1

Presentation Confidential between MTN and audience

Inside Mobile Internet(GPRS) Network

Presented by Mustafa Golam

Agenda

• GPRS Core/Access Network• Network Architecture• Protocols and Procedures• Node Functions• GPRS Access Channels Details

What is GPRS?• It provides packet radio access for mobile GSM and TDMA users. • GPRS is a migration step toward 3G networks. • GPRS allows network operators to implement an IP-based core architecture for data applications, which will continue to be used and expanded for 3G services for integrated voice and data applications. • The GPRS specifications are written by the ETSI, the European counterpart of the ANSI.

Goals– Open architecture– Consistent IP services– Same infrastructure for different air interfaces– Integrated telephony and Internet infrastructure– Leverage industry investment in IP– Service innovation independent of infrastructure

Benefits– Overlays on the existing GSM network to provide high-speed data service– Always on, reducing the time spent setting up and taking down connections– Designed to support bursty applications such as e-mail, traffic telematics, telemetry, broadcast services, and web browsing that do not require detected connection

GPRS Applications

GPRS Architecture• GPRS is a data network that overlays a second-generation GSM network. • This network provides packet data transport at rates from 9.6 to 171 kbps. • Multiple users can share the same air-interface resources simultaneously.• GPRS attempts to reuse the existing GSM network elements as much as possible, but

– To effectively build a packet-based mobile cellular network, some new network elements, interfaces, and protocols for handling packet traffic are required. • GPRS requires modifications to numerous network elements.

GSM Cellular Network Structure

The Service Areas are divided into PLMNs. Each operator providing GPRS services has its own PLMN. A PLMN is identified by the Mobile Country Code (MCC) and the Mobile Network Code (MNC).

An RA consists of one or more cells. One SGSN can handle several RAs. An MS may move between RAs with the SGSN area without roaming to a new SGSN. The size of an RA can range from a part of a city to an entire province, or even a small country.

A cell, either a GSM cell or a WCDMA Systems cell, is the smallest geographical unit in the GPRS service area. It is the basic unit of a mobile cellular network and is covered by one RBS.

An SGSN area, which consists of one or more RAs, is the region served by the same SGSN. It is either a GSM SGSN area, a WCDMA Systems SGSN area, or both (in a dual access SGSN). An SGSN area does not have to coincide with an MSC/VLR area.

GPRS Logical Network Architecture

MS BTS BSC

RNCBTSMS

SGSN GGSN

SGSN GGSN

MSC

SCP

HLR OSS

Service Networ

k

Service Networ

k

Gs

SMS

Um Abis Gb

Ge

Gn

GomGr

Gn

Gn

Gn

Gn

Um I IuGi

Gi

Gd

GSM

WCDMA

GPRS Network Elements

GPRS Subscriber Terminals• New terminals are required

– because existing GSM phones do not handle the enhanced air interface or packet data.

• A variety of terminals can exist, including – a high-speed version of current phones to support high-

speed data access, – a new PDA device with an embedded GSM phone, and – PC cards for laptop computers.

• These terminals are backward compatible for making voice calls using GSM.

GPRS BSS• Each BSC requires the installation of one or more PCUs and a software upgrade. • The PCU provides a physical and logical data • The BTS can also require a software upgrade but typically does not require hardware enhancements.• When either voice or data traffic is originated at the subscriber terminal,

– it is transported over the air interface to the BTS, and from the BTS to the BSC in the same way as a standard GSM call. – However, at the output of the BSC, the traffic is separated;

• voice is sent to the mobile switching center (MSC) per standard GSM, and • data is sent to a new device called the SGSN via the PCU over a Frame Relay interface.

GPRS Support Nodes• In the core network, the existing MSCs are based on circuit-switched central-office technology and cannot handle packet traffic. • Two new components, called GPRS support nodes (GSNs), are added:

– Serving GPRS support node (SGSN)– Gateway GPRS support node (GGSN)

GPRS Network Nodes

MSCHLR/AuCEIR

BSC

BTS

Um

PSTNNetwork

GSM & (E)GPRS Network ArchitecturePCU

EDAPGb

Gateway GPRSSupport Node(GGSN)

Charging Gateway (CG) Local

AreaNetwork

Server

Router

Corporate 1

Server

Router

Corporate 2

Datanetwork(Internet)

Datanetwork(Internet)

Billing System

Inter-PLMNnetwork

GPRSINFRASTRUCTURE

BorderGateway (BG)

Lawful InterceptionGateway (LIG)

GPRSbackbonenetwork(IP based)

Serving GPRSSupport Node(SGSN)

SS7Network

PAPU

Existing MTN NG Packet Core Network

Exiting GSN Topology

Existing Gn Interfaces

Existing Gi Interface Topology

Existing Gom Interface Topology

Radius Servers External Physical Topology

Online Gateway External Physical Topology

DNS Connectivity Overview

Logical Connectivity of Gn/Gp Interface After IP Redesign

Gp Topology After IP Redesign

Serving GPRS Support Node (SGSN)• The SGSN delivers packets to mobile stations (MSs) within its service area. • SGSNs send queries to home location registers (HLRs) to obtain profile data of GPRS subscribers. • SGSNs detect new GPRS MSs in a given service area, process registration of new mobile subscribers, and keep records of their locations inside a predefined area. • The SGSN performs mobility management functions such as handing off a roaming subscriber from the equipment in one cell to the equipment in another. • The SGSN is connected to the base station subsystem through a Frame Relay connection to the PCU in the BSC.

SGSN – Serving GPRS Support Node

SGSN handles the communication with MSs and the establishment of the connection between an MS and the PDN. Unless SGSN pool is used, an SGSN serves all GSM or WCDMA Systems (or both if it is a Dual Access SGSN) subscribers physically located within the geographical SGSN area. It forwards IP packets between all GPRS attached MSs within that SGSN service area and the GGSN. Connections between the SGSN and the MS and between the SGSN and the GGSN are handled through session management, that is, through the activation, modification or deactivation of PDP contexts.

Functionalities at a glance:

• Serves MSs in SGSN area• Mobility Management functions, e.g• Update Location, Attach, Paging,..• Security and access control:• Authentication, Cipher setting, IMEI Check...• Routing / Traffic-Management • Collecting charging data• Realises Interfaces: Gn, Gb, Gd, Gp, Gr, Gs, Gf• Controls subscribers in its service area (SLR)

SGSN - Functionality

• Routing Area Update (Location Registration)• Location Update for CS • Attach / Detach• P-TMSI allocation / reallocation• Authentication• Paging (PS & CS)

• Routing Area Update (Location Registration)• Location Update for CS • Attach / Detach• P-TMSI allocation / reallocation• Authentication• Paging (PS & CS)

Mobility Management

• PDP context activation• PDP context modification• PDP context cancellation

• PDP context activation• PDP context modification• PDP context cancellation

Session Management

Gateway GPRS Support node (GGSN)• GGSNs are used as interfaces to external IP networks such as the public Internet, other mobile service providers' GPRS services, or enterprise intranets.• GGSNs maintain routing information that is necessary to tunnel the protocol data units (PDUs) to the SGSNs that service particular MSs. • Other functions include network and subscriber screening and address mapping. • One or more GGSNs can be provided to support multiple SGSNs.

GGSN - Functionality

GGSN – Gateway GPRS Support NodeThe GGSN forwards uplink and downlink IP packets between the SGSN and the PDN. It also collects charging

information for each MS, for example, usage of GPRS network resources. Service Aware Charging & Control (SACC) enables the GGSN to charge based on different price levels. The GGSN handles session management, that is, activation, modification, and deactivation of PDP contexts for sessions between the GGSN and the SGSN, and between the GGSN and the PDN. Session management also includes dynamic IP address allocation & QoS negotiation. Security functionality can also be enabled in the GGSN, like firewall-filters, policies, etc to control payload.

Functionalities at a glance:

• Gi-,Gn-Interface: Interworking PLMN PDN• Routing Information for attached GPRS user• Screening / Filtering• Collecting charging data

GPRS Overview

                                                                                                            

Figure 1   GPRS Overview for GSM

                                                                                                            

Figure 2   GPRS Overview for WCDMA Systems

GPRS Terminals• The term terminal equipment is generally used to refer to the variety of mobile phones and mobile stations that can be used in a GPRS environment. • The equipment is defined by terminal classes and types.• Three classes of GPRS terminals are provided:

– Class A– Class B– Class C

Class A Terminals• Class A terminals support GPRS and other GSM services (such as SMS and voice) simultaneously. • This support includes simultaneous attach, activation, monitor, and traffic. • Class A terminals can make or receive calls on two services simultaneously. • In the presence of circuit-switched services, GPRS virtual circuits are held (i.e., placed on hold) instead of being cleared.

Class B Terminals• Class B terminals can monitor GSM and GPRS channels simultaneously but can support only one of these services at a time. • Therefore, a Class B terminal can support simultaneous attach, activation, and monitor, but not simultaneous traffic. • As with Class A, the GPRS virtual circuits are not disconnected when circuit-switched traffic is present. Instead, they are switched to busy mode. • Users can make or receive calls on either a packet or a switched call type sequentially, but not simultaneously.

Class C Terminals• Class C terminals support only sequential attach. • The user must select which service to connect.• Therefore, a Class C terminal can make or receive calls from only the manually selected (or default) service. • The service that is not selected is unreachable.• The GPRS specifications state that support of SMS is optional for Class C terminals.

GPRS Device Types (1/5)• In addition to the three terminal classes, each handset has a unique form (housing design). • Some of the forms are similar to current mobile wireless devices, while others will evolve to use the enhanced data capabilities of GPRS.

1. The earliest available type is closely related to the current mobile phone. – These are available in the standard form with a numeric keypad and a relatively small display.

GPRS Device Types (2/5)• PC cards are credit card-sized hardware devices that connect through a serial cable to the bottom of a mobile phone. • Data cards for GPRS phones enable laptops and other devices with PC card slots to be connected to mobile GPRS-capable phones.• Card phones provide functions similar to those offered by PC cards without requiring a separate phone. • These devices may require an ear piece and microphone to support voice services.

GPRS Device Types (3/5)• Smart phones are mobile phones with built-in voice, nonvoice, and Web -browsing services. • Smart phones integrate mobile computing and mobile communications into a single terminal. • They come in various form factors, which may include a keyboard or an icon drive screen.

GPRS Device Types (4/5)• The increase in machine-to-machine communications has led to the adoption of application-specific devices. • These black-box devices lack a display, keypad, and voice accessories of a standard phone.• Communication is accomplished through a serial cable. • Applications such as meter reading utilize such black-box devices.

GPRS Device Types (5/5)• Personal digital assistants (PDAs), such as the Palm Pilot series or Handspring Visor, and handheld communications devices are data-centric devices that are adding mobile wireless access. • These devices can either connect with a GPRS-capable mobile phone via a serial cable or integrate GPRS capability.• Access can be gained via a PC card or a serial cable to a GPRS-capable phone.

Mobility Management in Detail• MM States for GSM

MM States for GSM – Contd..

IDLEAn MS in the IDLE state is not attached. Therefore, the SGSN can neither locate or reach it, and no MM context exists for the subscriber. After a successful MS Attach procedure, the state is changed to READY.

READYAn MS in the READY state is attached, and an MM context and possibly a Packet Data Protocol (PDP) context exists for the subscriber. A signaling procedure or payload transfer is ongoing or has recently ended. The current location information is on cell level, that is, the location is known by the SGSN with an accuracy of the serving cell, and no paging is required to reach the MS. A ready timer defines the time the MS remains in READY state after a packet data transfer. When the timer expires, the state is changed to STANDBY_REACHABLE. After a successful MS Detach procedure, the state is changed to IDLE.

STANDBY_REACHABLEAn MS in the STANDBY_REACHABLE state is attached. The location information is on RA level, that is, no information on cell level is available and paging is required to reach the MS. A mobile-reachable timer defines the time the MS remains in STANDBY_REACHABLE state. When the timer expires, the state is changed to STANDBY_NOT_REACHABLE. After a successful MS Detach procedure, the state is changed to IDLE. Uplink payload transfer or a signaling procedure initiated by the MS will change the state to READY.

STANDBY_NOT_REACHABLEAn MS in the STANDBY_NOT_REACHABLE state is attached. The last known location information is on RA level, that is, no information on cell level is available. Contact is resumed as soon as the MS sends packet data and the state is changed to READY. If the MS sends no packet data, an implicit detach timer defines the time the MS remains in STANDBY_NOT_REACHABLE state. When the timer expires, the MS is implicitly detached, that is, the MM context is deleted, and the state is changed to IDLE. After a successful MS Detach procedure, the state is changed to IDLE. A signaling procedure will change the state to READY.

Data RoutingData routing requirements in GPRS network can be divided into two main categories:

1. Data packet routing and

2. Mobility management

Data Packet Routing• The main functions of the GGSN involve interaction with the external data network. • The GGSN updates the location directory using routing information supplied by the SGSNs about the location of an MS. • It routes the external data network protocol packet encapsulated over the GPRS backbone to the SGSN currently serving the MS. • It also decapsulates and forwards external data network packets to the appropriate data network and collects charging data that is forwarded to a charging gateway (CG).

Fig: Routing of Data Packets between a Fixed Host and a GPRS MS

Figure: Details• Three routing schemes are illustrated:

1. Mobile-originated message (path 1)• This path begins at the GPRS mobile and ends at the Host

2. Network-initiated message when the MS is in its home network (path 2)• This path begins at the Host and ends at the GPRS mobile

3. Network-initiated message when the MS roams to another GPRS network (path 3)• This path is indicated by the dotted line

• In these examples, the operator's GPRS network consists of multiple GSNs (with a gateway and serving functionality) and an intra-operator backbone network.

Inter-Operator Roaming• GPRS operators allow roaming through an inter-operator backbone network. • The GPRS operators connect to the inter-operator network through a border gateway (BG), which can provide the necessary interworking and routing protocols (for example, border gateway protocol [BGP]). • In the future, GPRS operators might implement quality of service (QoS) mechanisms over the inter-operator network to ensure service-level agreements (SLAs).• The main benefits of the architecture are its

– flexibility, – scalability, – interoperability, and – roaming attributes.

DNS Roaming

GRX / Roaming Partner

eDNS VPN

Gp VPN

SGSN

CN VPN

iDNS

eDNS

Gp

• The M-PBN recommends the use of an iDNS and eDNS to handle name resolutions.• The eDNS is connected to the eDNS VPN for DNS traffic. In this way the iDNS is not reachable

directly from external networks. This Mobile-PBN set-up provides a secure solution for all possible DNS Traffic scenarios

• Mobile-PBN describes a PS design solution for directly connected roaming partners as well as for those connected through a GRX.

• Secure connectivity using IPsec over the GRX or the Internet is also presented.

eDNS: Enhanced Domain Name System

GPRS Tunneling Protocol (GTP)• The GPRS network encapsulates all data network protocols into its own encapsulation protocol called the GPRS tunneling protocol (GTP). • The GTP ensures security in the backbone network and simplifies the routing mechanism and the delivery of data over the GPRS network.

Mobility Management• The operation of the GPRS is partly independent of the GSM network.• However, some procedures share the network elements with current GSM functions to increase efficiency and to make optimum use of free GSM resources (such as unallocated time slots).• An MS has three states in the GPRS system

– Active– Standby– Idle

The three-state model is unique to packet radio; GSM uses a two-state model (idle or active).

GPRS Tunneling Protocol (GTP)• The GPRS network encapsulates all data network protocols into its own encapsulation protocol called the GPRS tunneling protocol (GTP). • The GTP ensures security in the backbone network and simplifies the routing mechanism and the delivery of data over the GPRS network.

GPRS Interfaces (Again)

GPRS Network Protocol Stack

• The protocol between the SGSN and GGSN using the Gn interface is GTP. • This is a Layer 3 tunneling protocol similar to L2TP.

GPRS Network Protocol Stack• Note the Gn and Gi interface as IP, the underlying protocols are not specified, providing flexibility with the physical medium. • The most common physical interface used with GPRS is FE/GE. • This interface provides high bandwidth, low cost, and universal connectivity to other vendor equipment. • For the Gi interface, common interfaces are Serial, E1/T1 or Ethernet. • Running over the physical WAN interfaces can be a wide range of protocols including Frame Relay, ISDN, and HDLC.

GPRS Tunneling Protocol• The GTP tunneling protocol is a Layer 3 tunneling protocol. • The IP header identifies a session flow between the GGSN and SGSN. • The UDP header identifies the GTP application protocol (Port 3386).• The GTP header identifies the GTP tunnel session. • The payload identifies the session flow between the mobile station and the

remote host.

GPRS Tunneling Protocol• The GTP packet structure, typically has a fixed-size header and other

information called payload or information elements. • Currently, bits 1-5 of Octet 1 and Octets 7-12 are not in use. • TID is the tunnel ID that identifies a tunnel session. • The length field of GTP is different from the length field of IP.• In IP, the length includes the header; in GTP, length indicates only the GTP

payload.

GRPS Access Modes• The GPRS access modes specify whether or not the GGSN requests user authentication at the access point to a PDN (Public Data Network). • The available options are:

– Transparent—No security authorization/authentication is requested by the GGSN– Non-transparent—GGSN acts as a proxy for authenticating

• The GPRS transparent and non-transparent modes relate only to PDP type IPv4.

Transparent Access Mode• Transparent access pertains to a GPRS PLMN that is not involved in subscriber access authorization and authentication. • Access to PDN-related security procedures are transparent to GSNs.

– In transparent access mode, the MS is given an address belonging to the operator or any other domain’s addressing space. – The address is given either at subscription as a static address or at PDP context activation as a dynamic address. – The dynamic address is allocated from a Dynamic Host Configuration Protocol (DHCP) server in the GPRS network. – Any user authentication is done within the GPRS network. – No RADIUS authentication is performed; only IMSI-based authentication (from the subscriber identity module in the handset) is done.

Non-Transparent Mode• Non-transparent access to an intranet/ISP means that the PLMN plays a role in the intranet/ISP authentication of the MS. • Non-transparent access uses the Password Authentication Protocol (PAP) or Challenge Handshake Authentication Protocol (CHAP) message issued by the mobile terminal and piggy-backed in the GTP PDP context activation message. • This message is used to build a RADIUS request toward the RADIUS server associated with the access point name (APN).

GPRS Access Point Name• The GPRS standards define a network identity called an

access point name (APN). • An APN identifies a PDN that is accessible from a GGSN

node in a GPRS network (e.g., www.Cisco.com). • To configure an APN, the operator configures three elements

on the GSN node:– Access point

• Defines an APN and its associated access characteristics, including security (RADIUS), dynamic address allocation (DHCP), and DNS services

– Access point list• Defines a logical interface that is associated with the virtual template

– Access group• Defines whether access is permitted between the PDN and the MS

In GPRS, only the APN is used to select the target network.

APN List in MTN NG

APN

OJGGSN01 APGGSN01 ACCESSBANK.NET CLIENTLESS.VENTURI.TEST CLIENT.VENTURI.TEST ENSSOLUTIONS.TEST CLIENTLESS.VENTURI.TEST FoodConcept.net CRITICALRESCUE.NET IHSNIGERIAPLC.NET ENSSOLUTIONS.TEST KTEL.NET ETRANSACT.NET NSPORTSLOTTERY.NET NWGSUPPORT.TEST RSLOTTO.NET PLANNING.TEST SHAGAMUMFB.NET PREMIERLOTTO.NET SIMREG.MTNNIGERIA.NET RADIOSPEED.TEST VRAIE.NET SHAGAMUMFB.NET WGC.NET SKYEBANK.NET blackberry.net YOUPOCPTT.NET cocacola.net blackberry.net coolfm.net citran.net web.gprs.mtnnigeria.net cmail.gprs.mtnnigeria.net finconnekt.net flybook mms.gprs.mtnnigeria.net mtneye.gprs.mtnnigeria.net payphone.gprs.mtnnigeria.net ras.gprs.mtnnigeria.net uba.gprs.mtnnigeria.net unionbank.net web.gprs.mtnnigeria.net

GPRS Processes1. Attach process —Process by which the MS attaches (i.e,

connects) to the SGSN in a GPRS network2. Authentication process —Process by which the SGSN

authenticates the mobile subscriber3. PDP activation process —Process by which a user session is

established between the MS and the destination network4. Detach process —Process by which the MS detaches (i.e.,

disconnects) from the SGSN in the GPRS network5. Network-initiated PDP request for static IP address —Process

by which a call from the packet data network reaches the MS using a static IP address

6. Network-initiated PDP request for dynamic IP address —Process by which a call from the packet data network reaches the MS using a dynamic IP address

GPRS Tunneling Protocol (GTP)• The GPRS network encapsulates all data network protocols into its own encapsulation protocol called the GPRS tunneling protocol (GTP). • The GTP ensures security in the backbone network and simplifies the routing mechanism and the delivery of data over the GPRS network.

GPRS Interfaces

GPRS Network Protocol Stack

• The protocol between the SGSN and GGSN using the Gn interface is GTP. • This is a Layer 3 tunneling protocol similar to L2TP.

GPRS Network Protocol Stack• Note the Gn and Gi interface as IP, the underlying protocols are not specified, providing flexibility with the physical medium. • The most common physical interface used with GPRS is FE/GE. • This interface provides high bandwidth, low cost, and universal connectivity to other vendor equipment. • For the Gi interface, common interfaces are Serial, E1/T1 or Ethernet. • Running over the physical WAN interfaces can be a wide range of protocols including Frame Relay, ISDN, and HDLC.

GPRS Tunneling Protocol• The GTP tunneling protocol is a Layer 3 tunneling protocol. • The IP header identifies a session flow between the GGSN and SGSN. • The UDP header identifies the GTP application protocol (Port 3386).• The GTP header identifies the GTP tunnel session. • The payload identifies the session flow between the mobile station and the

remote host.

GPRS Tunneling Protocol• The GTP packet structure, typically has a fixed-size header and other

information called payload or information elements. • Currently, bits 1-5 of Octet 1 and Octets 7-12 are not in use. • TID is the tunnel ID that identifies a tunnel session. • The length field of GTP is different from the length field of IP.• In IP, the length includes the header; in GTP, length indicates only the GTP

payload.

GRPS Access Modes• The GPRS access modes specify whether or not the GGSN requests user authentication at the access point to a PDN (Public Data Network). • The available options are:

– Transparent—No security authorization/authentication is requested by the GGSN– Non-transparent—GGSN acts as a proxy for authenticating

• The GPRS transparent and non-transparent modes relate only to PDP type IPv4.

Transparent Access Mode• Transparent access pertains to a GPRS PLMN that is not involved in subscriber access authorization and authentication. • Access to PDN-related security procedures are transparent to GSNs.

– In transparent access mode, the MS is given an address belonging to the operator or any other domain’s addressing space. – The address is given either at subscription as a static address or at PDP context activation as a dynamic address. – The dynamic address is allocated from a Dynamic Host Configuration Protocol (DHCP) server in the GPRS network. – Any user authentication is done within the GPRS network. – No RADIUS authentication is performed; only IMSI-based authentication (from the subscriber identity module in the handset) is done.

Non-Transparent Mode• Non-transparent access to an intranet/ISP means that the PLMN plays a role in the intranet/ISP authentication of the MS. • Non-transparent access uses the Password Authentication Protocol (PAP) or Challenge Handshake Authentication Protocol (CHAP) message issued by the mobile terminal and piggy-backed in the GTP PDP context activation message. • This message is used to build a RADIUS request toward the RADIUS server associated with the access point name (APN).

GPRS Access Point Name• The GPRS standards define a network identity called an

access point name (APN). • An APN identifies a PDN that is accessible from a GGSN

node in a GPRS network (e.g., www.Cisco.com). • To configure an APN, the operator configures three elements

on the GSN node:– Access point

• Defines an APN and its associated access characteristics, including security (RADIUS), dynamic address allocation (DHCP), and DNS services

– Access point list• Defines a logical interface that is associated with the virtual template

– Access group• Defines whether access is permitted between the PDN and the MS

In GPRS, only the APN is used to select the target network.

GPRS Processes1. Attach process —Process by which the MS attaches (i.e,

connects) to the SGSN in a GPRS network2. Authentication process —Process by which the SGSN

authenticates the mobile subscriber3. PDP activation process —Process by which a user session is

established between the MS and the destination network4. Detach process —Process by which the MS detaches (i.e.,

disconnects) from the SGSN in the GPRS network5. Network-initiated PDP request for static IP address —Process

by which a call from the packet data network reaches the MS using a static IP address

6. Network-initiated PDP request for dynamic IP address —Process by which a call from the packet data network reaches the MS using a dynamic IP address

MSCHLR/AuCEIR

BSC

BTS

Um

PSTNNetwork

GSM & (E)GPRS Network ArchitecturePCU

EDAPGb

Gateway GPRSSupport Node(GGSN)

Charging Gateway (CG) Local

AreaNetwork

Server

Router

Corporate 1

Server

Router

Corporate 2

Datanetwork(Internet)

Datanetwork(Internet)

Billing System

Inter-PLMNnetwork

GPRSINFRASTRUCTURE

BorderGateway (BG)

Lawful InterceptionGateway (LIG)

GPRSbackbonenetwork(IP based)

Serving GPRSSupport Node(SGSN)

SS7Network

PAPU

MSC

PSTNNetwork

GPRSINFRASTRUCTURE

HLR/AuCEIR

Gateway GPRSSupport Node(GGSN)

Domain Name Server (DNS)

GPRSbackbonenetwork(IP based)

PDP Context Activation - 1

1. MS sends "Activate PDP Context Request" to SGSN

2. SGSN checks against HLR

Datanetwork(Internet)

Datanetwork(Internet)

Access Point

SS7Network

APN= "Intranet.Ltd.com"

2.Serving GPRSSupport Node(SGSN)

Access Point Name = Reference to an external packet data network the user wants to connect to

BSC

BTS

Um

1.

MSC

PSTNNetwork

GPRSINFRASTRUCTURE

HLR/AuCEIR

PDP Context Activation - 2Finding the GGSN

3. SGSN gets the GGSN IP address from DNS4. SGSN sends "Create PDP Context Request"

to GGSN

Datanetwork(Internet)

Datanetwork(Internet)

SS7Network

4.

Serving GPRSSupport Node(SGSN)

GPRSbackbonenetwork(IP based)

3.Domain Name Server (DNS)

Gateway GPRSSupport Node(GGSN)

Access Point

BSC

BTS

Um

DNS (Domain Name System) = mechanism to map logical names to IP addresses

MSC

GPRSINFRASTRUCTURE

HLR/AuCEIR

PSTNNetwork

PDP Context Activation - 3Access Point Selection

Access Point Name refers to the external network the subscriber wants to use

Datanetwork(Internet)

SS7Network

Serving GPRSSupport Node(SGSN)

GPRSbackbonenetwork(IP based)

Domain Name Server (DNS)

Gateway GPRSSupport Node(GGSN)

Access Point

APN="Intranet.Ltd.com"

Datanetwork(Internet)

BSC

BTS

Um

MSC

PSTNNetwork

GPRSINFRASTRUCTURE

HLR/AuCEIR

Datanetwork(Internet)

Datanetwork(Internet)

Access Point

APN="Intranet.Ltd.com"

Domain Name Server (DNS)

SS7Network

5.

Serving GPRSSupport Node(SGSN)

GPRSbackbonenetwork(IP based)

6.

Gateway GPRSSupport Node(GGSN)

BSC

BTS

Um

User (dynamic) IP address allocated5. GGSN sends "Create PDP Context Response" back

to SGSN6. SGSN sends “Activate PDP Context Accept“ to the

MS

PDP Context Activation - 4Context Activated

GPRS Attach Process (1)

GPRS Attach Process (1)

GPRS Attach Process (Steps)When a mobile subscriber turns on their handset, the following actions occur:

Step 1A handset attach request is sent to the new SGSN.

Step 2The new SGSN queries the old SGSN for the identity of this handset. The old SGSN responds with the identity of the handset.

Step 3The new SGSN requests more information from the MS. This information is used to authenticate the MS to the new SGSN.

Step 4The authentication process continues to the HLR. The HLR acts like a RADIUS server using a handset-level authentication based on IMSI and similar to the CHAP authentication process in PPP.

Step 5A check of the equipment ID with the EIR is initiated.

GPRS Attach Process (Steps)Step 6– If the equipment ID is valid, the new SGSN sends a location update to the HLR indicating the change of location to a new SGSN. – The HLR notifies the old SGSN to cancel the location process for this MS. – The HLR sends an insert subscriber data request and other information associated with this mobile system and notifies the new SGSN that the update location has been performed.

Step 7– The new SGSN initiates a location update request to the VLR. The VLR acts like a proxy RADIUS that queries the home HLR.

Step 8– The new SGSN sends the Attach Accept message to the MS.

Step 9– The MS sends the Attach Complete message to the new SGSN.

Step 10– The new SGSN notifies the new VLR that the relocation process is complete.

MS Attach Procedure

MS Attach Procedure

• The MS sends an Attach Request message to the SGSN. The Attach Request message contains information about the MS, which can identify itself with IMSI or P-TMSI. The RAI can correspond to either the SGSN trying to attach to, or an SGSN previously attached toIf the MS identifies itself with P-TMSI and the RAI correspond to an RA served by another SGSN than the one trying to attach to, the SGSN sends a Identification Request message to fetch the MM context from the previously used SGSN.

• The previously used SGSN returns the IMSI number and authentication information of the MS by sending an Identification Response message to SGSN. If the IMSI number is not stored in the SGSN or in the old SGSN it is requested from the MS through an Identity Request message. The MS sends the IMSI number to the SGSN in an Identity Response message.

• If no MM context for the MS exists anywhere in the network, the SGSN fetches the MM context and authentication data from the HLR and authenticates the MS. If the location of the MS is unknown in the HLR, the SGSN sends an Update Location message to the HLR.

• The HLR sends the relevant subscriber data to the SGSN. The SGSN sends an acknowledgement to the HLR that the subscriber data has been received. The HLR sends an acknowledgement to the SGSN that location update is performed.

• The SGSN allocates a new P-TMSI and a P-TMSI signature and, if ciphering is enabled, enters ciphering mode

• The SGSN sends an Attach Accept message to the MS. The Attach Accept message includes a list of equivalent PLMNs, if such a list has been defined

• The MS acknowledges the received P-TMSI and P-TMSI signature with an Attach Complete message.

GPRS Authentication Process• The GPRS authentication process is very similar to the CHAP with a RADIUS server. • The authentication process follows these steps:

1. The SGSN sends the authentication information to the HLR. The HLR sends information back to the SGSN based on the user profile that was part of the user’s initial setup.

2. The SGSN sends a request for authentication and ciphering (using a random key to encrypt information) to the MS. The MS uses an algorithm to send the user ID and password to the SGSN. Simultaneously, the SGSN uses the same algorithm and compares the result. If a match occurs, the SGSN authenticates the user.

GPRS Authentication Process

PDP Context Activation Process

PDP Context Activation Process1. The SGSN receives the activation request from the MS; for example, the MS requests access to the APN Cisco.com.2. Security functions between the MS and SGSN occur.3. The SGSN initiates a DNS query to learn which GGSN node has access to the Cisco.com APN. The DNS query is sent to the DNS server within the mobile operator’s network. The DNS is configured to map to one or more GGSN nodes. Based on the APN, the mapped GGSN can access the requested network.4. The SGSN sends a Create PDP Context Request to the GGSN. This message contains the PAP information, CHAP information, PDP request, APN, and quality of service information.

PDP Context Activation Process5. If operating in the non-transparent mode, the PAP and CHAP information in the PDP request packet is sent to the RADIUS server for authentication.6. If the RADIUS server is to provide a dynamic IP address to the client, it sends a DHCP address request to the DHCP server. In transparent mode, the RADIUS server is bypassed.7. If IPSec functionality is required, security functions occur between the GGSN and network access server (NAS).8. The GGSN sends a Create a PDP Context Response message to the SGSN.9. The SGSN sends an Activate PDP Context Accept message to the MS.

Detach Process initiated by MS• When a mobile subscriber turns off their handset, the detach process initiates. • The detach process is described below.

1. The MS sends a Detach Request to the SGSN.2. The SGSN sends a Delete PDP Context Request message to the serving GGSN.3. The SGSN sends an IMSI Detach Indication message to the MSC/VLR indicating the MS request to disconnect.4. The SGSN sends a GPRS Detach Indication message to the MSC/VLR.5. The SGSN sends the Detach Accept message to the MS.

Note: The GSN nodes must always respond to the detach request with a positive delete response to the MS and accept the detach request requested by the client. The positive delete response is required even if the SGSN does not have a connection pending for that client.

Detach Process initiated by MS

Network Initiated PDP Request For A Static IP Address

• The PDP protocol data unit (PDU) initiated from the network side is not fully specified by ETSI standards. • A connection request generated from the Internet/ intranet site specifies only the IP address of the client in the IP packets destined for the MS. • The requesting host provides no indication of the mobile device IMSI (i.e., the MAC address of the MS). • In mobile communications, all communications are based on the MS MAC address called the IMSI. • The IP address must be mapped to an IMSI to identify a valid GTP tunnel.

Figure: Network Initiate PDP (Static IP Address)

Network Initiated PDP Request For A Static IP Address

The following steps describe a PDP request initiated from the network side when the client has been assigned a static IP address.1. When the GGSN receives a packet, it checks its mapping table for an established GTP tunnel for this packet.2. When the GGSN locates the IMSI associated with this IP address, it sends a Send Routing Information message to HLR through an intermediate SGSN. The intermediate SGSN notifies the GGSN of the actual SGSN currently serving this client.3. On locating the appropriate SGSN, the GGSN sends a PDU Notification Request message to the serving SGSN.4. The SGSN sends a Request PDP Context Activation message to the MS and notifies it of the pending connection request.5. If the MS agrees to accept the call, it enters the PDP Context Activation procedure with the requesting GGSN.

Network Initiated PDP Request For A Dynamic IP Address

• The ETSI standards do not fully specify requirements for a network-generated PDP request when the client is dynamically assigned a temporary IP by a DHCP server.

• The following message sequence is Cisco’s implementation for this scenario. This method uses Cisco’s Network Registrar (CNR), which includes a DHCP, DNS, and an LDAP server.1. The host initiates a DNS query to obtain the IP address of the MS from a DNS

server. The DNS server resolves the client’s name to an IP address previously assigned to the client by the DHCP server.

2. The host sends a request to the GGSN for a connection using this IP address.3. The GGSN queries the LDAP server to obtain the MS IMSI. The LDAP server

stores a record for the MS with the client IMSI, name, and IP address.4. The GGSN sends a PDU Notification Request message to the serving SGSN.5. The SGSN sends a Request PDP Context Activation message to the MS and

notifies it of the pending connection request.6. If the MS agrees to accept the call, it enters the PDP Context Activation

procedure with the requesting GGSN.

Figure: Network Initiated PDP Request For A Dynamic IP Address

MS Attach Procedure

MS Attach Procedure

• The MS sends an Attach Request message to the SGSN. The Attach Request message contains information about the MS, which can identify itself with IMSI or P-TMSI. The RAI can correspond to either the SGSN trying to attach to, or an SGSN previously attached toIf the MS identifies itself with P-TMSI and the RAI correspond to an RA served by another SGSN than the one trying to attach to, the SGSN sends a Identification Request message to fetch the MM context from the previously used SGSN.

• The previously used SGSN returns the IMSI number and authentication information of the MS by sending an Identification Response message to SGSN. If the IMSI number is not stored in the SGSN or in the old SGSN it is requested from the MS through an Identity Request message. The MS sends the IMSI number to the SGSN in an Identity Response message.

• If no MM context for the MS exists anywhere in the network, the SGSN fetches the MM context and authentication data from the HLR and authenticates the MS. If the location of the MS is unknown in the HLR, the SGSN sends an Update Location message to the HLR.

• The HLR sends the relevant subscriber data to the SGSN. The SGSN sends an acknowledgement to the HLR that the subscriber data has been received. The HLR sends an acknowledgement to the SGSN that location update is performed.

• The SGSN allocates a new P-TMSI and a P-TMSI signature and, if ciphering is enabled, enters ciphering mode

• The SGSN sends an Attach Accept message to the MS. The Attach Accept message includes a list of equivalent PLMNs, if such a list has been defined

• The MS acknowledges the received P-TMSI and P-TMSI signature with an Attach Complete message.

MS-Initiated Detach Procedure

MS-Initiated Detach Procedure

• The MS sends a Detach Request message to the SGSN. The Detach Request message mentions whether the detach is due to a switch-off. • If it has active PDP contexts, the SGSN starts to deactivate them by sending one Delete PDP Context Request message for each PDP context to the Gateway GPRS Support Node (GGSN). • The GGSN acknowledges the deletion with a Delete PDP Context Response message. • In GSM, in case of an IMSI detach, the SGSN sends an IMSI Detach Indication message to the VLR. • A GSM MS can remain IMSI attached and perform a GPRS detach. The SGSN then sends a GPRS Detach Indication message to the VLR, which deletes the association with the SGSN and afterwards handles paging and location update without involving the SGSN. • If the detach is not due to a switch-off, the SGSN sends a Detach Accept message to the MS. • In WCDMA Systems, the SGSN-initiated Iu Release procedure is executed and communicated with the Radio Network Controller (RNC).

Intra-SGSN RA Update

Intra-SGSN RA Update

• The MS sends a Routing Area Update Request message to the SGSN. The Routing Area Update Request message includes information on the old RA and old P-TMSI, and if it is an RA update or periodic RA update.

• Security functions may be executed. Selective authentication applies. The SGSN validates the MS's presence in the new RA. If the check fails (for example, wrong P-TMSI signature) the SGSN authenticates the MS. If the authentication also fails, the SGSN logs the error and sends an RA Update Reject message. If all checks are successful, the SGSN updates the MM context for the MS. A new P-TMSI, including NRI in case the SGSN is a pool member, and P-TMSI signature, is allocated.

• A Routing Area Update Accept message is returned to the MS.. It also includes the P-TMSI

• The MS acknowledges the new P-TMSI with a Routing Area Update Complete message.

• The operator can choose to configure the SGSN to send network information to the MS in the GMM Information message.

Inter-SGSN RA Update

Inter-SGSN RA Update

• The MS sends a Routing Area Update Request message to the new SGSN. The Routing Area Update Request message includes information on the old RA. The SGSN analyzes the received RA identification and starts the inter-SGSN RA update procedure if the old RA is controlled by a cooperating SGSN.

• The new SGSN sends an SGSN Context Request message to the old SGSN to get the MM and PDP contexts for the MS.

• The old SGSN validates the old P-TMSI and responds with an SGSN Context Response message containing the APN-OI, IMSI number, MM context, and possible PDP contexts for the MS.

• Security functions are executed.

• The new SGSN sends an Update PDP Context Request message to the appropriate GGSNs. The GGSNs update their PDP context fields and return Update PDP Context Response messages.

• The new SGSN informs the HLR about the change of SGSN by sending an Update Location message to the HLR.

• The HLR sends a Cancel Location message to the old SGSN.

• The old SGSN acknowledges the location canceling with a Cancel Location Ack message.

• The HLR sends an Insert Subscriber Data message to the new SGSN.

• The new SGSN validates the presence of the MS in the new RA. If all checks are successful, the new SGSN returns an Insert Subscriber Data Ack message to the HLR. The HLR acknowledges the Update Location message by sending an Update Location Ack message to the new SGSN.

• The new SGSN responds to the MS with a Routing Area Update Accept message, including a new P-TMSI, and creates MM and PDP contexts for the MS.

• The MS acknowledges the new P-TMSI with a Routing Area Update Complete message.

Session Management – In Detail

PDP Context Activation Procedure

PDP Context Activation Procedure

• The MS sends an Activate PDP Context Request message to the SGSN. In addition to the requested QoS, the message contains information on the requested APN, the IP address, and PDP type, that is, IPv4 or IPv6 (optional parameters).

• The SGSN validates the Activate PDP Context Request using the information provided by the MS and the subscriber record. If allowed, the SGSN resolves the APN into a list of GGSN IP addresses and sends a Create PDP Context Request message to the first GGSN IP address in the list. For example, information on the MCC, MNC, IMEISV, and the radio access technology of the MS is sent to the GGSN.

• If the GGSN accepts the request, it responds with a Create PDP Context Response message. If the response from the GGSN indicates that the request was rejected, the SGSN may, depending on reject cause, try the second GGSN IP address in the list received from the DNS and so on until the list is exhausted. If the MS requests a dynamic address, the SGSN lets the GGSN allocate the dynamic address.

• If the validation was successful, the SGSN sends an Activate PDP Context Accept message to the MS.

MS-Initiated PDP Context Modification Procedure

MS-Initiated PDP Context Modification Procedure

• The MS sends a Modify PDP Context Request message to the SGSN.

• The SGSN sends the GGSN an Update PDP Context Request message, which includes the optional protocol configuration option.

• The GGSN responds with an Update PDP Context Response message to the SGSN.

• For GSM, the SGSN exchanges information related to ongoing user data transmission with the BSS, and initiates modification of the Packet Flow Context. The SGSN sends a Modify PDP Context Accept message to the MS

GGSN-Initiated PDP Context Deactivation Procedure

GGSN-Initiated PDP Context Deactivation Procedure

• The GGSN sends a Delete PDP Context Request message to the SGSN.

• The SGSN forwards the request to the MS in a Deactivate PDP Context Request message.

• The MS removes the PDP context and returns a Deactivate PDP Context Accept message to the SGSN.

• The SGSN returns a Delete PDP Context Response message to the GGSN indicating the deactivation is completed. If the MS was using a dynamic PDP address, the GGSN releases this PDP address and makes it available for subsequent activation by other MSs.

  Charging Postpaid Charging

Charging Data Records (CDRs), containing information about chargeable events, are transferred from the SGSN or GGSN either directly to a billing system or through a charging gateway.

The charging gateway offers functions to prepare GPRS charging information. It can, for example, check and eliminate double CDRs and verify that the CDRs have the right content. In addition, the charging gateway can examine the order of the CDRs and provide the billing system with charging data arranged in sequence. The Ericsson charging gateway product is called Multi Mediation (MM).

The billing system handles the billing of the customers, for example, keeping track of the account types and charging agreements. The CDRs used for postpaid charging can be sent over both a file-based output stream using FTP and a near-real-time output stream using GTP Prime (GTP')Hot billing, when the charging data is created immediately after service delivery, also uses CDRs for charging.

ChargingPrepaid Charging The CDRs used for prepaid charging can be sent over a near-real-timeoutput stream using GTP'. The prepaid charging function enables the SCP to have real-time control over subscriber service usage in the SGSN. The charging data is transferred from the SGSN usingCustomized Applications for Mobile Network Enhanced Logic (CAMEL) and the CAMEL Application Part (CAP) protocol. The GPRS Service Switching Function (gprsSSF) in the SGSN communicates with the GSM Service Control Function (gsmSCF) in the Charging Control Node (CCN). In the Ericsson solution the CCN converts the CAP protocol to the SC1+ protocol, which is used to transfer the charging information further to the SCP. The SCP is connected to a server containing information about the prepaid accounts.

Other than this, Ericsson has a SASN solution for real-time charging.

The SASN SolutionThe SASN is deployed in the operator’s IP network and captures the user traffic. By analyzing and classifying that traffic, it should identify the subscriber’s identity and the type of service being used, with no need for changes in neither the subscriber terminals nor the platforms providing theservice. Based on that knowledge, it should enable a number of sophisticated applications which can differentiate the traffic handling on a per service and subscriber basis. Some examples of applications are:• Content-based charging (service-differentiated charging or flowbased charging): SASN can generate charging information in real timeso the operator can apply price models that are service specific, both for pre-paid and postpaid subscribers.• Real-time prepaid. SASN acts as a control point of a real-time prepaid system. Operators can avoid fraud and increase the appeal oftheir data services to a broader customer base.• Content access control and filtering: SASN requests instructions from an access policy server and enforces its policy. Operator’s canimplement subscription management and parental control solutions, among others.Statistics: SASN accounting information can be exported to a data warehouse where the operator can gather valuable intelligence aboutthe network usage in terms of network reliability, most popular contents, etc.• Traffic marking and QoS management: SASN can enforce traffic shaping policies and request QoS changes to the access network, sothat operators can optimize the use of network resources on a perservice basis. SASN should also be able to mark packets based onthe service information. SASN is access agnostic and can be deployed in a number of networks, including GPRS, UMTS, WiFi, etc.

MSCHLR/AuCEIR

BSC

BTS

Um

PSTNNetwork

GSM & (E)GPRS Network ArchitecturePCU

EDAPGb

Gateway GPRSSupport Node(GGSN)

Charging Gateway (CG) Local

AreaNetwork

Server

Router

Corporate 1

Server

Router

Corporate 2

Datanetwork(Internet)

Datanetwork(Internet)

Billing System

Inter-PLMNnetwork

GPRSINFRASTRUCTURE

BorderGateway (BG)

Lawful InterceptionGateway (LIG)

GPRSbackbonenetwork(IP based)

Serving GPRSSupport Node(SGSN)

SS7Network

PAPU

GMSK & 8-PSK - Phase State Vectors

22,5° offset to avoid zero crossing

GMSK

8PSK(0,0,1)

(1,0,1)

(0,0,0) (0,1,0)

(0,1,1)

(1,1,1)

(1,1,0)

(1,0,0)

Time

Envelope (amplitude)

Time

Envelope (amplitude)

(0,0,1)

(1,0,1)

(d(3k),d(3k+1),d(3k+2))=

(0,0,0) (0,1,0)

(0,1,1)

(1,1,1)

(1,1,0)

(1,0,0)

8-PSK Modulation

EDGE GSM + EDGE Modulation 8-PSK, 3bit/sym GMSK, 1 bit/sym Symbol rate 270.833 ksps 270.833 ksps Bits/burst 348 bits

2*3*58 114 bits 2*57

Gross rate/time slot 69.6 kbps 22.8 kbps

• 8-PSK (Phase Shift Keying) has been selected as the new modulation added in EGPRS

• 3 bits per symbol

• 22.5° offset to avoid origin crossing (called 3/8-8-PSK)

• Symbol rate and burst length identical to those of GMSK

• Non-constant envelope high requirements for linearity of the power amplifier

• Because of amplifier non-linearities, a 2-4 dB power decrease back-off (BO) is typically needed, Nokia guaranteed a BO of 2 DB for BTS

3/8Phase states transitionsto avoid zero-crossing

GMSK and 8PSK BurstsdB

t

- 6

- 30

+ 4

8 µs 10 µs 10 µs 8 µs

(147 bits)

7056/13 (542.8) µs 10 µs

(*)

10 µs

- 1+ 1

(***)

(**)

10 8 10 10 8 10 t (s)

dB

-30

(*)

-6

+2,4

+4

-20

-2

(***)

(**)

2 2 22

7056/13 (542,8)s

(147 symbols)

0

GMSK Burst

8PSK Burst

Phase state vector diagram• Amplitude is not fixed• Origin is not crossed• Overshooting

8-PSK Modulation – Back-off Value

• Since the amplitude is changing in 8-PSK the transmitter non-linearities can be seen in the transmitted signal• These non-linearities will cause e.g. errors in reception and bandwidth spreading.• In practice it is not possible to transmit 8-PSK signal with the same power as in GMSK due to the signal must remain in the linear part of the power amplifier

Peak to Average of 3,2 dB

Pin

Pout

Back Off= 4 dB

Compression point

Peak to Average of 3,2 dB

Pin

Pout

Back Off= 4 dB

Compression point

• The back-off value is taken into account in link budget separately for UL / DL and bands: 900/850, 1800/1900)

• Too high MCA (8PSK) can lead to unsuccessful TBF establishment, if the MS is on cell border with low signal level (so the back-off is taken into account) and / or low C/I

Burst Structure• Burst structure is similar with current GMSK burst, but term 'bit' is replaced by 'symbol'• Training sequence has lower envelope variations • Seamless switchover between timeslots• In case of max output power only, back-off applied to 8-PSK

TSL1 TCH

GMSK

TSL2 TCH

GMSK

TSL3 TCH

GMSK

TSL4 TCH

GMSK

TSL5 PD T CH 8 - PSK / GMSK

TSL6 PD T CH 8 - PSK / GMSK

TSL7 PD T CH 8 - PSK / GMSK

TSL0 BCCH GMSK

P (dB)

t ( us )

EDGE Signal

1 2 3 4

1. Spectrum of Unfiltered 3pi/8 8psk modulation.

2. Filtered to fit GSM bandwidth.

3. Constellation after filtering: error vectors introduced.

4. Constellation after receiver Edge (equalised) filtering

GPRS Coding Schemes• GPRS provides four coding schemes: CS-1, CS-2 and with PCU2 CS-3, CS-4• PCU1 and 16 kbit/s Abis links support CS-1 and CS-2, the Dynamic Abis makes it possible to use CS-3 and CS-4• Each TBF can use either a fixed coding scheme (CS-1 or CS-2), or Link Adaptation (LA) based on BLER• Retransmitted RLC data blocks must be sent with the same coding as was used initially

Coding Scheme

Payload (bits)per RLC block

Data Rate (kbit/s)

CS1 181 9.05

CS2 268 13.4

CS3 312 15.6

CS4 428 21.4

More Data = Less Error Correction

Nokia GPRSPCU1

• CS1 & CS2 – Implemented in all Nokia BTS without HW change

• CS3 & CS4 – S11.5 (with PCU2) and UltraSite BTS SW CX4.1 CD1 (Talk is supporting CS1 and CS2)

Dat

a

Err

orC

orre

ctio

n

GPRS Coding Schemes

Nokia GPRSPCU2

CS-1

CS-2

CS-3

57 57 57 57 57 57 57 57

456 bits

MAC

USF BCS +4

puncturing

rate a/b convolutional coding

CS-1 CS-2 CS-3

RLC/MAC Block Size: 181 268 312

Block Check Sequence: 40 16 16

Precoded USF: 3 6 6

1/2 ~2/3 ~3/4

length: 456 588 676

0 132 220

Data rate (kbit/s): 9.05 13.4 15.6

interleaving

MAC

USF BCS

RLC/MAC Block Size: 428

BCS Size: 16

Precoded USF: 12

Data rate (kbit/s): 21.4

CS-4

20 ms

GPRS Coding Schemes

EGPRS Modulation and Coding Schemes

EGPRS modulation and coding schemes:Scheme Modulation Data rate

kb/s

MCS-9 59.2

MCS-8 54.4

MCS-7 44.8

MCS-6 29.6 27.2

MCS-5

8PSK

22.4

MCS-4 17.6

MCS-3 14.8 13.6

MCS-2 11.2

MCS-1

GMSK

8.8

Ref: TS 03.64

EGPRS Data Treatment Principle in RF Layer

User data

"Additional info" that does not require extra protection

Header part, robust coding for secure transmission

Adding redundancy

Puncturing of the coded info

BSC

BTS

• Class C Packet only

(or manually switched between GPRS and speech modes)

• Class B Packet and Speech (not at same time)

(Automatically switches between GPRS and speech modes)

• Class A Packet and Speech at the same time(DTM is subset of class A)

(E)GPRS Mobile Terminal Classes

(E)GPRS Multislot ClassesType 1

Multislot Classes 1-12- Max 4 DL or 4 UL TSL (not at same time)- Up to 5 TSL shared between UL and DL- Minimum 1 TSL for F Change- 2-4 TSL F Change used when idle

measurements required

Multislot Classes 19-29- Max 8 downlink or 8 uplink

(not required at same time)- 0-3 TSL F Change

Multislot Classes 30-45 (Rel-5)- Max 5 downlink or 5 uplink (6 shared)- Max 6 downlink or 6 uplink (7 shared)

Type 2

Multislot Classes 13-18- simultaneous receive & transmit- max 8 downlink and 8 uplink (Not available yet, difficult RF design)

DL

UL

DL

UL

1 TSL for F Change

1 TSL for Measurement

DL

UL

GPRS implementation• GPRS/EGPRS capable terminals are required• GPRS territory is required in BTS• Packet Control Units (PCUs) need to be implemented in BSCs• Gb interface dimensioning• GPRS packet core network dimensioning

• If CS3&CS4 will be implemented following units/items are required– PCU2 with S11.5 BSC SW– Dynamic Abis Pool (DAP) – EDGE capable TRXs – UltraSite and MetroSite BTS SW support

EGPRS Implementation• Can be introduced incrementally to the network where the demand is

– EGPRS capable MS

– Network HW readiness/upgrade (BTS and TRX)

– TRS capacity upgrade (Abis and Gb!)

– Dynamic Abis

GMSK coverage

8-PSK coverage

AA-bis

Gb

Gn

BTS

BTS

BSC

SGSNGGSN

MSC

More capacity in interfaces to support higher data usage

EDGE capable TRX, GSM compatible

EDGE capable terminal, GSM compatible

EDGE functionality in the network elements

Create a BCF

Create a BTS

Attach BTS to RAC

Enable EGPRS (EGENA/Y)

Define GPRS and EGPRS parameters

Enable GPRS (GENA/Y)

Create a TRX with DAP connection

Create handover and power control parameters

The steps to create radio network objects

Enabling (E)GPRS

RAC= Routing Area code

Create the dynamic Abis pool

Disable the GPRS in the cell

Lock the TRX

Delete the TRX to be connected to Dynamic Abis pool

Create a TRX which uses the dynamic Abis poolAll the TRXs that will be using EGPRS in the BTS must be attached to a dynamic Abis pool

Unlock the TRX

Enable EGPRS in the BTS (EGENA/Y)

Enable GPRS in the cell (GENA/Y)

Unlock the BTS

Lock the BTS

The steps to enable the (E)GPRS in BSC

Enabling (E)GPRS

To be considered:• When the TRX has been created with EDAP defined at BSC and EGPRS feature is enabled, the TRX

must be attached to EDAP on the BTS side also not to fail the configuration of BCF

• EDAP in BSC must be inside the TSL boundaries defined in the BTS side– When modifying EDAP the size of EDAP in the BTS has to be the same as the size of EDAP in the BSC

• Creating, modifying or deleting of EDAP in the BSC will cause a territory downgrade/upgrade procedure to all territories served by the PCU in question– The ongoing EGPRS/GPRS connections will pause and resume immediately

• The maximum EDAP size is 12 timeslots

• EDAP must be located on the same ET-PCM line as TRX signaling and traffic channels

• There are no specific commissioning tests concerning EDAP

• EDAP must be located on the same BCSU as Gb interface

Enabling (E)GPRS

(E)GPRS Protocol Architecture

L1

L2

IP

UDP

GTP

USERPAYLOAD

GGSN

L1

L2

IP

GPRS Bearer

GGSN

Relay

IP

GPRS IP Backbone

L1

L2

IP

GTP

L1bis

NW sr

BSSGP

SNDCP

LLC UDP

SGSN

Relay

Gn

Internet

L1

L2

IP

TCP/UDP

APP

Gi

User information transferUser information transfer

LLC

SNDCP

IP

TCP/UDP

APP

RLC

MAC

GSM RF

MS

RLC

MAC

GSM RF

BSSGP

NW sr

L1bis

BSS

Ciphering and reliable link

Um Gb

Compression, segmentation

FIXED HOST

(E)GPRS Logical Channels

GPRS Air Interface Logical Channels

CCCHCommon Control Channels

DCHDedicated Channels

PCHPaging CH

AGCHAccess Grant CH

RACHRandom Access CH

Existing GSM Channels

(Shared with GPRS Signaling in GPRS Release 1)

PACCHPacket AssociatedControl CH

PDTCHPacket Data TCH

NEW GPRS Channels

Functionality - ContentIntroduction • Network architecture and Interfaces• Mobile classes• Network Protocols• Multiframe and header structure• Air interface mapping – physical and logical channel

Procedures• State and Mobility Management

• GPRS Attach/Detach• Routing Area

• Session Management (PDP context)• Temporary Block Flow

• RLC/MAC Header• TBF Establishment

(E)GPRS Procedures - Content• Mobility Management and State Management

– Mobile States– GPRS attach– GPRS detach– Routing Area

• Session Management– PDP context activation

• Temporary Block Flow– RLC/MAC Header– TBF establishment

GPRS Mobility Management - Mobile States

MS location not known, subscriber is not reachable by the GPRS nw.

IDLE READY

STANDBY

READY Timer expiry

MOBILE REACHABLE Timer expiry

Packet TX/RX

GPRS Attach/Detach

MS location known to Routing Area level. MS is capable to being paged for point-to-point data.

MS location known to cell level. MS is transmitting or has just been transmitting. MS is capable of receiving point-to-point data.

Attach Procedure• The GPRS Attach procedure establishes a GMM context. This procedure is used for the following two purposes:

– a normal GPRS Attach, performed by the MS to attach the IMSI for GPRS services only – a combined GPRS Attach, performed by the MS to attach the IMSI for GPRS and non-GPRS services

• Attach procedure description– MS initiates by sending Attach Request– If network accepts Attach Request it sends Attach Accept

• P-TMSI, RAI

– If network does not accept Attach request it sends Attach Rejected– MS responds for Attach Accept message with Attach Complete (only if P-TMSI changes)

Detach Process• GPRS Detach procedure is used for the following two purposes:

– a normal GPRS Detach– a combined GPRS Detach (GPRS/IMSI detach, MS originated)

• MS is detached either explicitly or implicitly:– Explicit detach: The network or the MS explicitly requests detach.– Implicit detach: The network detaches the MS, without notifying the MS, a configuration-dependent time after the mobile reachable timer (MSRT) expired, or after an irrecoverable radio error causes disconnection of the logical link

Routing AreaThe Routing Area Update procedure is used for the

followings:• a normal Routing Area Update• a combined Routing Area Update• a periodic Routing Area Update• an IMSI Attach for non-GPRS services when the

MS is IMSI-attached for GPRS services.

• Routing Area (RA)– Subset of one, and only one Location Area (LA)– RA is served by only one SGSN– For simplicity, the LA and RA can be the same– Too big LA/RA increases the paging traffic, while too

small LA/RA increases the signaling for LA/RA Update

Routing Area Location

Area (LA)

Routing Area (RA)

SGSN

MSC/VLR

GS Interface

– Bad LA/RA border design can significantly increase the TRXSIG on LA/RA border cells causing the cell-reselection outage to be longer

– LA/RA border should be moved from those areas where the normal CSW and PSW traffic is very high

•PDP Context (Packet Data Protocol): Network level information which is used to bind a mobile station (MS) to various PDP addresses and to unbind the mobile station from these addresses after use

•PDP Context Activation– Gets an IP address from the network– Initiated by the MS– Contains QoS and routing information enabling data transfer between MS and GGSN– PDP Context Activation and Deactivation should occur within 2 seconds

Session Management - Establishing a PDP Context

PDP Context Request

155.131.33.55

MSC

PSTNNetwork

GPRSINFRASTRUCTURE

HLR/AuCEIR

Gateway GPRSSupport Node(GGSN)

Domain Name Server (DNS)

GPRSbackbonenetwork(IP based)

PDP Context Activation - 11. MS sends "Activate PDP Context Request" to

SGSN2. SGSN checks against HLR

Datanetwork(Internet)

Datanetwork(Internet)

Access Point

SS7Network

APN= "Intranet.Ltd.com"

2.Serving GPRSSupport Node(SGSN)

Access Point Name = Reference to an external packet data network the user wants to connect to

BSC

BTS

Um

1.

MSC

PSTNNetwork

GPRSINFRASTRUCTURE

HLR/AuCEIR

PDP Context Activation - 2Finding the GGSN

3. SGSN gets the GGSN IP address from DNS4. SGSN sends "Create PDP Context Request"

to GGSN

Datanetwork(Internet)

Datanetwork(Internet)

SS7Network

4.

Serving GPRSSupport Node(SGSN)

GPRSbackbonenetwork(IP based)

3.Domain Name Server (DNS)

Gateway GPRSSupport Node(GGSN)

Access Point

BSC

BTS

Um

DNS (Domain Name System) = mechanism to map logical names to IP addresses

MSC

GPRSINFRASTRUCTURE

HLR/AuCEIR

PSTNNetwork

PDP Context Activation - 3Access Point Selection

Access Point Name refers to the external network the subscriber wants to use

Datanetwork(Internet)

SS7Network

Serving GPRSSupport Node(SGSN)

GPRSbackbonenetwork(IP based)

Domain Name Server (DNS)

Gateway GPRSSupport Node(GGSN)

Access Point

APN="Intranet.Ltd.com"

Datanetwork(Internet)

BSC

BTS

Um

MSC

PSTNNetwork

GPRSINFRASTRUCTURE

HLR/AuCEIR

Datanetwork(Internet)

Datanetwork(Internet)

Access Point

APN="Intranet.Ltd.com"

Domain Name Server (DNS)

SS7Network

5.

Serving GPRSSupport Node(SGSN)

GPRSbackbonenetwork(IP based)

6.

Gateway GPRSSupport Node(GGSN)

BSC

BTS

Um

User (dynamic) IP address allocated5. GGSN sends "Create PDP Context Response" back

to SGSN6. SGSN sends “Activate PDP Context Accept“ to the

MS

PDP Context Activation - 4Context Activated

Temporary Block FlowTemporary Block Flow (TBF):• Physical connection where multiple mobile stations can share one or more traffic channels – each

MS has own TFI• The traffic channel is dedicated to one mobile station at a time (one mobile station is transmitting or

receiving at a time)• Is a one-way session for packet data transfer between MS and BSC (PCU)• Uses either uplink or downlink but not both (except for associated signaling)• Can use one or more TSLs

Comparison with circuit-switched:• normally one connection uses both the uplink and the downlink timeslot(s) for traffic

In two-way data transfer:• uplink and downlink data are sent in separate TBFs - as below

BSC

Uplink TBF (+ PACCH for downlink TBF)

Downlink TBF (+ PACCH for uplink TBF)

PACCH (Packet Associated Control Channel): Similar to GSM CSW SACCH

TLLI / TBF Concept

TBF (TFI + TSL)

MS SGSN GGSN

Internet or Intranet

GPRS CORE

BSS

TBF (RLC / MAC Flow)

TBF (LLC Flow)

PCUBTS

TLLI (SNDCP Flow)

P-TMSI

HLR

VLR

IMSITMSI

Multiple Mobiles and Downlink Transmission

TFI2

TFI5

TFI3

TFI2

MSs

BTS

The TFI included in the Downlink RLC Block header indicates which Mobile will open the RLC Block associated with its TBF

RLC Data Block

• Several mobiles can share one timeslot• Maximum of 7 Mobiles are queued in the Uplink• Mobile transmissions controlled by USF (Uplink State Flag) sent on DL (dynamic allocation)

TS 1

TS 2

TS 3

Uplink State Flag

• Mobile with correct USF will transmit in following Uplink block • Timeslot selected to give maximum throughput

New MS

Multiple Mobiles and Uplink Transmission

Multiple Mobiles and Uplink Transmission

USF = 1

USF = 2

USF = 3

USF = 3

MSs

BTS

RLC Data Block

The USF included in the Downlink RLC Block header identifies which Mobile will transmit in the following Uplink RLC Block

(E)GPRS Resource Allocation - ContentTerritory method

• Default and dedicated territory• Free TSLs

TSL Allocation• Scheduling with priority based QoS

Territory Method

TRX 1

TRX 2

BCCH TS TS TS TS TS TSSDCCHBCCH TS TS TS TS TS TSSDCCHBCCH TS TS TS TS TS TSSDCCH

TS TS TSTS TSTS TS TSTS TSTS TS TS TSTS TS TSTS

TS

TS

= (E)GPRS Territory/Dedicated capacity

= CSW Territory

TS= (E)GPRS Territory/Additional capacity

BCCH= Signaling

TS = Free TSL for CSW

TS= (E)GPRS Territory/ Default capacity

Territory border

EDAP, PCU and Gb Functionality - ContentEDAP

• Abis vs. Dynamic Abis• Channels carried on EDAP• EDAP limits• Abis PCM structure

PCU• PCU procedures• PCU types and limits

Gb• Gb protocols• Gb over FR• Gb over IP

Abis Basic Concepts – PCM frame (E1)One 64 kbit/s (8 bits) channel in PCM frame is

called timeslot (TSL)One 16 kbit/s (2bits) channel timeslot is Sub-TSLPCM frame has 32 (E1) or 26 (T1) TSLs

One Radio timeslot corresponds one 16 kbit/s Sub-TSL (BCCH, TCH/F etc.) and one TRX takes two TSLs from Abis

0 MCB LCB123456789

101112131415161718 TCH 0 TCH 1 TCH 2 TCH 319 TCH 4 TCH 5 TCH 6 TCH 7202122232425 TRXsig2627 BCFsig28293031 Q1-management

One TRX has dedicated TRXsig of 16, 32 or 64 kbit/s.

48 kbit/s isnot allowed.One BCF has dedicated BCFsig (16 or 64 kbit/s) for O&M

TRX1

Q1-management needed if TRS management under BSC

MCB/LCB required if loop topology is used

AbisBTS BSC

(E)GPRS Dynamic Abis Pool – EDAP Introduction

• Fixed resources for signaling and voice

• Dynamic Abis pool (DAP) for data

– Predefined size 1-12 PCM TSL per DAP

– DAP can be shared by several TRXs in the same BCF (and same E1/T1)

– Max 20 TRXs per DAP

– Max 480 DAPs per BSC

– DAP + TRXsig + TCHs have to be in same PCM

– UL and DL EDAP use is independent

– DAP schedule rounds for each active Radio Block

– Different users/RTSLs can use same EDAP Sub-TSL

0 MCB LCB1234 TCH 0 TCH 1 TCH 2 TCH 35 TCH 4 TCH 5 TCH 6 TCH 76 TCH 0 TCH 1 TCH 2 TCH 37 TCH 4 TCH 5 TCH 6 TCH 78 TCH 0 TCH 1 TCH 2 TCH 39 TCH 4 TCH 5 TCH 6 TCH 7

101112131415 EDAP EDAP EDAP EDAP16 EDAP EDAP EDAP EDAP17 EDAP EDAP EDAP EDAP18 EDAP EDAP EDAP EDAP19 EDAP EDAP EDAP EDAP20 EDAP EDAP EDAP EDAP21 EDAP EDAP EDAP EDAP22 EDAP EDAP EDAP EDAP232425 TRXsig1 TRXsig226 TRXsig327 BCFsig28293031 Q1-management

TRX1

TRX2

TRX3

EGPRSpool

Packet Control Unit (PCU) - Introduction• BSC plug-in unit that controls the (E)GPRS

radio resources, receives and transmits TRAU frames to the BTSs and Frame Relay packets to the SGSN

• Implements both the Gb interface and RLC/MAC protocols in the BSS

• Acts as the key unit in the following procedures:– (E)GPRS radio resource allocation and

management– (E)GPRS radio connection establishment and

management– Data transfer– Coding scheme selection– PCU statistics

• The first generation PCUs are optimized to meet GPRS requirements, i.e. non real time solutions (QoS classes "Background" and "Interactive“)

• The second generation PCUs (PCU2) supports the real time traffic requirements and enhanced functionality (GERAN) beyond (E)GPRS

Gb Interface - Introduction• The Gb interface is the interface between the BSS

and the Serving GPRS Support Node (SGSN)• Allows the exchange of signaling information and

user data• The following units can be found in Gb

– Packet Control Unit (PCU) at the BSS side– Packet Processing Unit (PAPU) at the GPRS IP

backbone side

• Each PCU has its own separate Gb interface to the SGSN

BSC

PCU

BSS

SGSN

PAPU

GPRS

Gb

Gb Interface• Allow many users to be multiplexed over the

same physical resource• Resources are given to a user upon activity

(sending/receiving)• GPRS signaling and user data are sent in the same

transmission plane and no dedicated physical resources are required to be allocated for signaling purposes

• Access rates per user may vary without restriction from zero data to the maximum possible line rate (e.g., 1 984 kbit/s for the available bit rate of an E1 trunk)

BSC

PCU

BSS

SGSN

PAPU

GPRS

Gb

RF PLANNING VS DATA PERFORMANCE

CONTENTS

• FREQ. PLANNING

• C/I VS THROUGHPUT GRAPHS

Frequency PlanningCombined interference and noise estimations needed for (E)GPRS link budget

Frequency allocation and C/I level• The existing frequency allocation has high impact on EGPRS performance• Loose re-use patterns will provide better performance for all MCSs

Data rate and network capacity• EGPRS highest data rates require high C/I, typ > 20dB for MCS-7, 8 & 9• Possibly no extra spectrum for EDGE so efficient use of the existing spectrum is very important• EGPRS traffic suited to BCCH use - typically the layer with highest C/I. But limited no. of TSLs

available on BCCH; may need to use TCH layer too

Sensitivity in tighter reuse and higher load• EDGE can utilize tighter reuse schemes and this is beneficial when planning for high load with

limited frequency resources• For systems with stringent spectrum constraints, EGPRS can offer good performance even with

tight re-use patterns (1/3 or 3/9). Load dependent

Data rate vs. CIR in Time (Field Measurement)

0

20

40

60

80

100

120

140

0 10 20 30 40

Time (s)

Thr

oug

hpu

t (kb

ps)

0

5

10

15

20

25

CIR

(dB

)

Data ThroughputApplication Throughput

TEMS-C/I-GMSKPoly. (TEMS-C/I-GMSK)

Good quality environment

Data rate vs. CIR in Time (Field Measurement)

0

20

40

60

80

100

120

0 10 20 30 40 50 60 70

Time (s)

Thr

oug

hpu

t (kb

ps)

0

5

10

15

20

25

CIR

(dB

)

Data ThroughputApplication Throughput

TEMS-C/I-GMSKPoly. (TEMS-C/I-GMSK)

Average quality environment

Data rate vs. CIR in Time (Field Measurement)

0

10

20

30

40

50

60

70

80

0 50 100 150

Time (s)

Thr

oug

hpu

t (kb

ps)

0

2

4

6

8

10

12

14

16

18

20

CIR

(dB

)

Data ThroughputApplication Throughput

TEMS-C/I-GMSKPoly. (TEMS-C/I-GMSK)

Worse quality environment

Q&A

Thank You