent

BackplaneLED drive

New disk(73 Gb from Hitachi) SCSI expander

MMS2 HW presentation:• Bigger capacity disk: 73 GB (vs 9 GB for MMS1)

• MMS2 boards are replacing MMS1 boards with the same functionality.

Mixed configurations MMS1 / MMS2 authorized:• MMS2 module can be mixed with MMS1 module for private or shared MMS

• Upward compatibility but no backward compatibility:

—A MMS1 module can be replaced by a MMS2 module

—A MMS2 module already installed must not be replaced by a MMS1 module.

Operations with MMS2• MMS2 boards can be introduced in a BSC e3 shelf in any of the place

available for the current MMS boards (private and shared). SW compatibility with V16.

3-15


Mass Memory Storage

Front Panel View

Removal Request Push Button

4 - MMS2 and Higher Capacity Disk

Installed disk Replacement disk

9 Gb/ 36 Gb flanged 9 Gb/ 36 Gb flanged

73 Gb 73 Gb

9 Gb/ 36 Gb flanged 73 Gb

73 Gb 9 Gb/ 36 Gb flangedXOnly replacement of a disk by a disk of same or

higher capacity is supported

The current MMS1 module (9 GB) houses a SCSI hard disk. The new MMS2 disk, introduced in V15.1, is a 73 GB device and it uses the same SCSI interface as the 9 GB disk.

3-16


Traffic Management Unit 1 - Main Functions

TMU

Traffic Management:Radio resource: TMG_RADConnection (setup, release, HO): TMG_CNXA interf. Messages (paging, incoming HO): TMG_MESAgprs interf. Messages: TMG_RPP

AterManagement:

TMG_COM

SS7 Management:SCCPMTP1, MTP2, MTP3

BTS Sitesupervision:

SPR

LAPD Management:Level 1, 2 and 3

TCU 3000 / TCU 2Gsupervision:

SUP-TCU / SPT

PCUSNsupervision:

SPP

The TMU is responsible for the BTS configuration and the main Call Processing functions:• GSM/GPRS traffic management,• GSM signaling (LAPD and SS7),• GPRS signaling (LAPD).

These functions are processed by six software modules.TMG_RAD:

• manages radio resources for a group of sites: allocation, modification and release of radio channels,

• manages the RSL dialog on the Abis and radio interfaces,• supervises coherence of allocated channels between the BTSs and the BSC.

TMG_CNX:• drives setup, release, assignment and handover,• asks for traffic connections.

TMG_MES:• codes/decodes A interface messages,• drives connectionless messages: paging, incoming HO.

TMG_RPP: codes/decodes Agprs interface messages.TMG_COM: allocation, release and administration of terrestrial circuits (CICs).SPR: Supervision of BTS sites (configuration and defense).For reliability purpose, the main Call Processing sub-functions use the Fault Tolerance service: for each sub-function there is one active entity on a TMU and one passive entity on another TMU.

3-17


Traffic Management Unit 2 - Call Processing and Traffic Management

Resource Allocator

Transparent Message Transfer

BSCtransaction

MSCconnection

Radioconnection Abis

Distr.Layer

LAPDA

Distr.Layer

SCCP

Call Processing

The Traffic Management Unit (TMU) is responsible for managing the GSM protocols in a large acceptance:

• provide processing power for GSM Call Processing,

• terminate GSM protocols (A, Abis and Ater interfaces),

• terminate low level GSM protocols (LAPD and SS7).

The GSM Call Processing function is responsible for the management of GSM communications:

• traffic management (connections and transfer of user information MS/MSC),

• network resource allocation (terrestrial circuits and radio resources),

• handover,

• radio measurements,

• power control.

The corresponding software is spread over all TMU modules, but is split into several entities:

• radio resource allocation: per BTS site,

• terrestrial circuit allocation: per TCU and per PCUSN,

• MSC connection and BSC transaction: internal criteria.

3-18


Traffic Management Unit 3 - TMU2 Introduction

MPC8560

Core @ 833 MHzCPM @ 333 MHz

2 MB SSRAM

Flash8 MB

PHY ATM77V106CPLD

UTOPIALevel 18 bit

ITM MIM

PHY ATM77V106

PLLMT9043

512 MB DDR333SDRAM

Optional PMC slot

Optional PMC slot

TDMclock

8 KHz

ATM25,6Mbps

Or51,2 Mbps

MTM

TMU2 HW Presentation:One Single board (TM+SBC+PMC).NTQE04BA

TMU2 Capacity:525 Erlang120 Lapd & 4 SS7TMU2=1.75TMU1

TMU2 Technical spec:Based on PQ3 processor (MPC8560)512 Mbytes RAM

TMU2 Compatibility:Mixed Configuration TMU/TMU2 allowedNo specific operation at introductionSW Compatibility V16

The distinction between TMU1 and TMU2 will be made thanks to a different PecCode value.

TMU2 is a mono-processor board, based on a Freescale PowerQuicc III using 512 MB SDRAM. It will be in charge of all tasks previously performed by the 3 TMU1 processors (SBC, PMC and TM sub-boards).120 LAPD and 4 SS7 ports are available for signaling channels.

TMU1: PecCode = NTQE04AATMU2: PecCode = NTQE04BA

3-19


ATM Subsystem ATM 25 Interface Distribution

OMUOAM TMU

TrafficManagement

TMU

TrafficManagement

OMUOAM

Active Passive


ControlNode

ATM SW

ATM Switch

ATM SW

ATM Switch

Active Active

The Control Node uses a duplex, star connectivity, with cell switching performed in both ATM SW modules at the center of the stars and the other Resource Modules at the leaves.From a hardware perspective, the ATM subsystem is the key factor for platform robustness and scalability.This subsystem provides reliable backplane board interconnections with live insertion capabilities. It has two main components:

• a pair of ATM switches (ATM SW module), working simultaneously,

• an ATM Adapter, located in each of the OMU and TMU modules.

The connections between modules use redundant ATM 25 point to point connections to ATM switches, allowing:

• high fault isolation, signal integrity,

• live insertion,

• backplane redundancy,

• scalability.

The backplane supports a redundant ATM 25 Mbps to any slot using the ATM 25 standard as defined by the ATM Forum.It carries all the internal signaling information, using the AAL1 and AAL5 protocols.

3-20


ATM Switch Module 1 - Functions

ATM SW

OMUOA&M

OMUOA&M


OC3 LinkSONET

(155 Mbit/s)ATM Switch

Opticalinterface

TMU

TMU

TMU

TMU

TrafficManagement

MUX

AAL5- AAL1 SAR

ATM Link (155 Mbit/s)

MessagingCommunication

ATM routing tableOA&M

6 x ATM Links (25 Mbit/s)

MUX6 x


MUX4 x ATM Links (25 Mbit/s)

MUX

ATM Physical Interface

UTOPIA

2 xATM Links (25 Mbit/s)

ATM SW

The ATM SW module provides a high performance interconnection between the OMU and TMU modules, as well as the ATM on OC-3 connectivity towards the Interface Node.Messaging provides a generic API to all entities which exchange messages to one another, using the UDP service only.Communication is in charge of all communication tasks:

• between processors of the module (TCP/IP),

• file transfer with the OMU module (TCP/IP and FTP).

The ATM routing table management configures on startup and allows modification of the routing table at run-time for AAL1 and AAL5.The OA&M Local Agent centralizes all administrative action relative to the module:

• configuration,

• performance measurement,

• fault notification.

The ATM SW is provisioned in a 1+1 active/active scheme, with both modules working simultaneously.

3-21


ATM Switch Module 2 - ATM Switching Principle

ATMSwitch

Cell 1 Cell 2

Port 1 Port 2Port 3

1 8 6 4 4 5

2 9

Input OutputPort VPI VCI Port VPI VCI

11 8 2 4 5

1 6 4 3 2 9

Switching Table

VC/VP ATM Switch: Input(Port, VPI, VCI) ® Output(Port, VPI, VCI)VP ATM Switch: Input(Port, VPI) ® Output(Port; VPI)

.

.

.

ATM switching consists first in establishing a virtual circuit for each communication using a virtual channel or VC and a virtual path or VP.These virtual circuits are established statically according to engineering rules, they are Permanent Virtual Circuits or PVCs.The main function of an ATM switch is to receive cells on a port and to switch those cells to the proper output port based on the VPI and VCI values of the cell.This switching is controlled by a switching table that maps input ports to output ports based on the values of the VPI and VCI fields.While the cells are switched through the switching fabric, their header values are also translated from the incoming value to the outgoing value.Addressing tables converting between VP/VC and slot number are loaded from ATM SW module at startup time and stored in the flash EPROM of the ATM part of all modules:

• AAL1 routing tables are dynamic,

• AAL5 routing tables are static.

3-22


BSC 3000: Interface Node

-Fill

er -

-Fill

er -

-Fill

er -

Shelf 1

Shelf 0

InterfaceNode

ATM

RM

ATM

RM

LSARC

LSARC

SIM

A

LSARC

LSARC

CEM

0C

EM 1

8k R

M 0

8k R

M 1

-Fill

er - LSA

RC

SIM

B

LSARC

0

1 2 3

45

Mandatory for

synchronization

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 3 4 7 8 11 12 13 14 159 102 5 6

The Interface Node is connected to the Control Node by four optical fiber cables with a standard ATM interface.There are four major hardware modules that make up the Interface Node:

• the Common Equipment Module (or CEM),

• the 8K subrate matrix Resource Module (or 8K RM),

• the Low Speed Access Resource Complex module (or LSA RC),

• the Asynchronous Transfer Mode Resource Module (or ATM RM).

The maximum configuration for the Interface Node is the following:• six LSA RC modules,

• two ATM RM modules,

• two 8K-RM modules,

• two CEM modules,

• two SIM modules.

The CEMs have special slots (slots 7 and 8, in shelf 0), and both are always provisioned.The LSA-RC module 0 is mandatory, as the Interface Node is synchronized through the PCMs of this slot (synchronizing PCMs 0-1-2-3-4-5).

3-23


Interface Node 1 - General Architecture

Ater

AbisBTS

TCU

ControlNode

S-links

S-links

Switching Unit

InterfaceNode

ATM RM

ATM/SlinkInterface

8K RM

8 kbit/s

CEM

64 kbit/s

LSA RC

PCMController

AgprsPCUSN

The Interface Node is composed of a controller (CEM) and a set of resource modules (RM) that are connected point-to-point to the CEM through the back-panel and communicate via a proprietary communication protocol called “S-link”. The main function of the modules are the following:The Common Equipment Module (or CEM) controls the BSC Interface Node Resource Modules, and provides:

• system maintenance,

• clock synchronization,

• traffic switching.

The Asynchronous Transfer Mode Resource Module (or ATM RM) adapts Time Slot (DS0) based voice and data channels of S-links to ATM cells for transmission over a Synchronous Optical NETwork (SONET), OC-3c interface.The 8K subrate matrix Resource Module (or 8K-RM) adds subrate switching capability to the Interface Node, as the CEM is only capable of switching at a TS (DS0) level (64 kbps circuits). The Low Speed Access Resource Complex or LSA RC is used to interface the BSC to both TCU and BTS using PCM links (E1 or T1).

3-24


Interface Node 2 - Detailed Architecture

BTS, TCU, PCUSN

Control NodeControl Node

BTS, TCU, PCUSN

InterfaceNode

Switching UnitIMC links

S-links

S-links

ATM RM

ATM/SlinkInterface

8K RM

8 kbit/s

LSA RC

PCMController

LSA RC

PCMController

ATM RM

ATM/SlinkInterface

8K RM

8 kbit/s

Active

CEM

64 kbit/s

Active

CEM

64 kbit/s

PassivePassiveDS512 DS512

The Interface Node architecture is based on a duplicated Common Equipment Module or CEM. Other modules: ATM-RM, LSA-RC and 8K-RM are connected to the CEMs via proprietary PCM serial links (S-links). The active CEM sources all PCM streams leaving the CEM complex. Both CEMs receive identical PCM traffic from all sources. The two CEMs can communicate with each other via the Inter Module Communication (IMC) links, in order to synchronize Call Processing and maintenance states.This results in a point-to-point architecture, which (when compared to bus architectures) provides:

• superior fault containment and isolation properties,

• fewer signal integrity related problems,

• easier backplane signal routing.

In addition to payload TSs (DS0), the S-links transport messaging channels, overhead control and status bits between the CEMs and the RMs.

3-25


ATM RM Logical Architecture

ControlNode

InterfaceNode

ATM RM

ATM/SlinkInterface

S-linksPCM

Physical layerOC-3

Interface

AAL1(LAPD, SS7)

DS0mapper

SPMmessaging

S-linksPCM AAL5

(OA&M, CallP)

SegmentationAnd

Reassemblysublayer

ConvergenceSublayer

ATMlayer


The main functions of the ATM-RM are:• terminating the OC3 optical interface using a single mode fiber,

• mapping TSs (DS0) from the six S-links, to ATM cells using AAL1 in Structured Data Transfer mode,

• relaying the contents of AAL5 cells to the CEM (BSC OA&M and Call Processing).

The ATM treatment is composed of three layers:• AAL layer (Convergence Sublayer and SAR Sublayer),

• ATM layer,

• Physical layer (OC3 interface).

The ATM RM is configured statically to associate one VP/VC, corresponding to a channel on the Control Node side, to one TS of the S-links to the CEM.The ATM RMs are provisioned in pairs to provide redundancy and connection protection. Both modules are used at the same time and the messages are duplicated.

3-26


Switching Unit 1 - Common Equipment Module

SwitchingMatrix 64K

CEM

PCM Clock

Interface NodeOA&M

8K IntegratedConnection

Manager

Switch Manager(Call Processing)

64KConnection

Manager

InterfaceNode

LSA RC

PCMController

LSA RC

PCMController

ATM RM

ATM/SlinkInterface

8K RM

8 kbit/s

The Common Equipment Module is the main module of the Interface Node.The CEM handles the following functions:

• channel connection management, (traffic switching),

• controls the Resource Modules of the Interface Node (downloading, testing, configuring),

• provides system maintenance, using the TML,

• clock synchronization,

• alarm processing.

The main function of the Switch Manager, is to establish, release and modify Abis/Ater connections in the Switching Matrix (switch fabric), under the control of Call Processing (TMU). Its other function is to establish 64 kbps connections for signaling links.The switch fabrics are updated on both CEMs to ensure consistency between them.The CEM is provisioned in a 1+1 hot stand-by redundancy scheme. One CEM is active, i.e. actually performing Call Processing functions, while the other is inactive, ready to take over if the active module fails.The messages between the IN-OA&M application (OMU) and the CEM are exchanged using the IP protocol over AAL5 ATM circuits. The IN O&M application handles only IP addresses and TCP/UDP ports.

3-27


Switching Unit 2 - Common Equipment Module and 8K RM

Switching UnitCEM

64 kbit/s

8K RM

8 kbit/s

LSA RC

PCMController

LSA RC

PCMController

Primary

CEM

The switching unit manages all the flow of connections sent by the Call Processing and BTS OA&M applications from the Control Node (TMU). The Integrated Control Manager or ICM software of the CEM is responsible for establishing connections between bearer channels, using a two stage matrix.The switching unit is composed of two types of module:

• the Common Equipment Module or CEM offers a 64 kbps matrix (switch fabric) only capable of switching at TS (DS0) level,

• the 8K RM is a subrate matrix Resource Module which provides a secondary stage of switching individuals bits within each TS.

Internal dialog between CEM and other modules (LSA RC and 8K RM) is carried out by reserved TSs (30 to 40) of the Primary S-link.

3-28


Switching Unit 3 - 8K-RM (SRT)

CEM

8K RM

Primary

SwitchingMatrix 8K2268 TSs

MessagingInterface

Clock

S-linkInterf.

ChannelSequencer

Main Bus

CEM

64 kbit/s

Active

Passive

The 8K RM is used for the bearer channels that have to be switched by the Interface Node, between the Abis interface and the Ater interface. The 8K RM or Subrate Matrix is a 4096 bit-to-bit switch, which can communicate with the two CEMs via:

• nine S-links, connected to the backplane.

The 8K RM is provisioned in a 1+1 (active/active) redundancy scheme.The active CEM module controls the switching activity of the two 8K RM modules, using the 36 reserved TSs of the Primary S-link:

• switch messaging (30 TS),

• synchronization (6 TS).

The S-link Interface extracts messaging for communication with CEMs and generates the reference clock.The Channel Sequencer performs rate adaptation and channel selection.The Switching Matrix performs channel switching at an 8 kHz frame rate, using an eight-bit matrix, working in parallel. The fanout is limited to 2268 Time Slots (payload).

3-29


Switching Unit 4 - DS512

Internal Switching ConnectionsCEM

64 kbit/s

8K RM

8 kbit/s

S-Links

CEM

DS512

New

External Switching Connections

101

In V14., the BSC 3000 can switch up to 2268 DS0 on Abis, Agprs and AterInterfaces.With the introduction of EDGE, this switching capacity needs to be increased, in order not to become a limiting factor.To increase the BSC 3000 switching capacity, four DS512 links (optical fibers) are established, between CEM and 8K-RM module.With this connection, the BSC 3000 DS0 capacity increases from 2268 DS up to 4056 DS0.

3-30


Switching Unit5 - Internal S-Link Connection

BTS

S-link

TCU,PCUSN

S-link

S-link

S-link

S-link

Interface Node

8K RM

8 kbit/s

S-link

S-link

SwitchingMatrix 64K

CEM

64 kbit/s

ATM RM

ATM/SlinkInterface

LSA RC

PCMController

LSA RC

PCMController

S-link

S-link

S-link

LSA RC

PCMController

Control Node

9 S-links

S-links = 256 Time Slots or DS0 (64 kbps)

9 S-links = 256 x 9 = 2304 Time Slots

As the 8K RM needs nine S-links to the CEM it has a fixed position into the Interface Node shelf no. 0.Whereas the LSA-RC and ATM RM only need three S-links for the back panel connection.Each S-link provides 256 Time Slots.

3-31


Switching Unit 6 - Switching LAPD and SS7 Time Slots

InterfaceNode

LSA RC

PCMController

ATM RM

ATM/SlinkInterface

CEM

64 kbit/s

LSA RC

PCMController

LSA RC

PCMController

ATM RM

ATM/SlinkInterface

TSa2

TSa1

TSb2

TSb1

TSc2

TSc1

TSa1

TSb1

TSc1

LAPD and SS7 messages arriving in AAL1 cells on both ATM modules are carried on two separate S-links to the CEMs.A Y-connection connects the two identical TSs to the required LSA module:

• in the ATM RM to LSA-RC direction, only the TS of the active plane is switched,

• in the LSA-RC to ATM RM direction, the TS is broadcast to both S-links.

S-links used for signaling are called Primary S-links.

3-32


Low Speed Access Resource Complex 1 - Functions

IEM

PCMMapperFramer

Transcoding

NRZHDB3/B8ZS

HDLCController

TIM

IEMSelection

LSA RC

Active

CEM

64 kbit/s

CEM

64 kbit/s

S-links

IEM

PCMMapperFramer

Transcoding

NRZHDB3/B8ZS

HDLCController

Passive

PCM (E1 or T1)

The Low Speed Access Resource Complex or LSA-RC is used to interface the BSC to the TCU, the PCU and the BTS. The LSA-RC is the PCM interface module. It is called “Resource Complex”, as it is made of three modules (taking three slots):

• two Interface Electronic Modules (or IEM), they are in 1+1 hot stand-by redundancy (field replaceable without service disruption),

• one Terminal Interface Module (or TIM), it is a passive switch that switches the PCM towards the active IEM. The TIM does not contain electronic components (very high MTBF) and provides LSA internal redundancy.

Main IEM functions:• the S-Link Mapper is responsible for transferring payload data between the

channels on the S-Link interface and the respective channels of the PCM30/DS1 Link interface,

• transcoding converts signals from NRZ to HDB3 (or B8ZS),

• the HDLC controller is used for LAPD level 2 treatment (only used in the TCU).

The BSC Interface Node can contain up to six LSA-RC modules to provide 126 PCM30 or 168 DS-1.

3-33


Low Speed Access Resource Complex 2 - Physical Architecture

IEMTIM

ToCTMx(CTU)

Spectrum Backplane

RC Mini Backplane

IEM

Bac

kpla

ne

RCMIEM

IEM

PCM

PCM

PCM

TIM

The LSA-RC module is made up of three modules (taking three slots):• two Interface Electronic Module or IEM, that are in 1+1 hot standby

redundancy and field replaceable without service disruption,

• one Terminal Interface Module or TIM.

These three modules are connected on a specific backplane called Resource Complex Mini-backplane or RCM.The RCM is designed for both IEM and TIM modules. It provides:

• Interface for 21 PCM E1 or 28 PCM T1,

• Matched impedance for 120 Ω or 75 Ω for E1 PCM, and 100 Ω for T1 PCM.

3-34


Low Speed Access Resource Complex 3 - LSA RC Front Panel

To search the previous PCM in fault

PCM Number with problem and type of problem

The red indicators indicates a fault condition on the span in the PCM (span) display:Loss Of SignalAlarm Indication SignalLoss of Frame Alignment – (Loss Of Frame Alignment)Remote Alarm Indication

To search the next PCM in fault

Indicates the one IEM module serving as synchronization reference

Problem with IEM module

IEM module operating and not to be removed

The LSA-RC is the PCM (or SPAN) interface module. The Spans can be checked in Automatic mode or in Manual mode, selected by the STOP pushbutton.In Manual mode the SPAN number is selected using the two up and down pushbuttons.The seven red indicators, indicate for the selected SPAN, the following faults:

• LOS: Loss Of Signal,

• AIS: Alarm Indication Signal,

• LFA: Loss of signal Frame Alignment, With T1 IEM the indication is LOF:Loss Of Frame Alignment

• RAI: Remote Alarm Indicator.

The BSC 3000 uses the first PCM ports from LSA logical No. 0 (the LSA in slots [4,5,6] shelf 0) as synchronizing PCMs. By default the synchronizing ports are:

• No. 0, 1, 2, 3, 4, 5 for E1 PCMS,

• No. 0, 1, 2, 3, 4, 5 for T1 PCMs.

3-35


Low Speed Access Resource Complex 4 - External Connection

CTB

CTMx1CTMx2CTMx3CTMx4CTMx5CTMx6CTMx7

Cable Transition Unit

LSA RC module

Rx cables

Tx cables

TIM

IEM

Active

IEM

Passive

RCM

Two versions of the LSA-RC module exist:• for International PCM30: 21 E1 PCMs, HDB3 coding,

• for North American DS-1: 28 T1 PCMs, AMI or B8ZS coding.

Each LSA is associated with a CTU (Cable Termination Unit). The CTU is housed in the PCM cabling interface (so-called Service Area Interface) and provides copper concentration.

The SAI is a cabinet, attached to the BSC frame, enabling front access to the PCM cabling. It can host up to six CTUs (plus two optional HUBs) in the BSCSAI and eight CTUs in the TCU SAI.

3-36


Low Speed Access Resource Complex 5 - IEM / IEM2

LSA RC module

TIM

IEM

Active

IEM2

Passive

RCM

LSA RC module

TIM

IEM2

Active

IEM2

Passive

RCM

The Interface Electronics Module (IEM) is a component of the Low Speed Access(LSA) module. It is the electronic interface for E1 or T1 PCMs. Two IEM instances are associated to each LSA, duplicated in a 1+1 protection scheme.LSA modules are located either in BSC 3000 Interface Node or in TCU 3000.

This IEM2 evolution is part of the normal life cycle management of the BSC/TCU 3000 H/W modules.

Mixed configurations of IEM1 and IEM2 modules are allowed on the same shelf. Therefore, an LSA may be equipped with two IEM1 modules, with two IEM2 modules, or with one IEM1 module and one IEM2 module.

3-37


TCU 3000: Transcoding Node

TranscodingNode

TranscodingNode

T

R

M

T

R

M

CEM

LSARC

1LSARC

2LSARC

3

LSARC

0

Shelf 1

Shelf 0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 152

1 3 4 7 8 11 12 13 14 159 102 5 6

Mandatory for

synchronization

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

T

R

M

S

I

M

S

I

M

T

R

M

T

R

M

S

I

M

S

I

M

T

R

M

T

R

M

T

R

M

T

R

M

Fille

r

Fille

r

CEM

The TCU 3000 is based on the Spectrum architecture, as is the Interface Node of the BSC 3000. One TCU 3000 cabinet consists in:

• two independent Transcoding Nodes (one sub-rack),

• one cabling interface area or SAI, which provides front access to the PCM cabling.

Each sub-rack supports twenty-eight modules (or slices) and two power interface modules or SIMs.The TCU 3000 uses the last PCM ports from LSA logical No. 0 (the LSA in slots [4,5,6] shelf 0) as synchronizing PCMs. By default the synchronizing ports are:

• No. 15, 16, 17, 18, 19, 20 for E1 PCMS,

• No. 22, 23, 24, 25, 26, 27 for T1 PCMs.

3-38


TCU 3000 Transcoding Node

TranscodingNode

BSCMSC

BSCMSC

IMClinks

S-links

S-links

Up to 4 LSA-RC modules

TRM

Vocoders

CEM

64 kbit/s

CEM

64 kbit/s

TRM

Vocoders

Up to 12 TRMs

ActivePassive

LSA RC

PCMController

LSA RC

PCMController

The Transcoding Node (or TCU) is composed of a controller (CEM) and a set of Resource Modules (RM) connected point-to-point to the CEM via S-links, through the backpanel.

One TCU is composed of the following physical entities:• two Common Equipment Modules (or CEM), identical to the BSC CEM

module, including:

—an OA&M processor,

—a 16 x 16 PCM link (32 TS) switching matrix,

—a circuit used to synchronize the time base on the clock taken from three of the PCM links connected to the MSC,

• up to twelve Transcoder Resource Modules (or TRM) which enable voice coding/decoding for Full Rate, Enhanced Full Rate and AMR traffic channels,

• up to four Low Speed Access RC modules (LSA-RC) which:

—are identical to BSC LSA RC modules,

—can manage up to 21 external E1 PCM (or 28 T1) links each.

3-39


Common Equipment Module 1 - Signaling Processing

Aterinterface

A interface

LAPD

SS7 TS SS7 TS

BSCMSC

PCM links

PCM links

CEM

64 kbit/s

LSA RC

PCMController

HDLCController

LSA RC

PCMController

HDLCController

TranscodingNode

CallProcessing

LAPD links established between the TCU and the BSC (on the Ater) are used for both OA&M and Call Processing functions located on the CEM:

• OA&M: management of the TCU under the control of the BSC:

—downloading and configuration, from BSC local disk,

—supervision: event reports are sent to the OMC-R through BSC.

• Call Processing: specific treatments performed by the TCU for each call, are initiated by the BSC:

—choice of the voice algorithm,

—Ater and A Time Slots to be used.

LAPD links are:• switched by the switching matrix of the CEM, coming from ATM-SW,

• processed by the HDLC Controller (up to four links) located on an LSA-RC module,

• carried on the Ater PCM TSs:

—Call Processing: one TS per LSA-RC module,

—O&M: one TS per TCU node.

SS7 Time Slots are simply switched through the switching matrix without transcoding process.

3-40


Vocoders

Common Equipment Module 2 - Information Switching and Processing

TranscodingNode

a a

4 3 2 1

BSC

MSC4 3 2 1

Speech or Data Switching

CEM

64 kbit/s

TRM

When the TCU 3000 receives a command to establish a communication of a given type on a given A interface circuit, it performs the connection between the A interface circuit, the appropriate transcoding resource and the Ater interface circuit.Thanks to this capability, it is not necessary for the MSC to manage A interface circuit pools. Speech flow carried on Time Slots is transcoded by TRM module voice coders so called vocoders. Each concentrated TS (a) to/from BSC is processed by the TRM module.Data flow are only adapted from 8 or 16 kbit/s to 64 kbit/s.Each processed TS (1), (2), (3), (4) is switched by the switching matrix of the CEM module to/from the MSC on A interface.

3-41


Transcoder Resource Module

Archipelago = 3 Islands



Island=

5 DSPs

TRM

DSP: Digital Signal ProcessorMLB: MaiL BoxPPU: Pre-Processing UnitSPU: Signal Processing Unit

SPU

SPUPPU SPU

SPU

VocodersDSPs

Frame synchronizationHandovers ….

ProcessorPower QUICC

MLBIsland

=5 DSPs

Island=

5 DSPs

S-LinksInterface

FR, EFR , AMR or TTY



The Transcoder Resource Module or TRM, performs the GSM transcoding function. The TRM supports 216 vocoders:

• Full Rate, Enhanced Full Rate (EFR) and AMR, voice coding/decoding,

• up to 14.4 kbit/s data rate.

A TRM contains one Processor (Motorola Power QUICC) and 45 DSPs (Motorola DSP 311), organized in three identical archipelagos, each of which can be assigned dynamically to a particular type of vocoder: FR, EFR, and AMR (from V14). Each archipelago is made of one MaiL Box DSP and three DSP islands.Each island consists of five DSPs:

• 1 PPU (Pre-Processing Unit) DSP managing frame synchronization, handovers, etc.,

• 4 SPU (Signal Processing Unit) DSPs managing the vocoding (six vocoders).

The TRM is provisioned in an N+1 load sharing redundancy scheme.A TCU 3000 sub-rack (Transcoding Node) can contain up to 12 TRM modules.The allocation of the vocoders, based on a dynamic process, is the result of a real-time adjustment, starting at the initialization of the TCU.When there are two or more types of vocoder to manage, the operator has to define for each TCU 3000 node the minimum capacity associated with each type of vocoder, in terms of number of communications to process. During this process, the TCU may have to modify the initial partitioning, in order to satisfy a larger number of requests than planned for a specific coder.If the operator wants the TCU 3000 to perform dynamic resource allocation, he needs to configure the minimum required capacity for each vocoder so as to leave some transcoding resources in the “free pool”.

3-42


Transcoder Resource Module 2 - TRM2 Introduction

SLIFS TRM2TD

MB

US

LHP

BU

S

common archipelago

POWER JEDI JTAG ITM Clock Drivers

SDRAM CTRLBILL

FLASH

SDRAM

QUICC

ARCHIPELAGO 1

ARCHIPELAGO 2

ARCHIPELAGO 3

MLB

DIA

PPU

PPU

PPU

SPU SPU SPU SPU

SPU SPU SPU SPU

SPU SPU SPU SPU

POWER

TRM2 HW presentation:• The TRM2 board is composed of 3 archipelagoes. Each one will be dedicated

to a codec type (FR, EFR, AMR, EFR_TTY)

• NTQE08BA for TRM2

TRM2 capacity = 33-60% more than TRM capacity

Operations with TRM2• TRM2 boards can be introduced in a TCU 3000 shelf in any of the places

available for the current TRM boards.

• SW Compatibility V16 & above.

• Mixed configurations with TRM1 and TRM2 in one TCU 3000 is authorized.

• TRM2 is ROHS compliant.

3-43


Vocoders

Vocoders

Internal PCM S-Link Allocation

BSC

MSC

S-link

S-link

S-link

S-link

S-link

S-link

TRM

TRM S-link

S-link

S-link

S-link

S-link

S-link

TranscodingNode

PCMController

LSA RC

PCMController

LSA RC

S-link

SwitchingMatrix 64K

CEM

64 kbit/s

PCMController

LSA RC

Any board of the Transcoding Node uses a three S-links connection.

3-44


Maintenance Trunk Module Bus

BSC 3000 TCU 3000

The Maintenance Trunk Module bus runs along the backplane.The MTM bus is a five wire multi-drop bus used to ease communication of test and maintenance commands or data between a system test/maintenance control module and up to 250 modules.Only one module, (the active OMU for the BSC), is assigned mastership of the bus, and is responsible for conducting MTM bus transactions. All other modules within the system are slave to the test bus, but have the capability of initiating communication to the master through the MTM bus interrupt capabilities.The ITM ASIC in direct control of the MTM bus and located on each transition module can be configured to operate as either an MTM bus master or a MTM bus slave interface device. Its outputs to the backplane are “open drain type”, so that failure of a power supply does not jeopardize the integrity of the whole bus.

3-45


Shelf Interface Module Power Distribution

Module

PCIU

SIM A

SIM B

Backplane

PUPS

PUPS

+3.3 V

+3.3 V

1

0

Shelf

A Feed

B Feed

Module

Each shelf has two Shelf Interface Modules, but one can supply all the modules (28). The two SIMs provide for the shelf:

• power supply (-48 V) EMI filtered,

• power switching (30 A) with soft start circuitry,

• CEM/PCIU alarm interfaces,

• craftsperson access.

In the case where a SIM module needs to be extracted (repair or upgrade), it is necessary to switch off the module and to disconnect the power feed on the faceplate.

3-46


Service Area Interface 1 - Overview

CTB

CTMx1CTMx2CTMx3CTMx4CTMx5CTMx6

CTU

CTU

CTU

CTU

CTU

CTU

CTU

CTMx7

Cable Transition Unit

LSA RC module

TIM

IEM

Active

IEM

Passive

Rx cables

Tx cables

RC Mini backplane

Service Area Interface

CTU

The Service Area Interface comprises seven CTU (Cable Termination Unit) modules which provide the physical interface between the LSA RC modules and the customer’s spans. Each CTU is associated with one LSA RC module and includes:

• one CTB (Cable Transition Board), equipped to mate the backplane with seven CTMx,

• seven CTMx (Cable Transition Module) which provide the following functions:

— terminate the cables that connect to the TIM board (LSA-RC) via CTB,

— provides connectors for terminating customer A and Ater PCMs,

— provide secondary surge protection, manual loopback switches, and passive electronics for impedance matching for PCM30 Coax connections.

The CTMx is available in three styles:• CTMC (PCM30 Coax) which provides three E1 PCMs,

• CTMP (PCM30 twisted Pair) which provides three E1 PCMs,

• CTMD (DS-1 twisted pair) which provides four T1 PCMs.

The CTU numbering, and the linking between the LSA and the CTU must respect the following principles:

• The operator must easily find the CTU corresponding to a LSA, in order to connect the LSA to the CTU.

• The operator must find the CTU associated to a LSA when the LSA is displaying span error on its faceplate (connection/loopback operation of the corresponding CTM).

3-47


0 1 2

3 4 5

6 7 8

12 13 14

15 16 17

18 19 20

9 10 11

Port Number on E1 - CTU

2 3

6 7

10 11

18 19

22 23

26 27

1

5

9

17

21

25

13 14 15

Port Number on T1 - CTU

0

4

8

16

20

24

12

Service Area Interface 2 - CTU Connection

On a BSC 3000 the PCM numbering is the following:LSA-RC number * 21 + (CTU Port Number) for E1 PCMsLSA-RC number * 28 + (CTU Port Number) for T1 PCMs

On a TCU 3000 the PMC numbering towards A Interface is:LSA-RC number * 21 + (20 – CTU port number) for E1 PCMsLSA-RC number * 28 + (27 – CTU port number) for T1 PCMs

For Ater PCMs, the connection is updated in the lsaPcmList parameter of LSA-RC object at OMC-R.

LSA-RC number

Slot Number

LSA-RC Number

Slot Number

LSA-RC number

5 0 5 0103 1 103 1109 2 109 2112 3 112 313 42 5

TCU 3000BSC 3000

3-48


Service Area Interface 3 - BSC

CTU#0

CTU#1

CTU#2

CTU#3

CTU#4

CTU#5

SAI

ATM

RM

ATM

RM

LSARC

LSARC

LSARC

LSARC

CEM

0C

EM 1

8k R

M8k

RM LSA

RC

LSARC

0

1 2 3

45Interface

Node

3-49


Service Area Interface 4 - TCU

CEM1

CEM0

LSARC

1LSARC

2LSARC

3

LSARC

0

CEM1

LSARC

1LSARC

2LSARC

3

LSARC

0

CTU#0

CTU#1

CTU#2

CTU#3

CTU#4

CTU#5

SAI

CTU#6

CTU#7

UpperTranscoding

Node

LowerTranscoding

Node

CEM0

3-50

Student notes:

4-1


nortel.com/training

November, 2006411-1075A-001.1603

Data Flow Exercises

Section 4

4-2


Objectives

After this module of instruction, you will be able to draw the data paths inside the BSC 3000 and TCU 3000 for the following:

> Traffic (Circuit Switch and Packet Switch)

> GSM Signaling

> Call Process Signaling

> OA&M

4-3


Contents

> Internal BSC Dialogues> Circuit Switch/Packet Switch Path> GSM Signaling Path> BSC 3000 and TCU 3000 Dialogue

4-4


Internal BSC Dialogues

LSARC

OMUOAM

LSARC

MMSMMS

ATM RM

ATM/PCMInterface

ATM RM

ATM/PCMInterface

Control Node

Interface Node

ToBTSs

ToTCUs

TMU

TrafficManagement

TMU

TrafficManagement

TMU

TrafficManagement

ATM SW ATM SW

OAM OMU

Switching Unit

CEM

64 kb/s

8K RM

8 kb/s

LSARC

PCMController

LSARC

PCMController

On the block diagram of the Control and Interface Nodes, trace the path for internal messaging:

• between TMUs,

• between the OMU and the CEM.

4-5


Circuit Switch/Packet Switch Path

LSARC

PCMController

LSARC

PCMController

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

TCU

CEM

8K RM

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

TRM

Vocoders

CEM

Transcoding NodeBSC

Switching Unit

BTS

ATM SW

ATM RM

ATM/PCMInterface

TMU

TrafficManagement

ATM SW

MSC

PCU

On the block diagram of the BSC and TCU, trace the path for circuit switch traffic and packet switch communication.

4-6


GSM Signaling Path

LSARC

PCMController

LSARC

PCMController

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

TCU

CEM

8K RM

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

TRM

Vocoders

CEM

Transcoding NodeBSC

Switching Unit

BTS

ATM SW

ATM RM

ATM/PCMInterface

TMU

TrafficManagement

ATM SW

MSC

On the block diagram of the BSC and TCU, trace the path for BTS/LAPD and MSC/SS7 signaling.

4-7


BSC 3000 and TCU 3000 Dialogue

LSARC

PCMController

LSARC

PCMController

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

TCU

CEM

8K RM

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

TRM

Vocoders

CEM

Transcoding NodeBSC

Switching Unit

BTS

ATM SW

ATM RM

ATM/PCMInterface

TMU

TrafficManagement

ATM SW

MSC

On the block diagram of the BSC and TCU, trace the path for Call Processing dialog and Operation and Maintenance between the BSC and the TCU.

4-8

Student notes:

5-1


nortel.com/training

November, 2006411-1075A-001.1603

BSC 3000 and TCU 3000 Operation

Section 5

5-2


Objectives


> Indicate the means which are used to operate and maintain a BSC 3000 and a TCU 3000• OMC-R• TML (Local Maintenance Terminal)• RACE (Remote ACcEss Equipment)

> Briefly describe the main operations• Software download• Start up and Shut down

5-3


Contents

> Operation and Maintenance > Object Model at the OMC-R > Software Architecture > Software Downloading > Startup

5-4


Operation and Maintenance Overview

OMC-R Remote ACcessEquipment

LocalMaintenance

Terminal

RACE TML

Operation and Maintenance

The BSC 3000 takes advantage of its high processing power, to perform many O&M tasks in parallel: for example it takes in charge the software upgrade of all its BTSs, once it gets the full software loaded from the OMC-R. It can download the software of up to 100 TRXs simultaneously, hence decreasing considerably the upgrade duration or the necessary time to bring back into service the whole BSS network after a cold restart.The hardware and software architecture of the BSC 3000 and TCU 3000 (one-to-one links between hardware modules, supervision software, supervision activity of passive modules) allow precise and immediate fault detection (both hardware and software failures).The simplicity of the hardware architecture allows the BSC to detect very precisely any hardware fault at a module level. Each hardware module is replaceable unit and is hot-insertion: when it has been detected as faulty, it can be replaced without stopping the BSC or the TCU and the new module will be automatically configured and put into service by the BSC.

5-5


Object Model at the OMC-R 1 - OMC-R/BSC Interface

Old New

BSC

BSC

The object model will converge towards the Q.3 object model of the OMC-R; in this way, the Q.3 mediation done in the OMC-R will become easier and more effective.Managed object modeling (list of objects, and their associated attributes, actions, notifications and counters), is equivalent to the one proposed on the Q.3 interface of the OMC-R Mediation Device. Main benefits:

• less mediation

— average mediation rate 4% instead of 55% network vision uniformity,

• single stream OA&M:

—design cost reduction OMC-R CM,

—BSC OA&M,

• hardware management:

—clear board identification, board restart, test triggering.

5-6


Object Model at the OMC-R 2 - bsc and transcoder Objects

Hardware

cn

mms

in

iem

ccmms omu tmu 8krm Lsacematm

Hardware

cem trmLsa*

iem

bsc

transcoder

softwarepcmCircuit

modules

boards

* = manually updated

Automaticallytriggered

New hardware objects are introduced into the OMC-R BSS Q.3 object model for each type of board or module to be managed in a BSC 3000 and TCU 3000. These objects will be used by the different OMC-R applications (configuration, fault, performance), exactly like the other Q.3 objects. For example, a fault related to a hardware module will be notified directly on the corresponding hardware object. These hardware objects will be made visible both in the “internal” Q.3 interface (MD/OMC-R) and in the “external” one (MD/NMS).The main objects are triggered automatically: bsc3GEqpt, cn and in.The LSA-RC shall be created manually by the operator at a specific position in the shelf (configuration data of the LSA-RC object).This creation results in the creation of the Resource Complex Management and the TIM:

• the IEMs follow standard plug & play module management,

• the TIM is always in the central position (x position),

• the two redundant IEM modules always surround the TIM (x-1 and x+1 position) at the OMC-R level.

All board objects are created automatically.

5-7


Object Model at the OMC-R 3 - BSC 3000 Control

BSC 3000

All hardware modules of the BSC 3000 & TCU 3000 are modeled and managed as logical objects. This allows both the BSC 3000 and the OMC-R to provide the operator with precise information and services on each individual hardware module:

• Board representation on the OMC-R GUI: The physical BSC board layout will be represented in the OMC-R GUI.

• Fault representation: A hardware problem can be tracked thanks to this new representation which allows faulty boards to be highlighted on the OMC-R GUI.

• Private Data collection: Dynamic data can be collected per boards to give theoperator specific information related to the boards/modules (Localization, Firmware identification, Inventory information).

• Maintenance actions: Actions can be performed for some boards/modules in order to prevent or to correct hardware problems (RESET BOARDS) or to trigger tests from the OMC-R.

• Performance measurement: New localization will be performed on the Q.3 and BSC/OMC interface which will significantly reduce the number of counters defined in the Q.3 interface. Thus, access to the observation report will be simplified.

5-8


Object Model at the OMC-R 4 - TCU 3000 Control

TCU 3000

Graphical view of a TCU 3000, with easy fault localization, due to module representation.

5-9


Application and Services Layer(GSM/GPRS Call Processing, OA&M, BSC and TCU services,

ADMinistration, Abis, Ater and Agprs access)

Platform Layer (Supervision, Startup, Load Balancing, Messaging)

Base OS Layer(Memory and disk access)

Hardware Abstraction Layer (Base Support Package, OS Kernel, drivers): AIX, VxWorks, VRTX

Hardware/Firmware

Software Architecture 1 - Software Layers

A basic software package provides common services to all software units:• The Application and Services layer (ASL), is a set of functional entities

providing the BSC/TCU services as GSM components: Call Processing, OA&M, Abis, Ater and Agprs access.

• The Platform layer is responsible for management of the platform: Supervision, Startup, Load Balancing, Messaging.

• The Base OS layer, is composed of the standard OS (AIX, VxWorks, VRTX) and off-the-shelf software running on the OS.

• The Hardware Abstraction Layer is responsible for making the upper layers independent with respect to the hardware; it is composed of a Base Support Package (flash), the OS kernel and the drivers required to manage the hardware.

5-10


Software Architecture 2 - BSC Software Architecture

ControlNode

InterfaceNode

OMU

GSMOA&M

Base OS AIX

Platform

CEM

SwitchMgt.

Base OS VRTX

SAPI/Base

INOA&M

8K RM

SwitchMgt.

Base OS VRTX

Base

RMOA&M

ATM RM

ATMMgt.

Base OS VRTX

SAPI/Base

RMOA&M

LSA RC

Base OS VRTX

Base

RMOA&M

TMU

GSMCall P.

Base OS VxWorks

Platform

BTSOA&M

PCUSNOA&M

TCUOA&M

TMU

GSMCall P.

Base OS VxWorks

Platform

BTSOA&M

PCUSNOA&M

TCUOA&M

The BSC 3000 Control Node software is divided in two main areas:• A "TMN front-end" area composed of the two OMUs: the software is composed of

centralized functions (OMC-R interface management, Data Base management, etc.) possibly duplicated in passive mode on the mate OMU.

• A "Traffic Management" area composed of the TMUs: the architecture is based on a scalability policy. This means that a BSC can be equipped with only one TMU with extension capability when more TMUs are provisioned (total = up to 14 TMUs). This implies a distributed software architecture to share the processing load over all the TMUs. In this way, the "GSM application" and the "Platform" layers are designed as distributed software.

The distribution criteria are closely linked to the managed objects:• the software in relation with the TCU should preferably be distributed per TCU

equipment,

• the software in relation with the PCUSN should preferably be distributed per PCUSN equipment,

• the software in relation with the BTS objects (BCF, TRX, TDMA, etc.) should preferably be distributed per BTS site,

• the software in relation with the MSC should be distributed only on software architecture criteria. In fact, the A interface objects are viewed as unmarked resources from the MSC point of view.

To achieve these goals, the "Load Balancing" and "Fault Tolerance" services provide respectively the capability to distribute the application entities over all the provisioned TMUs and the capability to protect the system from (or at least to reduce) the impact of failures.

5-11


Software Downloading 1 - BSC Downloading from the OMC-R

FTAM = File Transfer Access Management EFT = Set of files to be down loaded (Ensemble de Fichiers Transférables)

BSC

TMU

TrafficManagement

OMUOAM

ATM SW

OMC-R

FTAM

EFT Downloading

http://jjj.kk.lll... html

MMS(Disk)

TML

The BSS software (BSC, TCU and BTS) is downloaded into the BSC from the OMC-R. For each version and edition, the complete BSS software is delivered on a CDROM. This volume can be used at the OMC or TML level.It is compressed and divided into several files in order to download only the modified files between two versions and to reduce as much as possible the downloading duration. The BSC 3000 stores two versions of the BSS software. The new version will be downloaded in background without impacting BSC service. Both BSS software and BSC OS can be downloaded in background, or installed locally from the TML.There is no PROM memory on the BSC 3000 & TCU 3000 hardware module, with the exception of the ATM SW module (ATM switch).All firmware is in flash EPROM and can be modified and downloaded remotely by the system. The complete BSS software (BSC, TCU and BTS) is downloaded from the OMC-R to the BSC via FTAM. The OMC and BSC 3000 are connected through Ethernet and IP protocols. The throughput is up to 10/100 Mbit/s (Ethernet standard) if the OMC-R is locally connected to the BSC.When the BSC is remote, a minimum throughput of 128 kbps is necessary for the efficiency of OMC-BSC communication.

5-12


Software Downloading 2 - BTS and TCU Downloading

BSC

BTS

TCUCN

INBSC Disk

BSS Software

ATM RM8K-RM

LSA-RC

ATM SWTMU

PassiveOMU

ActiveOMUOA&M

LSA-RCTRM

BCFTRX

BTS downloadingThe BSC can download ten BTSs simultaneously per TMU. With ten “active” TMUs, 100 BTSs can be downloaded simultaneously. BSC 3000 support BTS Background Downloading since V16.

TCU downloadingTCU 3000 software is downloaded by the BSC 3000. It is compressed and divided into several files, in order to download only the modified files between two versions and to reduce the downloading duration as much as possible. The BSC 3000 stores two versions of TCU software.The new version can be downloaded as a background task, without impacting TCU service. The TCU software can also be installed locally from the TML. A set of LAPD connections is used for TCU management in normal operation. To download the TCU, supplementary LAPD connections must be setup. These connections pre-empt (or wait for) time-slots used for communications. A minimum of four LAPD channels can be managed per LSA module.Download of a set of files (size of about 20 Mbytes per TCU) lasts:

• with four LAPDs: about 20 minutes (requires a minimum of 2 LSAs),

• with eight LAPDs: about 10 minutes (requires a minimum of 3 LSAs).

5-13


Startup1 - BSC or TCU Cold Startup (MIB not built)

Board

Module

Control NodeBSC

This scenario applies to the C-Node modules:• OMU• TMU• ATM SW

The Hardware Startup Progress

Board Recovery

Module Recovery

Dead Office Recovery

The overall startup sequence describes how the BSC goes from its initial power-up state, with no software running, to a fully operational state where the applications are running and providing GSM service.This type of startup is called dead office recovery and first needs the entire Control Node startup sequence to be performed.The operator builds the network at OMC-R level and creates the BSC logical object. As soon as the OMC-R/BSC link is established, the BSC sends a notification indicating that a MIB build is requested.Upon receipt of this notification, the OMC-R triggers the MIB build phase:

• The MIB (Management Information Base) is built on the active OMU.

• The “Build BDA N+1” upgrade feature is provided on the BSC 3000, as in a BSC 2G.

• This phase ends with the creation of the MIB logical objects followed by the reception of a report build message.

5-14


Startup 2 - Board Startup: General Behavior

Non FTApplications

andFT creators

Fault Tolerant

Applications

Boot Local_OA&M(Software Manager)

Fault TolerantLocal Agent

Base OS

A module is said to be operational when all of its boards are operational.For each board, the startup sequence consists of three ordered steps:

• boot sequence,

• platform initialization,

• application initialization.

Application initialization covers both the creation and initialization of the GSM BSC applications, this phase is managed in accordance with the BSC configuration and available resources.Some boards are able to start autonomously, booting from non-volatile storage, whereas others must wait as they require the services of another board when operational.The Control Node is operational once application initialization has completed successfully and the BSC is operational when the Control and Interface Nodes are operational.

5-15


Startup3 - BSC or TCU Hot Startup (MIB built)

• Boards– Active OMU_SBC– Passive OMU_SBC– OMU_TM / TMU_TM– TMU_SBC– TMU_PMC– ATM SW

• Modules– OMU– TMU– ATM SW

• C-Node Startup• I-Node Startup• T-Node Startup

Since the MIB is already built, we only have to check the hardware configuration consistency. We must check that modules have not been introduced or removed when the BSC or the TCU was previously switched off. The BSC and TCU will have the same behavior as for a cold startup. The consistency between the new and the previous hardware configuration is checked out at the OMC-R level. Three cases may happen:

• A module has been extracted: the corresponding object is deleted on the MMI and in the MIB, and an alarm on the father object indicates the suppression.

• A module has been plugged into a previously-free slot: the corresponding object is automatically created on the MMI and in the MIB, and an alarm on the father object indicates the creation.

• A module has been replaced by another one:

—The object corresponding to the replaced module is deleted on the MMI and in the MIB.

—The object corresponding to the newly inserted module is created on the MMI and in the MIB.

—Alarms on the father object indicate the suppression and the creation.

5-16

Student notes:

6-1


nortel.com/training

November, 2006411-1075A-001.1603

BSC 3000 and TCU 3000 Maintenance and Enhanced Exploitability

Section 6

6-2


Objectives

> After this module of instruction, you will be able to understand the main benefits of BSC and TCU 3000 architecture for:• Fault Tolerance• Load Balancing• Overload• Software Upgrade• Hot insertion/extraction (Plug and Play)• Fault Management• Remote Access Equipment• Local Maintenance Terminal

6-3


Contents

> New Exploitability Principles > Fault Tolerance > Load Balancing> Overload > Fault Management> Software Upgrade > Upgrade and Build On Line Performances Improvements> Software Upgrade> Hot Insertion/Extraction > Fault Management > Remote ACcess Equipment RACE> Local Maintenance Terminal TML

6-4


New Exploitability Principles 1 - Redundancy

BSC Defense

TMUTMU

TMU +

N+P redundancyAutomatic reconfiguration

Active-activeredundancy

OMU A OMU B ATM SW A

ATM SWB

TMU

Hot standby

The BSC 3000 and TCU 3000 provide carrier-grade availability. All hardware modules are totally redundant, including PCM interface modules. But unlike the current BSC 12000, total duplication of all critical BSC hardware is not required, and a board failure does not entail a switch over to a whole set of passive boards. In the BSC 3000 and TCU 3000, the modules work according to one of the following three modes:

• in hot stand-by (active/passive) mode: OMU, CEM, IEM (LSA-RC module); a single faulty board has no impact on the BSC or TCU and multiple faults also have no impact, providing that one module (or IEM board) works in each pair, 8K-RM (SRT),

• in parallel (both modules are simultaneously active): ATM SW (ATM switch) + ATM RM, shared MMS + private MMS,

• in N+P mode: TMU, TRM, the modules work in load sharing, processing both active and passive processes, and P failures will preserve the nominal capacity.

The Fault Tolerance algorithm implemented in the BSC Control Node allows fast fault recovery, by reconfiguring the software activity on working modules, without impacting service.

6-5


New Exploitability Principles2 - Cell Group Concept

• BSC 2G— up to 2 CPU-BIFP boards (CPUE) dedicated to the Call Processing

— cellGroup = collection of BTS managed on the same board

• BSC 3000— up to 14 TMU modules dedicated to the Call Processing

— cellGroup = collection of BTS sites

2 cellGroup

96 cellGroup

To manage the BTS sites a new concept is introduced with the BSC 3000: the Cell Group.Each site (and all the cells and TRXs belonging to this site) is held by a Cell Group. A Cell Group (called CG) can hold several sites. The CG entity is instantiated into an active and a passive instance, which are located on different TMUs. The CG is in charge of all the Call processing related to the BTSs (Supervision of the BTS, Call processing of all the communications in these cells).The distribution of BTS sites per Cellgroup is an internal algorithm which can be only partially controlled by the operator and thus can be configured either:

• automatically and statically by the ADM application,

• by the operator from the OMC through an optional parameter (Number of estimated TRX) transmitted at the site creation.

Each Cellgroup is able to manage up to 300 Erlangs.Each TMU module is able to manage:

• an average of 300 Erlangs,

• up to 100 TRXs,

• up to 16 Cellgroups (8 actives and 8 passives).

6-6


New Exploitability Principles3 - Cell Group Management

New Site

BSC 3000

LoadBalancing

The Cell Groups are determined at boot time by the Load Balancing function, according to data associated with the cells:

• when a BTS is added to the BSC, it is added to an old or a new Cellgroupthanks to the same algorithm,

• when a cell or a TRX is added to a BTS, the corresponding Cellgroup has more load.

The redistribution of the sites into Cell Groups is a complex task, which is normally performed by the BSC by respecting the CG dimensioning rules and CG capacity objectives so defined:

• 54 CG per BSC,

• 10 sites maximum per CG,

• 18 CG per TMU,

• 75 TRXs maximum per CG,

• Maximum of 16 TRX per Cell and 48 TRX per site ( note linked to CG allocation but maximum site size in V15.1)

Due to the software links complexity, a site must be placed in a CG by a BSC at its creation, and cannot be moved to another CG after that. The only way to move a site from one CG to another one is to delete it and then to re-create it. Another possibility is to perform an on-line build (with complete service loss of the whole BSC for a few minutes).

6-7


• A table predefines the cell Erlang load from 1 to 16 TRX (by default Erlang B law, 2% blocking rate)

• estimatedSiteLoad (Class 3, object: btsSiteManager): The customer can modify the values (mErlang) in the table and set the estimated load in a cell.

Range: [0, 1100] Erlang

New Exploitability Principles 4 - Estimated Site Load Parameter

The BSC 3000 Load Balancing feature uses a table that predefines a cell Erlangload from 1 to 16 TRX: ERLANG_PER_N_TRX_CELL BSC Data ConfigThe values in milliErlang can be modified by the customer without service interruption (class 3).By default, this table is filled with the Erlang B law results (2% blocking rate).In V15.1, a possibility to set the estimated Erlang load of the site is offered with the use of the parameter estimatedSiteLoad, which is a class 3 parameter with object the btsSiteManager.This parameter is used at the site creation, to define the Erlang consumption of the new Cell Group, by setting the Erlang consumption to a different value from the one defined by the ERLAN_PER_N_TRX_CELL table.

6-8


New Exploitability Principles 5 - Fault Tolerance, Load Balancing and Overload

> Two kinds of software:Fault Tolerance entities "launched" by the FT application and supervised by FT

non FT entities

> Two kinds of software:Fault Tolerance entities "launched" by the FT application and supervised by FT

non FT entities

GSM applications may be either Fault Tolerant (FT) or non Fault Tolerant (non FT).A Fault Tolerant application is an application that is replicated.Load Balancing only applies to Fault Tolerant applications, it relies on the following FT primitives to balance FT applications between TMUs:

• CREATE, to create a passive entity,

• FLUSH, to synchronize a passive entity on an active one,

• SWACT, to switch activity from an active entity to a passive one,

• KILL, to destroy a passive or an active entity.

6-9


New Exploitability Principles 6 - BSC 3000 Support BSS Based Solution

MSCVLR

MSCVLR

SMLCSMLC

GMLCGMLCBSCBSCBTSBTS

Lh

LgAbis

A

MSMS

Um

LocationApplications

LocationApplications

HLRHLR

Lb

BSC and SMLC directly exchange BSSMAP-LE and BSSLAP messages over the Lb logical interface

Support the four location methods

In the new BSS architecture, Nortel follows the 3GPP specification concerning the Lb interface.The Lb interface is used only for LCS application and relies on SS7.There are two SS7 interfaces: A and Lb. The BSC has to manage the dialog with multi distant point codes (SMLC and MSC). This requires having a SCCP and a MTP3 layer multi SSN and multi DPC. Each interface (A and Lb) relies on one distinct physical route from the BSC.As the SMLC, the MSC and the BSC are part of the same SS7 network, the set of sccp parameters should be identical for Lb or A interface.

6-10


Fault Tolerance 1 - Fault Tolerance Software

Module #1

Module #1 Module #2

ActiveInstance Passive

InstanceCurrent context update

Fault ToleranceSoftware

Module #2

ActiveInstance

SWACTFailure

A Fault Tolerant application is an application which can survive a hardware fault. For the Control Node, this is done by having a single active instance on a given module (TMU or OMU) and having one (or more) replica instances called passive, located on a different module.The active instance of the FT application runs the application code, whereas the passive instance does not.The passive instance(s) of the FT application are simply kept “up-to-date” with the current context of the active instance.Therefore, if the hardware with an FT application is running and fails, the passive instance can take over and continue to run the application without any break in service. The previous passive instance becomes the new active instance. The process of changing a passive instance to an active instance or vice versa is called a SWitch of ACTivity or SWACT.

6-11


Fault Tolerance 2 - Example: Swact on TMU Failure

> The GSM Core Process is a set of FT applications managing a set of sites (Cellgroup):

TMG_RAD for radio resource management TMG_CNX for connection management (setup, release, assignment, HO)TMG_MES for A interface messages (paging, incoming HO)TMG_L1M for Layer 1 managementSPR for BTS site supervisionSPT for TCU supervisionTMG_RPP for PCUSN supervisionOBS for observations

> The GSM Core Process is a set of FT applications managing a set of sites (Cellgroup):

TMG_RAD for radio resource management TMG_CNX for connection management (setup, release, assignment, HO)TMG_MES for A interface messages (paging, incoming HO)TMG_L1M for Layer 1 managementSPR for BTS site supervisionSPT for TCU supervisionTMG_RPP for PCUSN supervisionOBS for observations

TMU#1

A1

TMU#2 TMU#3

P1

P2 A2

P3 A3

TMU#1

A1

TMU#2 TMU#3

A1

P2 A2 P2

P3 A3

P1 Passive Core ProcessA1 Active Core Process

P1

All processing relative to a cellgroup is executed on a single TMU. The corresponding passive (or redundant) process is executed on another TMU. In this example, there are three groups of process, each of which is composed for Fault Tolerance purposes of one active process “Ai” plus one passive process “Pi”with i as the application identifier.The three active processes are distributed over three TMUs. The passive processes related to one active process do not run on the same TMU as the active process, but on another TMU.The passive processes are directly and continuously updated by their corresponding active processes, using internal messaging.On failure of the TMU1, the Fault Tolerance algorithm performs a SWACT by “electing” the passive A1 processes as “Active”. The figure shows the new distribution of processing over the two available TMUs.A TMU managing 300 Erlangs, processes around three HandOver per second, thus less than one external HO per second. A maximum of 10 messages per second (50 bytes of payload) are exchanged between two TMUs.For connection management, each TMU exchanges up to 50 messages per second (20 bytes of payload) with the Interface Node. At swact time (failure) the messaging activity between OMUs and TMUs needs a bandwidth of from 2.5 to 4 Mbytes during 2 or 3 seconds.

6-12


Load Balancing1 - Principle

TMU#1

A1

A2

TMU#2

P2

P1

TMU#1

A1

TMU#2

P2 A2

P1

P3A3 P3A3

P4 A4A4 P4

The purpose of the Load Balancing function is to distribute processing in an optimal way between the TMUs and to use the BSC resources optimally.This is performed by distributing the processes related to the different Cellgroups(i.e. sets of cells belonging to the same process) “equally” over the TMUs. The distribution of Cell groups and redundant processes is also done automatically by the system at boot time. Load Balancing allows a redistribution of Cell groups on the TMUs, without disturbing the calls and is executed:

• when a TMU module fails or comes into operation (for hardware or operator reasons),

• when Cell groups are modified (to add BTSs),

• when an imbalance of the TMU CPU load is detected by the BSC: in this case, the load balancing can be done during non-busy hours.

6-13


Load Balancing 2 - Example: Adding a TMU

TMU#1

A2

TMU#2

P1

P2

A3P3

A1

TMU#1 TMU#2 TMU#3

P1A2

P3

A1

A3

P2

In the system, the processor load of each TMU depends mainly on the number of BTSs/cells/TRXs to manage, and the related amount of traffic. When there are modifications to a BTS configuration (addition of TRX) or to a BSC configuration (addition of TMUs) the Load Balancing service allows redistribution of the processing with the best use of the BSC resources. The chart gives an example of the use of Load Balancing when a TMU is added to the BSC.The initial configuration of the BSC is 2 TMUs, and one more is added and provisioned for traffic management:

• the BSC automatically computes a new distribution and applies it,

• the re-distribution is achieved without exposure time by:

—adding new passive members to the groups,

—swapping their activity,

—suppressing the useless passive members.

The Load Balancing operation is achieved by using the Fault Tolerance service. In fact, the redistribution of the processing is obtained by “electing” active processes with the best location distribution (best applies here to taking into account all the parameters that specify the LB criteria).This “election” leads to several SWACTs achieved by Fault Tolerance.

6-14


Overload 1 - Principles

Paging

RACH

HandOver

Locat.Update =CPU Load

Memory

OMUOAM TMU

TrafficManagement

ATM SW

CEM

64 kbps

Disk

The BSC 3000 robustness in overload conditions is ensured by a centralized overload control mechanism, which is based on the same principles as for the overload control implemented for BSC 12000 in BSS release V12.Overload management is the function that allows the system to correct sporadic peaks of load (on the system) without any loss of service on still-established communications and still covered areas. The TMU and CEM modules that can reach the overload state are monitored. The overload management is a dynamic and reactive process.Each module reports its synthetic load to the OMU, which controls globally the load state of the BSC and triggers the appropriate action according to the module in overload (TMU or CEM) and to the level of overload.There are four parameters to observe in order to be able to manage a correct overload:• cpu load,• system memory occupancy,• messaging level: queues and delays.

In nominal mode, the main peaks of load are generated on TMUs, due to Call Processing needs, and also due to the amount of other processes, for example, management of a large number of BTSs, and several TCUs. This is based on the thresholds allowed per domain of software, such as:

• GSM Call Processing,• BTS_OAM, • TCU_OAM, • Platform management.

The goal is to prevent the named entities from using more resources than those allowed by given threshold(s).

6-15


Overload2 - TMU Mechanism

Level 2 = 90%

Level 1 = 80%

overLoad levels

Level 3 = 100%

Only Traffic management operations are taken into account in this mechanism Current communications are maintained, (except for HO incoming requests above threshold 3)

50% List of messages filtered:• Paging request• Channel request (non emergency)• all first Layer 3 (non emergency)• HO request (traffic reason)• HO request (O&M reason)• Directed retry

All messages are filtered

2/3 messages are filtered

1/3 message is filtered

Hysteresis is applied at each threshold.

Time

TMU modules are relatively independent one with respect to the other in terms of overload handling. Since a TMU module manages the traffic of a group of cells, when a TMU module is in overload, it will filter partially the new incoming traffic requests related to the group of cells it manages.Three overload levels are defined for each monitored processor.For each level, some of the new traffic requests are filtered:

• level 1 (80% of processor load): traffic reduction by around 33% by filtering one request out of three of the following messages:

— Paging Request,— Channel Request (not Emergency Call),— all first Layer 3 messages (not Emergency Call),— handover for traffic reason,— handover for O&M reason,— directed retry,

• level 2 (90% of processor load): traffic reduction around 66% by filtering two requests out of three of the above messages,

• level 3 (100% of processor load): no new traffic is accepted by filtering all previous and following messages:

— all first layer 3 messages,— all Channel Requests,— all handover indications,— all handover requests.

6-16


Fault Management1 - Impact on Service in the Control Node

What happens when a module is down:

No impact on traffic

No disturbance of traffic or slight delay

The corresponding OMU module is down

Swact on the other OMU, no traffic impact

1

TMU

TMU

TMU

ATM

SW

ATM

SW

TMU

TMU

TMU

TMU

SIM

B

TMU

TMU

MM

S pr

ivat

eM

MS

shar

ed-F

iller

--F

iller

-

MM

S pr

ivat

eM

MS

shar

ed

TMU

TMU

TMU

SIM

A

OM

U

TMU

OM

U

TMU

1 3

21 4

-Fill

er -

-Fill

er -

2

3

4

Control Node behavior in the case of a faulty module.

There is no impact on the traffic in the case of:• faulty OMU,

• faulty ATM-SW.

In the case of a faulty MMS:• Private MMS: the corresponding OMU will be lost,

• Shared MMS: no effect on traffic or OAM.

In the case of a faulty TMU:• loss of the communication being established and being handed over, no more

duplex mode until passive processes are reestablished,

• if there are not enough TMU modules to handle the processes, then we may loose traffic.

6-17


Fault Management 2 - Impact on Service in the Interface Node

ATM

RM

ATM

RM

LSARC

LSARC

SIM

ALSARC

LSARC

CEM

0C

EM 1

8k R

M 0

8k R

M 1

-Fill

er - LSA

RC

SIM

B

-Fill

er -

-Fill

er -LSA

RC

-Fill

er -

What happens when a module is down:

The connections can be sent with delay to the C-node

No disturbance on traffic

No more communication

The connections are switched over to the second IEM

1 2

1

1

2

3

4

3 44

Interface Node behavior in the case of a faulty module.

There is no impact on traffic in the case of:• a faulty 8K-RM.

There will be delay in the order of connections in the case of:• a faulty ATM-RM,

• a faulty CEM.

In the case of a faulty LSA-RC:• short-cut of the signal on a PCM, there is no impact on signaling, in the case of

a faulty IEM,

• loss of all PCM connections on the corresponding LSA-RC, in the case of a faulty TIM.

6-18


Fault Management3 - Impact on Service in the Transcoder Node

SI

M

L S A

CEM

CEM

Fille

r

Fille

rWhat happens when a module is down:

The connections can be sent with delay to the I-node

No signaling disturbance, but light problems on the telephony can appear

The connections are down

No impact on traffic, swact on other IEM 4 4

1

3

TRM

TRM

TRM

TRM

TRM

TRM

TRM

TRM

TRM

SI

M

2

TRM

TRM

TRM

L S AL S A

L S A

1

2

3

4

Transcoding Node behavior in the case of a faulty module.

In case of a faulty CEM:• loss of communications being established, for the active processes on the

corresponding CEM.

In case of a faulty TRM:• slight loss in voice quality,

• no impact on signaling.

In case of a faulty LSA-RC:• short-cut of the signal on a PCM, but there is no impact on signaling, in the

case of a faulty IEM,

• loss of all PCM connections on the corresponding LSA-RC, in the case of a faulty TIM.

6-19


Software Upgrade 1 - Overview

Software upgrade from version N to version N+1:Uses a Zero Downtime mechanism based on the replicated architecture involved by Fault Tolerance and Load BalancingThe operator does not have to move on site and only one person must be able to control remotely the whole upgrading sequence from the OMC-R.

A new version and edition of software can be downloaded remotely without any operational impact, only modified files in the new version are downloaded.Before any upgrade procedure, the equipment (BSC/TCU) must check its hardware (flash memory checksum).The execution of the upgrade is ordered by the OMC-R and controlled by the BSC (in the OMU module), after the complete transfer of new files. Only the modules that have modified software are downloaded again.

The first phase of the upgrade software can be made a long time before the upgrade of a module. This allows the upgrading data to be transferred to the MIB (Managed Information Base) located in the ”shared” disk located of the control node. This operation is done when the BSC 3000 is working without any service disturbance (except bandwidth reduction.)Then, the control node sends upgrade orders to the CEM module that manages the upgrade of the concerned module itself, without breakdown of the services that are running.

6-20


V15

OMU_P OMU_A

CC1_1 CC1_2

TMU_1 TMU_2 TMU_N

IEM_A

IEM_P

IEM_P

IEM_A

IEM_P IEM_A ATM_1 ATM_2

8K_A 8K_P

CEM_P CEM_A

V15.1

CN

IN

UPGRADE TYPE 4

Upgrade and Build On Line Performances Improvements (1/5)

In the current behavior, the CN is upgraded first and then IN upgrade is triggered. This serialization of CN/IN upgrades was chosen to prevent interoperability issues. In particular, this serialization prevented having an IN in N+1 release that interact with a CN in N release.This serialization of CN and IN upgrades can be alleviated provided that interoperability is no more an issue when CN and IN are in heterogeneous releases: CN in N release and IN in N+1 release. The IN upgrade will be triggered as soon as the OMUs, CC1s and the first TMU have been upgraded successfully. New CC1 upgrade behaviour ( previously no check done on the ATM-RM status)

• Check that both ATM_RM are enable/on-line before

• Beginning the upgrade

• Upgrading CC1

ATM-RM new behaviour• In V15.1 release, The ATM-RM is expected not to reset when it detects a loss

of signal on the OC3 fiber.

6-21



• Active OMU Application restart instead of OMU reset• Save AIX start up and disk mounting

• New MIB Activation Protocol:• Clear config request * to IN/TCU instead of reset

Target Downtime reduction ( Half of V14.3)

* clean-up of all resources on IN/TCU)

BUILD ONLINE

In V14.3, the complete restart of the BSC is triggered by upgrade control node manager that sends a control reset request to hardware management. Upon reception of this control node reset request, hardware management resets first the TMUs, the CC1, the passive OMU and finally the active OMU.Actually, there is no need to reset the active OMU; only applications need to be restarted to load the new data from the new MIB. This saves the AIX startup latency and shared disks mounting. The average AIX star-up latency is around 4 minutes. For that purpose, upgrade CN sends a control node restart to hardware management that triggers a backplane control node restart.The backplane control node restart triggers the following actions:

• Stop all applications on the active OMU

• Reboot of the OMU_TM of the active OMU

• Restart all applications on the active OMU

• Check the shared disk

The clear config req impacts the upgrade control node manager that must not reset the IN/TCU before resetting the control node during the activation of the new MIB.

6-22



• OMU application restart used to restart the CN instead of resetting it during an Upgrade type 6 & 7

• Save AIX start up and disk mounting

• Clear config request* to TCU instead of reset

• New OMU flash upgrade protocol• Replace CN reset by an OMU restart• Parallelize multiple tasks

* clean-up of all ressources on TCU)

UPGRADE OFF LINE

The omu application restart will be used to restart the control node instead of resetting it during an upgrade type 6 or type 7. This OMU application restart requires to reset the active OMU at the end of the offline upgrade making sure that low level deliveries are loaded on the active OMU. This OMU reset must be synchronized with load balancing and IN events.The clear config can be leveraged during an offline upgrade to gracefully restart the TCU instead of resetting it. For that purpose, the upgrade control node manager will send a CLEAR_CONFIG_REQ to TCU instead of a RESET_REQUEST.This way the TCU will be ready to be configured very shortly. Note that the IN must still be reset since control node and IN are upgraded at the same time.A new OMU flash upgrade protocol has been proposed to decrease significantly the upgrade offline downtime. This protocol relies on the OMU application restart to shorten the latency of OMU flash upgrade. Precisely, this new protocol replaces each control node reset by an OMU restart. Furthermore, this protocol enables also to parallelize multiple tasks that were previously serialized.

6-23



OMU Active Startup

config IN

Config TCU

2min30

Downtime before first call: ~ 5min

IN Critical Path

2minTCU Clear Config

1min

Active CP Startup

1min30

OMU Passive Startup

2min30

Passive CP Startup

1min30

1min30

3min

2min

1min

Downtime before full duplex: ~ 7min

V16 BSC Start up chronogram during an offline activation

Main improvement: • OMU passive startup is postponed at the end of the control node startup

concurrently to the startup of the passive core processes among TMUs.

• IN critical path duration decreases to two minutes (see section 4.7.1). This enables the IN critical path latency to overlap entirely with the active OMU startup duration. Note that the requirement differs for IN and TCU critical path duration improvement. IN critical path must not exceed 2 minutes whereas TCU critical path can be a little bit longer without any impact on the overall BSC down time.

• Core processes are started-up concurrently on different TMUs.

6-24



Offline upgrade TCU without having to lock the TCU

BSCOMC TCUTGE backgroundTcuUpgrade (TCU e3, TC3vveeddpp.LIV, offline,…)

Upgrade offline req

2034:begin

ftp load to flash

Upgrade ack with reset

Init_dialog_req

Init_dialog_ack

PCM configuration

Updates TCU S/W links to new version

Upgrade offline req2034:begin

2024:cleared2034:END

2024:cleared2034:END

Upgrade ack w/o resetStartup Upgrade

2024:cleared

TCU auto reset

Fuzzy Period

All Boards Flash Updatedwith new N+1 release

SharedDisk

Check On upgradeconditions OK

RGE OK

Currently, the TCU offline upgrade protocol requires to lock the TCU prior to activate the upgrade. This TCU lock incurs a very long interruption of service because the offline upgrade includes the TCU boards flash download via the LAPD channels;hence through a limited bandwidth. Recent performances measurements have shown that the TCU software download is longer than IN software download by an order of magnitude.

A major improvement is to trigger an offline upgrade TCU without having to lock the TCU prior the offline upgrade activation; hence keep the TCU in service during the software download.

6-25


Software Upgrade2 - OMU Software

Software upgrade may only impact some parts of the software entities:

base OS (AIX),platform (OAM, FT, messaging, LB, …),applications (supervision, performance and fault management, software downloading, lapd and SS7 management, ....

The previously active OMUbecomes passive

is reset and boots with the new software version

OMU#1 OMU#2Active OMU Passive OMU

A1 P1

A2 P2

A3 P3

Base OS (N) Base OS (N)

Platform (N) Platform (N)

Applications (N) Applications (N) The passive OMUis reset and boots with new software

and becomes active

Upgrading of the passive OMU can be separated into two phases: • application software upgrade,

• AIX upgrade.

The master module of each Node is upgraded first: OMU (C-Node) and CEM (I-Node or T-Node).Application software upgrade:

• The passive OMU is reset and boots with the new software version.

• When the passive OMU has entirely recovered and correctly updated, the OMU performs a swap and the new active OMU runs the new softwareversion.

• The new passive OMU is then reset to boot on the new version.

AIX upgradeAIX is the operating system of the BSC 3000, hosted in the private disks (MMS).The upgrade of AIX is very different if it is a new version or an update. When a complete re-installation is required, the private disk of the OMU has to be erased and re-written. This is done via the other OMU, which acts as boot server, and takes between half an hour and one hour. During the installation the OMU is not bootable and the BSC is in a phase without OMU redundancy.AIX updates are made with “file-sets” which can be installed online. A reboot of the OMU may be necessary. If so, it is done on the passive OMU.

6-26


Software Upgrade 3 - TMU Software: Principle

1° TMU is isolated2° TMU reset and boots with new

software3° TMU joins the group to retrieve the

processes it hosted previously

TMU#1

A1

P4

N+1

TMU#2

A2

P1

N

TMU#3

A3

P2

NTMU#4

P3

A4

N

TMU#3

A3

P2

NTMU#1

A1

P4

N

TMU#2

A2

P1

N

TMU#3

A

P2

NTMU#4

P3

A4

N

TMU#4

P3

A4

N

A1

TMU#2

A2

P1

NP4

TMU#1

Isolation+

reset

The TMU upgrade is the most complex, because call processing is managed by TMUs in real time during the upgrade, these modules are in N+P “load sharing”redundancy and furthermore, the upgrade is performed without any interruption of service.The advantage of redundancy during a software upgrade is to manage “N” and “N+1” versions together during transient states of the system with minimal risk. The two software versions, N and N+1 are assumed to be fully compatible. The upgrade is always executed concurrently with GSM traffic management remaining active. TMU modules are upgraded one by one as follows:

• One TMU is relieved of all its processes so that active processes and passive processes are supported entirely by the other TMUs.

• When isolated, the TMU resets and boots on the new software version: the TMU flash is rewritten at this time.

• Once recovered, the TMU (N+1 version) joins the TMU group (N version) to retrieve the applicative processes it hosted previously to the upgrade.

6-27


Software Upgrade 4 - TMU Software: Upgrade Wave

TMU#1

A1

P4

N+1

TMU#2

A2

P1

N

TMU#3

A3

P2

NTMU#4

P3

A4

N

TMU#3

A3

P2

NTMU#4

P3

A4

N

TMU#4

P3

A4

N

TMU#1

A1

P4

N+1

TMU#2

A2

P1

N+1

TMU#1

A1

P4

N+1

TMU#1

A1

P4

N+1

TMU#2

A2

P1

N+1

TMU#3

A3

P2

N+1

TMU#2

A2

P1

N+1

TMU#4

P3

A4

N+1

TMU#3

A3

P2

N+1

Thanks to N+P replication, no downtime should occur during this upgrade wave

To maintain the traffic management activity during the upgrade, the upgrade is performed by “waves”, by one set of boards at a time:

• First, all the traffic is transferred to TMUs that are in version N.

• The other TMUs are isolated (the size of the wave is a configuration parameter).

• The isolated TMUs are upgraded (software downloading, initialization, etc.).

• To avoid service interruption, passive members are first created on the newly upgraded boards.

• Finally, activity is transferred to them.

• During the period of coexistence of the two releases, some restrictions may apply depending on the compatibility level between both versions: no handover between N and N+1 area, etc..

6-28


Software Upgrade5 - CEM or RM Software Upgrade (ATM-RM, 8K-RM, IEM)

> Software upgrade may impact only some parts of software entities:

base OSplatform (OAM, messaging, …)applications (software downloading, etc.)

Active CEM or RM Passive CEM or RM

A1

A2

A3

Base OS (N) Base OS (N)

Platform (N) Platform (N)

Applications (N) Applications (N)

P1

P3

P2

For the CEM modules and the RMs, with the following redundancy factor: 1+1, the upgrading of this protection group is done as follows:

• loading of the software packages is running inside the passive RM or the passive CEM module,

• a SWACT is running between:

—the passive CEM module and the active CEM module,

—the active RM and the passive RM.

6-29


Software Upgrade 6 - TRM Software Upgrade

Final Step: The newsoftware is downloading on each TRM

TRM1

N+1

TRM2

N+1

TRM6

N+1

TRM8

N+1

TRM7

N+1

TRM9

N+1

TRM10

N+1

TRM3

N+1

TRM4

N+1

TRM5

N+1

STEP 1• Soft blocking on

the first module• Load sharing on

each other TRM modules

• Load N+1 release on TRM1

TRM1

N

TRM2

N

TRM3

N

TRM4

N

TRM5

N

TRM6

N

TRM7

N

TRM8

N

TRM9

N

TRM10

N

TRM1

N+1

TRM2

N

TRM3

N

TRM4

N

TRM5

N

TRM6

N

TRM7

N

TRM8

N

TRM9

N

TRM10

N

TRM2

N

TRM3

N

TRM4

N

TRM5

N

TRM6

N

TRM7

N

TRM8

N

TRM9

N

TRM10

N

TRM1

N+1

TRM2

N+1

TRM3

N

TRM4

N

TRM5

N

TRM6

N

TRM7

N

TRM8

N

TRM9

N

TRM10

N

STEP 2• Soft blocking

on the second module

• Load sharing on each other modules

TRM1

N+1

STEP 3 to Final Step

For the TRM with the following redundancy factor: N+P (P=1), the upgrading of the protection group is done as follows:

• a “soft blocking” is sent to the TRM concerned,

• the new communications are distributed to another TRM,

• when the communications in progress inside the TRM concerned are accomplished, then the software upgrading is done.

6-30


Hot Insertion/Extraction 1 - Overview

Hot module insertion or extraction without service interruption

Automatic or half-automatic plug and play

configuration capability

Easy hardware maintenance or extensionby simply extracting or plugging modules

The hardware modules of the BSC 3000 have a hot insertion and extraction capability. This means that a hardware module can be replaced or added in the equipment without shutting down the machine even partly and without any impact on service.Furthermore, the BSC 3000 offers “plug & play” (or auto discovery) capability both for equipment startup and for module hot insertion.The modules are automatically detected, started and configured allowing an easy and efficient maintenance of BSC 3000 and TCU 3000 hardware equipment. The BSC 3000 and TCU 3000 report information about their hardware configuration automatically to the OMC-R. Because of this architecture, the “hot plug & play” feature does not apply to the LSA-RC module: TIM and RCM boards are not involved. Module extractionWhen a module is extracted, a notification is sent to the OMC-R: this notification is a state change to “disabled/notInstalled” of the object that was previously in the slot. On reception of this state change, the OMC-R deletes the corresponding logical object and removes it from the HMI and the MIB.An alarm is generated at OMC-R level on the father object to indicate that a module has been removed.Hot extraction of the module can be performed without any tools, but the OMU and MMS modules requires an operator action on the frontface pushbutton, using a pencil.

6-31


7° Creation TGE and RGE

2° HardwareInsertionevent

Hot Insertion/Extraction 2 - Hot Insertion Procedure

6° ObjectCreationNotification

7°Board appearance(Local Manager)

3°MOD and Q.3 logical identifiers allocation

4°Instance Storage

5°Spontaneouscreation on Q.3

1°Module insertiondetection

8°Moduleeventreportingbegin

BDE MIB(BDA)

MODQ.3Local or ExternalManager

MD-R BSC/TCU 3000

TGE = Transaction Globale d’ExploitationRGE = Réponse Globale d’Exploitation

Module insertionC-Node (Control Node) and I-Node (Interface Node) objects are automatically created when the user creates the BSC 3000 object on the OMC-R.The Platform sends notifications indicating the hardware configuration. This hardware configuration is detected on the corresponding platform object (C-Node, I-Node, LSA or T-Node).This information is stored on the MMS disk and sent to the OMC-R. It can be read on the MMS disk, even when a module is out of service.The information is also stored at OMC-R level and can be displayed upon operator request.Module hot insertion may be described as follows:

• module insertion by the craftsperson,

• hardware detection and BIST,

• front panel LED state depending on BIST results,

• verification by the craftsperson that the LED state is correct,

• hardware detection notification including BIST results sent towards the OMC-R,

• the module is created at the OMC-R and is displayed on the HMI.

6-32


Fault Management 1 - Remote Maintenance Capability

> Technical status:• Hardware references• BIST (Built-In Self Test results)• Hardware faults...

> Reset/Switch-over Remote tests:

• on-line tests• off-line tests

Easy maintenance platform:• The network operator can remotely trigger module reset or switch-over.• He can also trigger on line or off-line test from its network management center.• The network operator has a permanent technical status at network management center level.

BSC 3000 & TCU 3000 hardware management from the OMC-R is based on the hardware detection capability of the new generation platform. All faults concerning the components of an object are reported to the OMC-R.The FM application is hierarchically structured: the processor, module, I-Node, C-Node, BSC and each level of the OA&M function are able to detect, analyze, filter and react to a fault if their level is able and authorized to manage this fault, because of the potential system complexity of the fault.For example, an ATM fault detected between the ATM SW and TMU modules can not be corrected directly by the TMU/OA&M application, but only by the Control Node/OA&M application located on the OMU module.The OMU is the FM master module for the Control Node and the BSC 3000, it stores the fault events on circular files and sends them to the OMC-R.The CEM is the FM master module for the Interface Node. Two kinds of information are sent by the Interface Node in the case of equipment failure:

• the state changes treated by the I-Node/OA&M application,

• the details of the fault, forwarded to the OMC-R for maintenance purpose.

There are two levels of fault:• faults that do not impact the availability of the object: failure of an IEM (LSA-RC), a

CEM, an ATM or an 8K-RM,

• faults that make the object unavailable: failure of both cards or modules, failure on a TIM of an LSA-RC.

In the case of hardware failure, a craftsperson needs to repair the failure by changing the faulty module.

6-33


Fault Management 2 - On Board Inventory Information

Fast detection

Fix without any service interruption

Reliable diagnostic

Report with accurate identification

TMU boardversion xx

shelf 2slot 3

serial xxxxxTM

U

TMU

TMU

ATM

SW

ATM

SW

TMU

TMU

TMU

TMU

SIM

B

1 2 3 4 5 6 7 8 11 12 13 14 15

TMU

TMU

MM

SM

MS

-Fill

er -

-Fill

er -

MM

SM

MS

TMU

TMU

TMU

SIM

A

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

9 10

OM

U

-Fill

er -

TMU

-Fill

er -

OM

U

2

2

TMU

The faults events sent to the OMC-R contain all the necessary information for supervision and maintenance: type of fault, criticality, service impact, impacted hardware. Hardware failures are notified directly on the related hardware module, so that the OMC-R can display the failed equipment precisely to the operator.On board inventory information for the equipment (BSC 3000 and TCU 3000):

• Physical location,

• Site,

• Unit,

• Floor,

• Row Position,

• Bay Identifier.

For the FRUs (Field Replaceable Units):• Serial number (Corporate Standard 5014.00 compliant),

• Module Name (generic name of the module family),

• Module type (PEC code = product engineering code),

• Hardware release,

• Hardware position (shelf, slot).

6-34


Fault Management 3 - LED of all Modules in BSC 3000 and TCU 3000 (except MMS modules)

(*): indicates that a board must be flagged for replacement or for any other reason, in order to avoid errors by the maintenance staff

Red Green Meaning

Not powered

BIST running

Module is active

Module is passive

Alarm state

Path finding (*)

All BSC 3000 & TCU 3000 modules have the same two LEDs on the upper part of the front face of each module to facilitate on-site maintenance and to reduce the risk of human error. This table gives the description, combinations and states of the red LED and the green LED for each module (except the MMS module) inside the BSC 3000 cabinet and the TCU 3000 cabinet.

6-35


Fault Management 4 - LED of MMS Modules in the BSC 3000

Red Green Meaning

The MMS module is not powered

The MMS module is locked. It is not operational (disk is updating or stopping)

The disk is operational and updated (unlocked)

Alarm state

Path finding: the MMS module can be removed

The MMS module is not managed or not created

Read/Write operation on the disk

This table gives the description, combinations and states of the red LED and the green LED for the MMS modules in the BSC 3000 cabinet.The round yellow led is blinking on the disk to indicate, read/write operation.

6-36


Remote ACcess Equipment RACE1 - HTTP/RACE Server on an OMC-R WorkStation

IP NetworkIntranet/ Internet

Terminals Server

OMC-R Server

OMC-RSite

BSC3000

BSC Site

BTSS4000/S2000E

BTSS8000S12000S18000

BTS Site

ETHERNET1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

RACETMLTML

RACE

ModemModemModem

Modem

Modem

RACE Server

PSTN

Firewalls

Modem

RACE

TML

The Remote ACcess Equipment offers a Web interface to the OMC-R.It provides the users with a convivial interface similar to the one of Graphic MMI and all the functionality of the ROT is available on this new feature.The RACE was first developed to replace the ROT, but it could also be used as a particular OMC-R WorkStation.The advantages of this new product are the following:

• The interface is user-friendly, it is close to the interface of the OMC-R. Thus, the tool is easy to manipulate for the user used to the OMC-R interface.

• Compared to the ROT, which has been developed with tools that are now obsolete, the RACE is implemented using new technologies, object oriented.

• The RACE is able to ensure a secure access to the network, which was no longer guaranteed with the ROT.

• Thanks to the Web-oriented conception, operations and maintenance of radio subsystems can be done from a remote site without requesting an OMC-R on-site operator:

—by using PSTN and any kind of secured connection system,

—via BTS or BSC equipment using the BSC-OMC link within the BSS,

—through LAN.

6-37


Remote ACcess Equipment RACE 2 - Overview

HTTPserver

MmiWWW Kernel

Web browser

OMC-R

Server

Real time information

RACE Client OMC-R WorkStation/RACE server OMC-R Server

This new application is composed of Web pages and Java applets that can be run through a Web navigator (Netscape or Internet Explorer).This new application is adapted to individual operator needs: when the operator must work from home, or when operations from BTS or BSC sites are required.A better presentation of the data allows the customers to save time: for instance, an operator had to modify a list of parameters and could make a mistake:

• with the ROT, it was mandatory to re-enter all the information,

• with the RACE, using the “Back” button of the navigator, he just has to modify the wrong parameters.

The unique requirement to let this feature run is to have a Web browser, which brings two advantages:

• all data are stored on the server and are downloaded at connection, so the installation of a RACE client is done very quickly and then there is almost no upgrading to be provided on the client side,

• the operator can use a PC to connect to the OMC-R; such an OMC-R station is cheaper than a Unix station.

Finally the RACE can run on either an OMC-R WorkStation or an OMC-R server, with a standard Internet browser for Unix.

6-38


Physicalpath

Manager

HardwareManager

ATMManager

S/WBus

HTTPServer

Local Maintenance Terminal TML 1 - Overview

LAN TestManagement

TML BSC 3000

BSC 3000

HTMLJAVA

Testserver

InterfaceNode

access

The Local Maintenance Terminal or TML (Terminal de Maintenance Locale) application is a java applet stored in the BSC disk.The TML hardware is a PC: it works under Windows and behaves like a Java browser.The TML can be connected to the BSC OMU through Ethernet connections.The TML can be plugged onto a hub that can be hosted in the SAI of the BSC 3000.The TML interface is independent of the BSC 3000/TCU 3000 software evolutions.The TML allows a first BSC 3000/TCU 3000 installation to be performed.It allows the customization parameters of the BSC 3000/TCU 3000 to be read and modified:

• BSC number,

• IP address,

• PCM type, etc..

The configuration information on the different hardware modules can be read from the TML:

• board identification and states,

• software version,

• software and patch markers.

6-39


Local Maintenance Terminal TML2 - Principle

TML PC 3000 Platform

http://mmm.ii.jjj.kk/BSC3000. html

Download html pageand Java applet

Try connectionSend USER and PASSWORD

Send commands

Receive answers

HTMLJAVA

WEBBrowser

TMLApplication

HTTPserver

TestServer

Using a web browser, the TML operator loads an HTML page (through HTTP) holding the TML applet. The TML applet is then downloaded to the TML PC using the HTTP server.Once the TML software is loaded in the TML PC, it is possible to start a test session. The messages exchanged between the TML and the BSC are done through a TCP/IP connection.The TML communicates with the “Test server” software module.The TML accesses the MIB for:

• modification of commissioning data:

—OMC-R link definition (IP, direct, …),

—PCM trunk setup,

—physical location definition (name, floor),

• checking software and hardware marking information.

6-40

Student notes:

7-1


nortel.com/training

November, 2006411-1075A-001.1603

BSC 3000 and TCU 3000 Provisioning

Section 7

7-2


Objectives

After this module of instruction, you will be able to:

> Provision a BSC 3000 and a TCU 3000 Cabinet

> Define • TMU number in the Control Node of a BSC 3000 Cabinet• LSA-RC in the Interface Node of a BSC 3000 Cabinet and of a TCU

3000 Cabinet

> Define TRM number in Transcoding Nodes of a TCU 3000 Cabinet

7-3


Contents

> BSC 12000HC and BSC 3000 Comparison> TCU 2G and TCU 3000 Comparison> BSC 3000 Provisioning > Mixed TMU1/TMU2 Configurations> TRM2 Dimensioning> BSC 3000 Provisioning > BSC 3000 and TCU 3000 Configurations

7-4


36 / 032 / 2TCU 2G / TCU 3000

BSC 12000HC and BSC 3000 Comparison

BSC 12000HC (6 cabinets )

BSC 3000 vs BSC 12000HC BSC 3000 and TCU 3000 (2 cabinets)

600/120600/1201000/205Floor load (kg/m2_ lb/ft²)540/11881620/3564570/1254Weight (kg/lb)

W: 156/61 D: 60/23 H: 200/78

W: 468/182 D: 60/23 H: 200/78

W: 96/37 D: 60/23 H: 220/86Cabinet dimension (cm/in )

1.75.12.0Power consumption (kW)

126037803112A circuits

48/48144/144126/168E1/T1 links

61816SS7 links40120567LAPD links138414500BTS160480600Cells3209601000TRX120036003000Erlang

For one BSC 12000HCFor three BSC 12000HC For one BSC 3000Maximum values

12 / 0

Flexibility: there will be no predefined and limited number of configurations as for the BSC 12000HC.A set of product configurations fitting the needs of a given customer closely, in terms of processing, signaling, PCM connectivity can be delivered, given that these configurations remain within the minimum and maximum product configurations.For example, if a 2000 Erlang BSC with the same PCM and signaling connectivity is delivered to two customers having networks with different traffic profiles, the number of TMUs in each BSC can be different, say 8 TMUs for one and 10 for the other. A BSC 3000 of maximum capacity is ‘equivalent’ to three BSC 12000.

7-5


TCU 2G and TCU 3000 Comparison

600/120600/1201000/205Floor load (kg/m2_ lb/ft²)

283/6238490/18678570/1254Weight (kg/lb)

W: 78/30 D: 60/23 H: 200/78

W: 78/30 D: 60/23H: 200/78

W: 96/37 D: 60/23 H: 220/86

Cabinet dimension (cm/in)

1302.0Power consumption (kW)

120 for E1 links92 for T1 links

3780 for E1 links2760 for T1 links

1944A-circuits for E1/T1 links

1+430+12020+67 for E1 links24+89 for T1 links

Ater + A for E1/T1 links

For one TCU 2GFor thirty TCU 2GFor one TCU 3000Maximum values of

TCU 2G (30 shelves)

BSC 12000HC (6 cabinets)

TCU 3000 vs TCU 2G

BSC 3000 and TCU 3000 (2 cabinets)

7-6


BSC 3000 Provisioning 1 - BSC 3000 versus BSC 2G

N + P

1 + 1

1 + 1

OAM

ATM SW

OMU

TMU

TrafficManagement

300 E

Control andSwitchingChain BPassive

Control andSwitchingChain AActive

Total nbr of TMUs

GSM Object

SITE (w/ 1 LAPD Channel) 240 300 300 360 420 480 500 500 500 500120 150 150 180 210 240 270 300 300 30080 100 100 120 140 160 180 200 200 200

Not standard

13 14

SITE (w/ 2 LAPD Channel)SITE (w/ 3 LAPD Channel)

6 71 2 3 4 5 128 9 10 11

N+P Redundancy

The redundancy concept of the BSC 3000 / TCU 3000 is different from the 2G BSC/TCU. We are not speaking any more about two redundant chains, but about a per module or per card (LSA) redundancy.Except for the TMU module for which an N+P redundancy is implemented, all the modules are 1+1 redundant.Entire Call Processing of the BSC 3000 is based on the TMU module and the dimensioning for this module is based on the estimated traffic load (maximum 300 Erlang per TMU). The estimated traffic for each site is calculated by the BSC 3000 by taking in account the sum of each cell traffic (based on the number of TRX per cell). The BSC 3000 estimates alone the number of TMU needed to reach the capacity required by the Site/Cells/TRX configured by the OMC. If the calculated number of TMUS is less than the installed TMUs, then the BSC notifies to OMC via Load Balancing Anomaly how many TMUs are needed in order to reach the required capacity.Due to the fact that in case of the BSC 3000 there are no more fix configuration (like type1...5 in case of BSC 2G), the OMC-R verify only the maximum dimensioning of the BSC 3000.

7-7


BSC 3000 Provisioning2 - Number of TMUs

– N is the minimum number of TMUs to run all active processes

– P is the minimum number of TMUs needed to run all the passive processes.

– 2 is the number of TMUs needed to run SS7 active and passives processes.

CapacityNumber of TMU

600Erlang

900Erlang

1200Erlang

1500Erlang

1800Erlang

2100Erlang

2400Erlang

2700Erlang

3000Erlang

MP

SS7

Total

212

5

312

6

412

7

512

8

622

10

822

12

922

13

1022

14

722

11

In order to have a well balanced processing load between TMUs, two mechanisms have been implemented:

• The BSC 3000 makes a site distribution per Cell Group using a special algorithm so that a number of equally charged Cell Groups can be obtained:

—a Cell Group is a logical entity containing several sites,

—all the Cells and the TRXs belonging to one site are in the same Cell Group.

• The Cell Groups are then distributed over the existing TMUs. A TMU is capable of managing up to 16 Cell Groups (8 active and 8 passive). The active Cell Groups from one TMU will have their passive instance on another TMU.

Concerning TMU redundancy, let us define the following: • M is the minimum number of TMUs needed to run active processes: (Fault

Tolerance).

• P is the minimum number of TMU needed to run passive processes.

• 2 is the number of TMUs needed to run active and passive SS7 processes.

Taking into account these considerations, the BSC 3000 capacity can be defined.

7-8


LAPD link Dimensioning:

• Up to 567 LAPD channels per BSC 3000

• 62 LAPDs per the TMU module (2 for TCU LAPDs)

• Engineering recommendation = 40 BTS LAPDs per TMU

• Dependence between the number of TMUs and LAPDs:

BSC 3000 Provisioning 3 - Abis LAPD Channels

Total number of TMUsGSM Object 1 2 3 4 5 6 128 10 11

LAPD Channel (Abis) Not standard 120 180 300 360 420 480

13 14

540 567

7

240

9

300

LAPD link dimensioning:• Up to 567 LAPD channels can be configured at the BSC 3000 level.

• 62 LAPD can be handled by each TMU module (2 are reserved for TCU LAPDs).

• The dependence between the number of TMUs and LAPDs that can be handled by the BSC 3000 is presented in the table.

• The BSC 3000 engineering recommendation is 40 BTS LAPDs per TMU: this engineering margin will provide enough processing capacity on the TMU or the GPRS LAPDs.

7-9


BSC 3000 Provisioning 4 - TMU2

998776655Total

222222222SS7

111111111P

665443322N

3000 Erlang

2700 Erlang2400 Erlang2100

Erlang1800 Erlang

1500 Erlang

1200 Erlang900 Erlang600 ErlangNumber of TMU2 vs Capacity

987654Total

222222SS7

111111P

654321N

3000 Erlang

2625 Erlang

2100 Erlang

1575 Erlang

1050 Erlang525 ErlangNumber of TMU2 vs Capacity

> Configurations allowed

> LAPD dimensioning

567550440330220110Not supportedLAPD Channel

987654321Total TMU2 number

TMU2 objective is to have a capacity being 1.75 times the current TMU1 one in terms of Erlangs processing capabilities and twice in terms of signaling (LAPD/SS7) ports offered. This means that the maximum BSC 3000 Erlangscapacity may be reached with only 9 TMU2 (7 instead of 12 for GSM call processing applications and 2 for SS7 management).

7-10


Mixed TMU1/TMU2 Configurations

TMU1 TMG

TMU2TMG

1 2 3 4 5 6 7 8 9 10

1

2

3

4

5

6

7

8

9

10

11

0

00 NA 600 900 1200 1500 1800 2100 2400 2700 3000

525 825 1125 1425 1725 2025 2325 2625 2925 3000

1050 1350 1650 1950 2250 2550 2850 3000 3000

1575 1875 2175 2475 2775 3000 3000 3000

2100 2400 2700 3000 3000 3000 3000

2625 2925 3000 3000 3000 3000

3000 3000 3000 3000 3000

3000 3000 3000 3000

3000 3000 3000

3000 3000

3000

3000

Erlang Capacity vs TMU1 and TMU2 number (SS7 TMU and redundant TMU not taken into account)

300 Erlang redundancy



825 Erlang redundancy + additional TMG TMU for redundancy or to support harder call profiles



> Erlang Capacity Vs TMU1 & TMU2 number

As TMU2 capacity is bigger than TMU1, dimensioning rules regarding the number of needed TMUs for a chosen target erlang capacity are modified.

7-11


Mixed TMU1/TMU2 Configurations> LAPD Capacity Vs TMU1 & TMU2 number

TMU1 TMG

TMU2TMG

1 2 3 4 5 6 7 8 9 10

1

2

3

4

5

6

7

8

9

10

11

0

0108 162 216 270 360 420 480 567 567

110 160 218 272 326 380 470 530 567 567

220 270 328 382 436 490 567 567 567

330 380 438 492 546 567 567 567

440 490 548 567 567 567 567

550 567 567 567 567 567

567 567 567 567 567

567 567 567 567

567 567 567

567 567

567

567

LAPD Capacity vs TMU1 and TMU2 number (SS7 TMU and redundant TMU not taken into account)







As TMU1/TMU2 mix configuration is supported, a faulty TMU board can be replaced by a board from a different type. However, as board capacity depends on its type, it has to be checked about the overall BSC capacity (for example, if a faulty TMU2 is replaced by a TMU1, BSC capacity will decrease in terms of supported erlangs and offered number of LAPD ports).The type of a TMU installed in a Control Node cabinet is given in an explicit way by the result of a “Display Marker” action.

7-12


TRM2 Dimensioning

Dimensioning figures:• FR archipelago capacity: 96 circuits (vs 72 for TRM board)• EFR archipelago capacity: 96 circuits (vs 72 for TRM board)• AMR archipelago capacity: 96 circuits (vs 60 for TRM board)• EFR_TTY archipelago capacity: 84 circuits (vs 48 for TRM board)

Thus the capacity of a TRM2 using three FR, EFR or AMR codec will be 288 circuits.

Nb of TCU shelves/BSC

Nb of TRM2 w/o

redundancy)

Nb of TRM2 (with redundancy)

Nb of LSAs E1

Nb of voice

channels

Nb of SS7 channels

Nb of Ater

LAPD

Capacity (Erl)

1 1 1+1 1 288 3 2 2471 2 2+1 2 576 4 4 5211 3 3+1 2 864 5 4 7981 4 4+1 3 1152 7 6 10781 5 5+1 3 1440 9 6 13591 6 6+1 4 1728 9 8 16411 7 7+1 4 1944 16 8 1923

Dimensioning for configurations without any EFR_TTY code configured

The dimensioning rules, regarding the number of needed TRM boards in a TCU 3000 cabinet, take into account the TRM capacity in terms of maximum number of terrestrial circuits that can be managed.

7-13


BSC 3000 Provisioning 5 - GPRS Impact

BSC 3000V15

PSPDN

HLR

MSC

AMAP-D

VLRPSTN/ISDN

SGSN GGSN

MSC - Mobile Switching CentreVLR - Visitor Location RegisterHLR - Home Location RegisterBSS - Base Station SystemEIR - Equipment Identity Register

PCUSN - Packet Control Unit Support NodePSPDN - Packet Switched Public Data NetworkSGSN - Serving GPRS Support NodeGGSN - Gateway GPRS Support Node

Ater

BTS

PCUSN

SMSC EIR

TCU 3000V15

Agprs

Abis

GPRS entails no BSC capacity decrease in terms of processing. In other words, the processing power of the TMU and of the other processing boards is not a limiting factor for GPRS dimensioning. Only the PCM connectivity (Abis + Ater + Agprs) and the circuit switching capacity of the BSC 3000 have to be taken into account for GSM and GPRS network engineering.In urban areas, the BSC 3000 has enough PCMs available so that the GPRS introduction can be done without any PCM dimensioning constraints. For example a maximum capacity BSC 3000 managing a BSS network made mainly of S444 BTSs, will need around 90 PCMs for Abis and Ater, out of 126. Therefore, whatever the GPRS profile is, there will be enough additional PCMs available for Agprs.In rural areas (BTS S111 & S222), all PCMs might be used for voice service only. The introduction of GPRS can then impact the BSC 3000 capacity in terms of the number of managed BTSs & TRXs.The maximum circuit switching capacity of the BSC 3000 (2268 64-kbit/s circuits) shall be taken into account in the dimensioning of a voice + GPRS network. The switching capacity is not a limitation for voice-only and for low-speed GPRS services (CS1/CS2).For high-speed data services, since the radio time-slots carrying those services require more circuits on Abis and Agprs (2 to 4 times more than for voice and low-speed packet data), the BSC 3000 switching capacity limit can be reached for some network configurations, especially for high data penetration (for example 8 radio TS per cell for GPRS).The impact on BSC 3000 capacity in terms of the number of managed TRX has to be determined on a case-by-case basis, according to the network configuration.

7-14


BSC 3000 and TCU 3000 Configurations 1 - Min and Max Configurations

163SS7 links3112620A interface circuits (BSC 3000)1944200A interface circuits (TCU 3000)

84/11221/28E1/T1 PCM (TCU 3000)126/16842/56E1/T1 PCM (BSC 3000)

567120LAPD links600360Cells500120BTS

1000360TRX3000600Erlang

MaxMinBSC 3000 and TCU 3000 dimensioning

This table gives the dimensioning factors for the BSC 3000 & TCU 3000 in minimum and maximum configurations.

BSC configuration• The minimum configuration is a 600 E, which translates to 3 TMUs (2+1 for

redundancy) and 2 LSAs (42 E1 or 56 T1 PCMs).• The maximum configuration is a 3000 E, which translates to 12 TMUs (10+2

for redundancy) and 6 LSAs (126 E1 or 168 T1 PCMs); the TCU function will require two Transcoding nodes.

Between these two configurations, all configurations can be offered, never less some product engineering rules are defined to avoid inconsistency between the number of TMUs and the number of LSAs.

TCU configuration• The minimum configuration is a 200 E TCU 3000, which translates, in the case

of Enhanced Full Rate, to 2 TRM modules (1+1 redundant) and 1 LSA (21 E1 or 28 T1 PCMs).

• The maximum configuration is a 1800 E: up to 11 TRMs (10+1 redundant) and 4 LSAs in each of the 2 nodes of a TCU cabinet. The TCU 3000 cabinet can be connected to the same BSC or to 2 different BSCs.

Note: The TCU 3000 can have a maximum of 12 TRMs modules if required.Between these minimum and maximum configurations, all configurations can be offered. Nevertheless, in the TCU 3000 the number of TRMs and the number of LSAs are directly related to the required A interface capacity.

7-15


BSC 3000 and TCU 3000 Configurations 2 - BSC 3000 and TCU 3000 Typical Examples

63/843

7+11200E

105/140480

51+18+2

2400E

84/1124

9+11800E

126/168567

61+1

10+23000E

42/5621/28Nb of E1/T121LSA

3+11+1TRM600E200ETCU 300063/8442/56Nb of E1/T1

300120Nb of LAPD32LSA

1+11+1TMU SS75+12+1TMU Traffic

1500E600EBSC 3000

Nortel Networks will define some market model configurations (rural, semi-urban, urban, etc.) and some optional extension kits (comprised of TMU, TRM, LSA) in order to satisfy most of the product configurations required by customers:

• a rural type of configuration, with a relatively low number of TMUs (because the traffic capacity is low) and a maximum number of LSAs (because many small BTSs used for coverage need to be connected),

• an urban type of configuration, with a high number of TMUs (high traffic capacity) and a relatively low number of LSAs (because BTSs have many TRXs per cell, and there are relatively few BTSs to be connected to the BSC).

Market models and market packages are defined both to optimize the end-to-end supply chain from the order to the delivery of the products to the customer, and to satisfy most of the configurations requested by the customers. The market packages allows a market model to be modified by adding extension kits, to fit as closely as possible to the customer request.

7-16

Student notes:

8-1


nortel.com/training

November, 2006411-1075A-001.1603

Exercises Solutions

Section 8

8-2


Objectives

> After this module of instruction, you will be able to understand the main data flows for BSC 3000 and TCU 3000:• Traffic (Circuit switch and Packet switch)• GSM Path Signaling• Call Processing Signalization• OA&M

8-3


Contents

> Internal BSC Dialogues> Traffic (Circuit and Packet Switch) Path> GSM Signaling Path> BSC 3000/TCU 3000 Dialogue

8-4



LSARC

OMUOAM

LSARC

MMSMMS

ATM RM

ATM/PCMInterface

ATM RM

ATM/PCMInterface

Control Node

Interface Node

ToBTSs

ToTCUs

TMU

TrafficManagement

TMU

TrafficManagement

TMU

TrafficManagement

ATM SW ATM SW

OAM OMU

Switching Unit

CEM

64 kb/s

8K RM

8 kb/s

Plane 1Plane 2

LSARC

PCMController

LSARC

PCMController

Two paths are established simultaneously using the two planes.

8-5



LSARC

OMUOAM

LSARC

MMSMMS

ATM RM

ATM/PCMInterface

ATM RM

ATM/PCMInterface

Control Node

Interface Node

ToBTSs

ToTCUs

TMU

TrafficManagement

TMU

TrafficManagement

TMU

TrafficManagement

ATM SW ATM SW

OAM OMU

Switching Unit

CEM

64 kb/s

8K RM

8 kb/s

Plane 1 Plane 2

LSARC

PCMController

LSARC

PCMController

Two paths are established simultaneously using the two planes.

8-6


PCU

Traffic (Circuit and Packet Switch) Path

LSARC

PCMController

LSARC

PCMController

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

TCU

CEM

8K RM

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

TRM

Vocoders

CEM

Transcoding NodeBSC

Switching Unit

ATM SW

ATM RM

ATM/PCMInterface

TMU

TrafficManagement

ATM SW

MSC

BTS

8-7


GSM Signaling Path

LSARC

PCMController

LSARC

PCMController

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

TCU

CEM

8K RM

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

TRM

Vocoders

CEM

Transcoding NodeBSC

Switching Unit

ATM SW

ATM RM

ATM/PCMInterface

TMU

TrafficManagement

ATM SW

MSC

BTS

Full TS LAPD signaling LAPD signaling on ATM

BTS LAPD signaling

8-8


GSM Signaling Path

SS7

LSARC

PCMController

LSARC

PCMController

TCU

CEM

64 kb/s

TRM

Vocoders

Transcoding Node

BSC

BTS MSC

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

8K RM

8 kb/s

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

CEMSwitching Unit

ATM SW

ATM RM

ATM/PCMInterface

8K RM

8 kb/s

TMU

TrafficManagement

ATM SW

MTP1 & MTP2: TMU AMTP3 & SCCP: TMU B

A B

For SS7 signaling, always two TMUs are involved.One TMU will manage MTP1 and MTP2 layer.The second TMU will manage the following layer of SS7 Signaling (MTP3, SCCP).

8-9


BSC 3000/TCU 3000 Dialogue

LSARC

PCMController

LSARC

PCMController

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

TCU

CEM

64 kb/s

8K RM

8 kb/s

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

TRM

Vocoders

CEM

64 kb/s

Transcoding NodeBSC

Switching Unit

BTS

ATM SW

ATM RM

ATM/PCMInterface

TMU

TrafficManagement

ATM SW

MSC

Same TMU

OA&M and Call Processing (1/2)

8-10


BSC 3000/TCU 3000 Dialogue

LSARC

PCMController

LSARC

PCMController

LSARC

PCMController

OMUOAM

Control Node

TMU

TrafficManagement

TCU

CEM

8K RM

ATM RM

ATM/PCMInterface

LSARC

PCMController

Interface Node

TRM

Vocoders

CEM

Transcoding NodeBSC

Switching Unit

BTS

ATM SW

ATM RM

ATM/PCMInterface

TMU

TrafficManagement

ATM SW

MSC

Different TMUs

OA&M and Call Processing (2/2)

8-11

Student notes:

8-12

Student notes:

Nortel BSC3000 userguide

Documents

bsc tcu

trademarks of nortel

nortel networks logo

property of nortel networks

training purposes only1

training purposes onlyvisit

training purposes onlynortel

training purposes only411