Title: UMTS Core Network Overview Core Network Architecture and
1UMTS Core Network OverviewCore Network
Architecture and Migration
2Products for the new Core Network Architecture
3Ericsson server-gateway split migration path
4Ericsson server-gateway split migration path
- AXE-10 system comprises both the network control
and connectivity layers in the same node. It is
important that these already deployed nodes can
be upgraded in an evolutionary way. - In a first step the AXE-10 MSC will be divided
internally into a server and a MGW. - In the next step, a physical split will be made.
- The MSC server will now be based on
AXE-10/AXE-810 and the MGW will be based on the
new developed Cello Packet Platform (CPP).
5AXE10 Platform Evolution, Upgrade Existing Nodes
AXE10
MAP
BSSAP
N-ISUP
GCP
(to other MGWs)
RANAP
MGW Handler
GCP
MSC Server
int. i/f
Media Gateway Function
int. i/f
Connection Handler
Payload Processing Switching
User data
User data
MSC/VLR
6- Ericssons strategy for server development is
building on our existing GSM/GPRS products. - The internal structure of the MSC/VLR makes it
easy to perform a logical split of the
functionality into server and MGW functionality. - Support for the new Iu interface (RANAP) is added
- Add functions for controlling Media Gateways (GCP
to control other MGW) - There is also an internal interface to enable the
MSC Server to control its own Media Gateway
functionality. - The Media Gateway part handles the payload
processing and switching for the user data.
7WPP Platform Evolution, Upgrade Existing Nodes
WPP
MAP
GTP-C
Gb
GCP
(to other MGWs)
RANAP
GCP
MGW Handler
SGSN Server
int. i/f
Media Gateway Function
int. i/f
Connection Handler
User data
User data
Payload Processing Routing Re-tunneling
SGSN
8New MSC and SGSN servers and MGW
MSC server
SGSN server
MAP
MAP
N-ISUP
MSC control logic (APT)
SGSN control logic
GTP-C
RANAP
RANAP
MGW Handler
MGW Handler
MSC Server
SGSN Server
GCP
GCP
GCP
Connection Handler
Payload Processing
Routing/switching
MGW
9New Media Gateway
Media Gateway concept
Media Gateway
- Key element in the connectivity/transport layer
- MGWFGGSN
- At the edge of the Core Network connectivity
layer - -- connect the Core Network to access networks
and to external networks - Optimised for packet transmission
- -- It supports ATM, IP as well as STM
GGSN
MGWF
10New Media Gateway(Continued)
Media Gateway concept
Media Gateway
- Embedded Real Time IP Router
- ATM/AAL2 switch
- --to bear compressed voice
- Interface functions for different transport
standards and bitrates - -- different transport standards -- ETSI, ANSI
and so on - -- different bitrates -- for example from 1.5/2
Mbps up to 155 Mbps, STM, ATM, Ethernet
GGSN
MGWF
11New Media Gateway
MGWF
Media Gateway
- Speech processing
- The transcoders can be placed at the edge of the
Core Network - allowing significant savings in
transmission. - Setup/release of user data bearers
- QoS IP routing switching
- Data volume counting
- Security for packet based traffic
GGSN
MGWF
12New Media Gateway
GGSN
Media Gateway
- IP address allocation
- --Address translation (translates IP address to
IMSI for downlink IP packets received from the
external network) - --IP address allocation
- --Packet forwarding
- QoS IP routing
- Secure connections to ISPs/Intranets and
Internet - Volume based charging (CDR generation)
- Mobile IP, Foreign Agent
- --Support for a (Mobile IP) Foreign Agent (FA).
This FA enables roaming with other access
networks, for example CDMA 2000 and fixed, using
the IETF Mobile IP standard.
GGSN
MGWF
13Ericsson MSC Server
- Based on the well proven AXE 10 platform in use
in more than 100 countries - Non-stop operation
- -- AXE10 Hardware redundancy allows non-stop
operation that is for replacing Hardware Devices - In-service upgrade
- -- Since AXE10 CP/RP are duplicated, the new SW
on one side (Standby) can be loaded while the
other side is still running the old SW
(Executive). - Scalable, load-sharing and fault tolerant
architecture with built-in redundancy
CP Cabinet
14R9 Solutions
15- This is AXE10 in R9 for MSC Server ..
- ATM-IWU is ATM InterWork Unit, is used as a
interface for UMTS, such as for GCP - IPNX, IPNA is for supplying Ethernet int.
163G MSC Server and TSC Server Nodes
- The TSC and MSC Server nodes will have the same
AXE10 HW platforms. Only the dimensioning of
these nodes is different. - MSC-server (GS connected) for CN2.0 with 300k
subscribers (not to scale) Signalling Bandwidth
(with overhead) 2 3 Mbit/s-- 2 Mbit/s
signalling capacity per ALI unit, one redundant
ALI-unit in GDM magazine-- IPNX Ethernet switch
with 8 ports used only for CP-APG40 connection. - It is expected that a non GS version of the ALI
unit only for signalling will be developed.
17Ericsson SGSN Server
- Carrier class performance
- Fault tolerant architecture with built-in
redundancy - Non-stop software upgrading
- Scalable capacity
- Web-based OM
18Ericsson Media Gateway
- Multiprocessor architecture with scalable
capacity and robustness - In-service SW/HW upgrade support
- ATM and AAL2 switching
- STM, AAL1, AAL2 and AAL5 termination services
- High capacity real-time IP router
- Web based element management
- On-line documentation
Subrack with boards
193G Backbone network
Media GW
IP backbone AXI 500 products
ATM backbone AXD 300 products
AXD301
AXI 540
Media GW
Media GW
AXI 520
CELLO
CELLO
Media Gateway CELLO products
Media GW
20- Cello Media Gateway on the edge of the backbone
network, connecting with other networks. - Also have a number of ATM and IP products which
complement the Media Gateway and form the
backbone network. - For an IP transport backbone, we have the AXI 500
family of products. - AXI 540 -- a gigabit Edge Aggregation Router
which aggregates traffic from several access
routers. - -- AXI 540 is from Ericsson.
- AXI 520 -- an IP backbone router with speeds of
20 Gbit/s (or 40 Gbit/s if throughput is measured
bi-directionally). From the AXI 540, the traffic
goes to the highest level(AXI 520) in the
backbone. - -- AXI 520 is form Juniper Ericssons
partnership. - For an ATM backbone, we have the AXD 300
products. - The AXD301 is a scalable, high performance
multi-service ATM switching system with switching
capacity that is scalable from 10 Gbps up to 160
Gbps. (High availability, upgrades from smaller
to larger switch sizes can be done without
interruption of ongoing services).
21TDM Connectivity Network
22Main Limitations of TDM Transmission Technology
As Connectivity Backbone
- Very inflexible bandwidth allocation. Limited to
64kbps or nx64kbps channels. - The CSS nodes do not support subrate switching.
So coded speech cannot be transported
efficiently. - Introduction of subrate switching is regarded as
waste of money. - Supports only constant bit rate transport, or
complex dynamic rate adaptation needed. - No statistical savings possible, unused bandwidth
cannot be used for other type of traffic. - There is no separate protocol available for
set-up of TDM bearers. - Very tied coupling with the call control
signalling. So only hop-by-hop routing possible.
23ATM Connectivity Network
24ATM (Asynchronous Transfer Mode)
- ATM is a transmission technology that uses fixed
size packets called cells. - ATM networks are connection-oriented -- So a
connection (corresponding to a fixed route and
reserved resources) must be established before
data transfer can take place. Data is guaranteed
to arrive in sequence. - Public ATM networks typically use SDH as the
transport technology. - ATM is a packet mode technique, but the delay in
the network can be kept to a minimum because the
cells have a fixed length.
25ATM Cell format
A cell is a 53-byte packet 5 bytes of
header/descriptor 48 bytes of payload
26The contents of an ATM cell header
27- The cell header is divided into different fields
- VPI(Virtual Path Identifier)VCI(Virtual Channel
Identifier) - -- As address field, Specify a logical channel
number/identity. - -- Identifies the circuit and provides a unique
link address between two network nodes. - PTI (payload type identifier)
- -- Specifies whether the cell contains user
information or information to be used by the
network itself for operation and maintenance. - CLP (cell loss priority)
- -- Specifies the priority level of the cell (out
of two possible levels) if there is not space
enough for all cells. - HEC (header error control)
- -- Contains a check value, which is used by
nodes in the network and at the receiving end to
detect any distortion of the header (bit error).
28Labelled multiplexingin ATM allows flexible use
of bandwidth
An ATM network consists of ATM nodes and links.
A stream of ATM cells that carry information for
the corresponding services is transferred in each
link. Unused space in the cell stream consists
of empty cells, called idle cells
29The principle of ATM switching
30- In an ATM switch, ATM cells are transported from
an incoming logical channel to one or more
outgoing logical channels. - A logical channel is indicated by a combination
of two identities - 1. The number of the physical link
- 2. The identity of the channel the virtual path
identifier (VPI) and the virtual channel
identifier (VCI) on the physical link. - Switching of cells through an ATM node requires a
tie between the identities of incoming and
outgoing logical channels. - Two transport functions that are needed in the
ATM switch are described below they are also
compared with the corresponding functions in a
circuit -mode switch.
31- The first function can be compared to the change
of time slot numbers in circuit-mode switch. This
is the function that transfers a voice sample
from an incoming time slot to an outgoing time
slot. - In an ATM network, the identities of the
different logical channels correspond to the time
slots. - The identity is composed of two values in two
different fields in the header of the cell VPI
and VCI. They have the same task as the time slot
in a circuit-switched system that is, to
identify each individual connection on each
physical link between two nodes. - In the Figure above, the digits in the cell
header specify the channel identity ( VPI and
VCI) - The second function can be compared to the
function for space switching in a circuit-mode
switch. - In Figure above, by payload A being shifted from
the upper physical input to the lower output, and
vice versa for payload B.
32Protocol Reference Model
- The three lowest layers in the protocol reference
model are - Layer 1, the physical layer
- Layer 2, the ATM layer and
- Layer 3, the AAL layer.
33ATM adaptation layers and classes
- To enable transfer of both data,voice and
multi-media services, the information must be
adapted to the network in different ways. - ATM has been divided into four service classes
(A, B, C and D) on the basis of three
parameters(synchronous or asynchronous
services,constant or variable bit
rate,Connection-oriented or connectionless
transfer). - Four protocols (AAL 1, AAL2, AAL 3/4 and AAL 5)
are defined for each one of the classes. - Note that AAL is not part of the cell header.
34Segmentation and multiplexing of different
broadband services
Information flow of different services is
segmented and packed into the information field
of ATM cells, Then be multiplexed into one
common cell flow with a larger bandwidth.
35AAL1
36AAL1
- AAL 1 for the realisation of circuit-switched
connections with constant bit rate and minimal
delay. - AAL 1 supports class A services i.e. voice and
video traffic. - Since voice traffic is error tolerant, no error
control (CRC) is required. - However, what is important in the case of voice
transmission is that cells are received in the
exact sequence in which they were sent, and that
they arrive at a constant rate gt AAL1 assures
sequence numbers.
37AAL2
- AAL2 can be used multiplex several mini-cells
into an ATM cell payload. - AAL2 primarily used to the transport of
compressed voice. - Why use AAL2?
- Lets imagines a stream of voice information as
water flowing from a tap and the ATM cell payload
as a bucket. - Then that the payload will take a finite amount
of time to fill it -- this is called
packetization delay. - At 64 kbps(uncompressed voice) the packetization
delay is approximately 6ms. - If the speech is compressed, the data rate is
reduced. This will increase the amount of time it
takes to fill the payload. gt packetization
delay will be lengthened . If speech rate is 8
kbps(AMR), the packetization delay is as high as
47 ms.
38Packetization Delay
39- Delay budget is used to avoid echo and unnatural
speech perception. - The delay budget can be as small as 30ms.
- It is totally unacceptable to have a
packetization delay that is such a large
proportion of the total delay budget. - To reduce the impact of packetization delay,
mini-cells from several users are multiplexed
onto a single ATM payload. - A unique Channel ID (CID) identifies each stream
of data.
40AAL2 Multiplexing
41AAL2 Logical Link Connections
- Each stream of data is referred to as an AAL2
Logical Link Connection (LLC). - Several LLCs can be on the same established ATM
Virtual Channel Connection - An AAL2 LLC is defined as a bi-directional
point-to-point connection. The same CID value is
used in both directions. The set of available
CIDs at each end is known. - The use of LLCs introduces another level of
multiplexing into the ATM network. - The CID is similar to a CIC value in SS7 the
major difference being that the CID value does
not represent a physical circuit, but rather a
logical connection
42AAL2 header
43AAL2
- CID Field
- The channel identifier field identifies the
individual user channels within the AAL2, and
allows up to 248 individual users within each
AAL2 structure. The CID field is actually 8 bits,
thus allowing a theoretical maximum of 255
individual users. However, several CID values are
reserved for management functions and future
functions, hence the figure of 248 users. - LI Field
- The length identifier identifies the length of
the packet payload associated with each
individual user, and assures conveyance of the
variable payload. The value of the LI is one less
than the packet payload and has a default value
of 45 octets, or may be set to 64 octets.
44AAL2
- UUI Field
- One current use for the User-to-user field is to
negotiate a larger Maximum Transfer Unit (MTU)
size for IP. This function was originated from
third generation mobile telephony standard,
Universal Mobile Telephony Standard (UMTS)
development. AAL2 is important as it will be
fundamental to this UMTS standard. Carrying IP
traffic is perceived as a major application in
UMTS. - STF
- The Start Field identifies the location of the
start of the next packet within in the flow.
45AAL5, variable bit rate
46Step 1 Establishment of ATM VCCs
47Step 2 Establishment of AAL2 Connections
48IP Connectivity Network
49MPLS solution
The Multi-Protocol Label Switching (MPLS)
protocol is an interstitial, layer 2.5 protocol
which complements and enhances the IP protocol,
in that it offers an alternative method of
forwarding IP packets, while reusing the existing
IP routing protocols (e.g., OSPF, BGP). MPLS can
run on top of numerous L2 technologies
(PPP/Sonet, Ethernet, ATM, FR, WDM Lambdas, etc.)
50MPLS
- For the first step into migration to an all IP
core network we propose the usage of MPLS paths
(LSPs) end to end between the nodes of the CN
(e.g. MGW to MGW). Each LSP carries traffic from
only one specific service class. The LSPs are
created with a specific set of characteristic
parameters (carried in MPLS signalling) according
to the type of traffic (voice, signalling, etc)
it is intended to carry. - LSPs may be set up either static or dynamic
according to the actual bandwidth need between
the MGWs. The bandwidth for the tunnels is
reserved and modified (increase/decrease) on an
aggregated basis (and not on a call-by-call base,
H.245) thus minimising the signalling traffic in
the core network.
51MPLS
- Moreover the operators have the ability to
perform proper traffic engineering, by defining
alternative paths in their network to provide
better load control on the network nodes but also
to redirect traffic to predefined paths making
best use of the network resources also in case of
network failures. - Another advantage of using LSPs is the enhanced
security the operator can apply inside the core
network, which is more relevant if the network is
used for other than UMTS services too.
52How does MPLS work?
LSP
- LER - ingress examines IP header, classifies,
assigns MPLS header and forward towards the next
hop. - LSR - a router that examines the MPLS header only
and based on the forwarding table sends packet to
next LSR. IP header never examined at an LSR. - LER - egress removes MPLS header, examines IP
header, forwards to destination using IP routing. - MPLS label a shim header used for forwarding and
queuing decisions at the LSR - MPLS domain LER LSR LSR LSR LER
- LSP MPLS path between ingress LER to egress LER
53MPLS Label ...
54MPLS Forwarding Table ...
Interface
Label In
Label Out
Interface
O3
I2
27
8
O4
I2
31
12
I1
O2
O3
Label 27
LSR
Label 8
I2
O4
Label 31
Label 12
55- Router A computes path to Router D, shortest
path is A-F-D with metric 20
- Router B and Router C are underused
- SGSN/GGSN traffic results in congestion and
packet loss in Router F
IP1
IP1
IP1
IP1
IP1
IP3
IP3
IP3
IP3
IP3
56- Routing packets through LSP, MPLS forwarding
table
- Better bandwidth usage, less congestion and
packet loss
IP1
IP1
IP2
IP2
IP3
IP3
IP3
IP3
57Label Distribution with CR-LDP
58Establishment of Logical Channel per Call
59Other Nodes
Service enablers
Services/applicationlayer
Application
Servers
Application
Servers
GMSC/Transit
MSC
SGSN
HLR/AuC/FNR
Server
Server
Server
SGW
Control
PSTN/ISDN
MGW
Connectivity
InternetIntranets
MGW GGSN
GSM
EDGE
WCDMA
Control
User data
60Protocol Stack Interwork in Cello SGW
SCTP Stream Control Transmission Protocol
RFC2960 M3UA MTP3 Users Adaptation layer
61CN-OSS
62Product and Network Migration
63MSC/GSN Product Migration
Installed base
New generation product line
MSC
GSN
Media Gateway
64MSC architecture today
MSC Control
A
Switching
65MSC Server/Media Gateway architecture
MSC Server
A
MGWF
66Introduce Media Gateway Connect UTRAN
67New stand-alone MSC server
MSC Server
MSC Server
A
MGWF
Iu
68Same model applies to GPRS
SGSN Server
GTP-C
SGSN Server
Gb
BSS
MGWF
GGSN
MGW
Iu
UTRAN
69GSM, GPRS, UMTS combined
SGSN Server
MSC Server
GCP
MGWF
A,Gb
MGWF
Iu
70Later release - New GSM traffic to MGW
GSN Server
MSC Server
GCP
MGWF
A,Gb
MGWF
GSM growth
Iu
71Support of multi-vendor environment-
interworking examples
Common HLR
SGSN Any vendor
MAP
GGSN
GTP
GGSN Any Vendor
Iu
SGSN Server
H.248
Radio Access Network
MGW
MGW
ISUP
MSC Server
GMSC/ Transit Server
ISUP
GMSC/ TSC Any Vendor
MSC Any Vendor
User data Signalling
72Multi-vendor environment - layered network
architecture
SGSN Server
HLR
MAP
MAP
BICC/ISUP
GMSC/ Transit Server
MSC Server
BICC/ISUP
H.248, GTP-C
RANAP
H.248
UTRAN
PSTN/ISDN PLMN Internet/ Intranet
MGW/ GGSN
MGW
GTP-U,...
Connectivity Backbone NW ATM, IP...
73T h a n k Y o u !
74Welcome...
- Instructor Edward Liu Lu
- Senior Training Engineer
- Core Network Section,
- ETC/F/CD
- E-mail address
Edward.liu_at_etc.ericsson.se - Tel. No 65615599-13912