LTE Architecture Part 2: EPS Architecture

Summary

This script offers an insightful journey into the LTE and EPC architecture, using the analogy of a road trip from Istanbul to Izmir in Turkey. It explains the network components like eNodeB, Serving Gateway, P-Gateway, MME, and HSS, and their roles in managing data flow and mobility. The interfaces S1, S5, and S11 are detailed, highlighting the separation of control and user planes. The script also delves into roaming concepts, explaining how LTE supports seamless connectivity across different networks and countries, and touches on the evolution of voice services in LTE networks.

Takeaways

• 🚀 The LTE and EPC architecture is explained through an example of a car journey from Istanbul to Izmir, Turkey, illustrating the connectivity and mobility aspects of the network.
• 📡 The Enhanced Node Base Station (eNodeB) is the term for the base stations and antennas that provide connectivity along the route.
• 🌐 To access the internet, a user needs an IP address, which is hosted at the P-Gateway, allowing for continuous connectivity even when moving between base stations.
• 🔄 The Serving Gateway (S-Gateway) acts as an anchor for mobility, ensuring seamless transitions between base stations without the P-Gateway being aware of each movement.
• 🔗 Interfaces S1-U and S5 are crucial for the architecture, with S1-U connecting the eNodeB to the S-Gateway and S5 connecting the S-Gateway to the P-Gateway.
• 🛡️ The Home Subscriber System (HSS) is responsible for storing user credentials, identity, and subscription data, essential for network access authentication.
• 👮‍♂️ The Mobility Management Entity (MME) manages the authentication signaling and mobility between base stations, reducing the traffic to the HSS and hiding user mobility.
• 🤝 The X2 interface facilitates handovers between base stations, while the S10 interface is used for handovers between different MMEs controlling subsets of base stations.
• 🌍 Roaming is a key feature of LTE, allowing users to maintain connectivity when outside their home network, with the architecture supporting roaming through various mechanisms.
• 🔑 The user's identity in LTE contains the domain name of the home operator, which is crucial for authentication and roaming scenarios.
• 📈 The architecture supports both home-routed traffic and local breakout (LBO) for roaming, with the latter being common for voice services to reduce latency.

Q & A

What is the LTE and EPC architecture?

The LTE and EPC (Evolved Packet Core) architecture is a network structure used in telecommunications that includes the Evolved Universal Terrestrial Radio Access Network (E-UTRAN) and the EPC, which handles the core network functions.

What is the role of eNodeB in LTE architecture?

The eNodeB, or enhanced Node Base Station, is responsible for the radio communication with the user equipment. It is the base station that provides the antennas and facilities for wireless signal transmission along the route of the user's journey.

Why is a P-Gateway necessary in the LTE architecture?

The P-Gateway (Packet Data Network Gateway) is necessary because it serves as an anchor point for the user's IP address, allowing the user to maintain a consistent IP address even as they move between different base stations.

What is the purpose of the Serving Gateway in LTE?

The Serving Gateway (S-Gateway) anchors the user's mobility between different eNodeBs, ensuring that the user's IP packets are routed correctly without the P-Gateway needing to be aware of each base station change.

What are the S1, S5, and S11 interfaces in LTE architecture?

The S1 interface connects the eNodeB to the S-Gateway, the S5 interface connects the S-Gateway to the P-Gateway, and the S11 interface is used for signaling messages between the Mobility Management Entity (MME) and the S-Gateway, primarily for handling mobility during handovers.

What is the function of the Home Subscriber Server (HSS) in LTE?

The HSS stores the user's subscription data, including their identity, username, password, and other relevant information, and is crucial for the authentication process when a user tries to access the network.

What is the Mobility Management Entity (MME) and its role in LTE?

The MME is responsible for managing the mobility of the user equipment within the LTE network. It handles the signaling for authentication, and controls the handover process between different eNodeBs.

What is the difference between the user plane and control plane in LTE?

The user plane is responsible for carrying the actual user data traffic, while the control plane manages the signaling and control messages that are essential for the operation of the network, such as handovers and authentication.

What is the significance of the X2 interface in LTE?

The X2 interface is a direct connection between two eNodeBs and is used for handover procedures, allowing the transfer of user data and control information during the transition from one base station to another.

What is roaming in the context of LTE and how does it work?

Roaming in LTE refers to the ability of a user to use their mobile services in a country or network different from their home network without any special configuration. The network identifies the user's home operator and authenticates them, allowing access to services while sharing revenue between the home and visited operators.

What are the two main types of roaming architectures in LTE?

The two main types of roaming architectures in LTE are home-routed traffic, where the user's IP address and P-Gateway are in the home network, and local breakout (LBO), where the user's traffic is routed through the visited network's P-Gateway for services like Voice over LTE.

Outlines

00:00

📡 Introduction to LTE and EPC Architecture

This paragraph introduces the LTE (Long-Term Evolution) and EPC (Evolved Packet Core) architecture through an analogy of a road trip from Istanbul to Izmir in Turkey. The explanation begins with the necessity of antennas and base stations (eNodeB) for connectivity and the concept of a Packet Data Network Gateway (P-GW) that maintains a constant IP address for the user as they move between base stations. The architecture includes a Serving Gateway (S-GW) to anchor mobility, making the transition between base stations invisible to the P-GW. The paragraph also outlines the interfaces S1-U and S5, distinguishing between the data plane (solid lines) for IP packet transfer and the signaling interface (dotted lines) for control messages. The Home Subscriber System (HSS) is introduced as the storage for user credentials and subscription data, emphasizing the importance of authentication in accessing the network.

05:01

🚀 Understanding Mobility Management in LTE

The second paragraph delves into the mobility management aspect of the LTE network. It discusses the role of the Mobility Management Entity (MME) in handling authentication signaling and its interface with the Home Subscriber System (HSS). The MME is portrayed as a proxy that shields the HSS from the traffic generated by mobile users as they move between base stations. The paragraph also explains the S11 interface, which is crucial for controlling handovers between base stations. The concept of the X2 interface for direct communication between eNodeBs during handovers and the signaling involved in this process is introduced, providing a comprehensive view of how the LTE network manages mobility and maintains seamless connectivity.

10:03

🌐 Exploring Interfaces and Terminology in LTE

This paragraph provides an in-depth look at the various interfaces and terminologies used in the LTE architecture. It explains the S5 interface between the Serving Gateway (S-GW) and the Packet Data Network Gateway (P-GW), which includes both user plane and control plane aspects. The paragraph also introduces the S10 interface for handling mobility between different MMEs controlling distinct sets of base stations. Key terms such as the Evolved Universal Terrestrial Radio Access Network (E-UTRAN), User Equipment (UE), and the LTE interface (Uu) are defined, offering clarity on the components and interactions within the LTE system. The architecture diagram in specifications is briefly mentioned, highlighting the complexity and detail-oriented nature of the LTE system design.

15:03

🔄 Roaming and Its Impact on LTE Architecture

The fourth paragraph discusses the concept of roaming within the LTE architecture. It explains how roaming is facilitated by the inclusion of the home operator's domain name in the subscriber's identity, allowing the visited network to authenticate and connect the roaming user. The paragraph details the roaming architecture, distinguishing between home routed traffic, where the P-GW is located in the home network, and local breakout, where traffic exits through the visited network. The importance of inter-operator billing and the role of the MME in identifying roaming users are also covered, providing insights into the operational aspects of LTE networks when it comes to roaming.

20:05

🌍 Roaming Architectures and Their Operational Dynamics

This paragraph expands on the operational dynamics of roaming architectures in LTE networks. It contrasts the non-roaming architecture, where all network entities are located within the home network, with roaming scenarios. The paragraph describes two common roaming architectures: home routed traffic, where the P-GW remains in the home network, and local breakout, where traffic is routed through the visited network. The discussion includes the roles of visited and home operators in authentication, traffic management, and revenue sharing. It also touches on the importance of local breakout for voice services, emphasizing the need for quick access to the internet and the preference of operators to manage traffic for billing accuracy.

25:07

📞 LTE as a Packet-Switched Domain and Voice Support

The final paragraph addresses misconceptions about LTE's support for voice services. It clarifies that while LTE is a packet-switched domain and does not support traditional circuit-switched voice calls, it does support Voice over LTE (VoLTE). The paragraph explains the distinction between the two domains and highlights the humorous video about the misconception for further exploration. It also mentions the separation of control and user planes in LTE, with the MME handling control plane functions and the S-GW and P-GW managing user plane traffic. The paragraph concludes by emphasizing LTE's design as a packet-switched network, which is a departure from the 3G architecture that included a separate entity for control functions.

Keywords

💡LTE

LTE stands for Long-Term Evolution, which is a standard for wireless broadband communication for mobile devices and data terminals. It is a 4G technology that provides high-speed data and voice services. In the video, LTE is the main theme, as the script discusses the LTE architecture and its components, illustrating how it enables high-speed internet access while traveling from Istanbul to Izmir in Turkey.

💡EPC

EPC stands for Evolved Packet Core, which is the core network architecture used in LTE networks. It is responsible for data transmission and is designed to handle the increased data demands of mobile broadband. The script explains the EPC architecture through the analogy of a car journey, highlighting its role in managing data flow and maintaining IP connectivity across different base stations.

💡eNodeB

eNodeB, or enhanced Node B, is the term used for the base stations in an LTE network. They are responsible for the radio communication with the user equipment. The script mentions eNodeBs as the initial point of contact for a user wishing to connect to the internet, emphasizing their importance in providing coverage along the route from Istanbul to Izmir.

💡P-Gateway

The P-Gateway, or Packet Data Network Gateway, is a node in the EPC architecture that connects the LTE network to external packet data networks such as the internet. The script describes the P-Gateway as the host for the user's IP address, routing internet packets to and from the user's device.

💡Serving Gateway

The Serving Gateway, or S-Gateway, is an anchor point in the EPC architecture that manages the user's mobility between different eNodeBs without the need to update the P-Gateway with every movement. The script explains how the Serving Gateway facilitates seamless connectivity as the user moves from one base station to another.

💡HSS

HSS stands for Home Subscriber Server, which is a database in the EPC architecture that stores user-related information such as authentication credentials and subscription data. The script mentions the HSS in the context of storing usernames and passwords, and its role in authenticating users who wish to access the network.

💡MME

MME stands for Mobility Management Entity, which is responsible for managing the mobility of users within the LTE network. It handles the signaling and control aspects of user mobility. The script describes the MME as a proxy entity that manages authentication signaling and communicates with the HSS for user identity validation.

💡S1-U

S1-U is the interface name between the eNodeB and the Serving Gateway in the EPC architecture. It is responsible for the user plane data transfer. The script refers to S1-U as the interface that carries the data packets between the base station and the Serving Gateway.

💡S5

S5 is the interface between the Serving Gateway and the P-Gateway in the EPC architecture. It includes both user plane and control plane functionalities. The script explains S5 as a key interface that allows internet packets to reach the Serving Gateway and then be forwarded to the appropriate eNodeB.

💡Roaming

Roaming refers to the ability of a mobile user to automatically and seamlessly connect to a visited network when they are outside their home network's geographic coverage area. The script discusses roaming in the context of LTE, explaining how the architecture supports users making calls and browsing the internet in a different country without additional setup.

💡E-UTRAN

E-UTRAN stands for Evolved Universal Terrestrial Radio Access Network, which is the radio access network part of the LTE system. It includes the eNodeBs and the interface to the user equipment. The script uses E-UTRAN to describe the radio access network that includes the base stations and the connection to mobile devices.

Highlights

Introduction to LTE and EPC architecture through the example of a car journey from Istanbul to Izmir in Turkey.

Explanation of eNodeB (enhanced Node Base Station) as the base stations and antennas providing connectivity along the journey.

The role of the P-Gateway in hosting the IP address and routing internet packets to the user's device.

The Serving Gateway's function as an anchor for mobility between base stations without visibility to the P-Gateway.

Interface names and their purposes in the LTE architecture, such as S1-U between eNodeB and Serving Gateway, and S5 between Serving Gateway and P-Gateway.

Differentiation between data plane (solid line) and signaling interface (dotted line) in the LTE architecture.

The Home Subscriber System (HSS) as the storage for user credentials and subscription data.

The Mobility Management Entity (MME)'s role in handling authentication signaling and mobility management.

The S11 interface's purpose for controlling handovers between base stations.

The X2 interface for direct communication between eNodeBs for handover purposes.

The EUTRAN term referring to the radio access network including eNodeBs and the interface to the mobile.

The User Equipment (UE) term for the mobile device, including the SIM card and mobile equipment.

The Evolved Packet Core (EPC) encompassing the MME, Serving Gateway, P-Gateway, and HSS.

The LTE system as a commercial term for the entire network, including both EUTRAN and EPC.

Roaming support as a key concept in LTE architecture, allowing users to maintain connectivity across different networks.

The distinction between home routed traffic and local breakout in roaming scenarios, impacting network architecture and operations.

LTE as a packet-switched domain only, with voice services supported through Voice over LTE (VoLTE).

Transcripts

now we will learn about the LTE and EPC

architecture which is the evolve packet

core architecture the way we will cover

it is through an example that's always

the best way to learn about any

architecture you're in the car and

you're traveling from Istanbul to Izmir

these are two cities in Turkey and it

will take you forever to get out of this

omble traffic so you're most of the time

going to be just in the precincts of

Istanbul you're sitting with your friend

who's driving and you are wanting to

connect to the internet what is the

first thing that you're going to see

will be a bunch of antennas and base

stations that need to be provided to you

all around all along the way on your way

to Izmir and these base stations and

antennas they are called een or DS or

enhanced node base station so we love

acronyms and so we create acronyms for

for them but these are essentially base

stations that that you are seeing you're

also wanting to connect to what's the

internet and for that you will need to

have an IP address so one way would

would be to give you an IP address at

each of these base stations but then

your IP address would change as you keep

going along so the way the architecture

works is that there is a gateway that

gets created which is called as P

gateway and your IP address is actually

host a home at this P gateway and all

your packets from the internet basically

come to this IP address over here and

then from this P gateway the packets

will make their way to the base station

so one of the way to get your package to

your base station would be to connect T

gateway to each of these base stations

and then your packets would come down

this way but this would also mean that

every time you move from one base

station to another base station your

your part of the packet would have to be

changed from the left to the right and

that would mean that every base station

that you move would be visible to the P

gateway in order to move away from that

the architecture which was created was

that there was another anchor that got

added and that

Anker is called the serving gateway the

serving gateway is the one that is

anchoring your mobility from one base

station to the other so that this

mobility between these base stations is

not visible to the packet gateway now

the things that you've seen is we have

drawn these lines which are showing the

interfaces that so call the interfaces

to the various nodes and so these are

the interfaces that's one and this and

the way the architecture works it calls

each of these interface by a name the

interface name between the base station

and the serving gateway is called s1 U

and the interface between the serving

gateway and P gateway is called s5

another thing you will notice is I've

shown solid lines and I've shown a

dotted line the solid line here is for

what's called your data plane so this is

where your IP packet goes both ways

it'll go up and then coming down from

the internet will come down this way and

the dotted line over here is actually

what's called a signaling interface so

there is some signaling traffic that

also goes and we will cover a little bit

more about the signaling traffic in a

few in the next slide set but the dotted

line is not your data packet but some

signaling which we will cover when

you're when you go beyond the particular

serving gateway now what you've got in

your IP packet and they're able to go

back and forth is this it is there

something else which is nothing yes the

most important part is how do you get

access to this network not anybody can

get access to this network and in order

to do that you need to have some

username and password and you need to

have a place where this username and

password is stored and that node is

called the HSS so that is your home

subscriber subsystem which stores your

username and your password and the

detail

of this username and password my writing

skills are terrible with the Spence will

be covered a little bit later this is

also your identity as it's called your

identity and your sub your your password

and your other subscription data so your

subscription data is stored in this HSS

over here now in order for you to get

authenticated you will need to

essentially send your credentials to the

HSS now here again from as you if you

had access the system through the C note

B your interface would have been through

here but then if you access it to

another e node B you would have to have

an interface from there so essentially

you could have an interface from each

one of these base stations to your home

subscriber subsystem to get yourself

authenticated this would again mean that

as you kind of move from one base

station to the other then you need to

get reoriented over here your HSS will

need to we'll see all this mobility and

the way the architecture was created was

to hide this mobility and also limit the

amount of traffic that goes to the HSS

in order to do that what was created was

this Mme which is called the mobility

management entity and if these acronyms

are there in the next slide so at the

moment just bear with me so this is the

proxy entity for which all your

authentication signalling and all your

signaling from your base station are

goes to and it is the one that then will

request for your identity and request

for your password validation from the

HSS so that's for your mobility

management entity there's actually one

more interface that that is from this

mobility management entity and that is

for control of your handovers itself and

that interface is called the s11

interface and this interface is

primarily for handling mobility as you

go from one base

station to the other this base station

when you come here has to tell the

serving gateway look I am serving this

Yui please send any packets coming from

the Internet to the Yui to me but in

order to do that he doesn't have a

direct signaling interface so the way

the signaling goes the signaling will go

wire the Mme and then the mme will tell

the serving gateway that hey look this

mobile is no longer at a node b1 he has

moved to e note b 2 so this s 11

interface over here is meant for

carrying signaling messages between the

mme to control the serving gateway this

is pretty much close to what the

architecture is there is one more

interface and that is the interface

between the base stations themselves and

this is used for handover so when you

move from this base station to to this

next base station what happens as some

of your internet packets may have

already come down to this old base

station in the meantime while you move

over here and that packet is transferred

from your old base station to the new

base station that's the solid blue line

the dotted blue line is again for

signaling messages for the space station

to tell this neighboring base station

hey this mobile is coming can we please

handle him and then this base station

goes back to the source base station

tell yes I can handle him and then the

mobile essentially moves over to the

space station this is the core part of

the EPC Eve or package core or the LT

architecture this is this

these are the main interfaces these are

the main network elements we thought

that the architecture was built of

network elements the serving gateway P

gateway a node B the mobility management

entity and the HSS and a bunch of

interfaces which are either signaling

interfaces or user plain interfaces or

data interfaces and for each of these

interface we give some interface name to

them s5 is the one between serving

gateway and T gateway and s5 includes

both user plain interface and a

signaling interface so now what happens

there's one more interface and that

we'll cover next is now you've gotten

out of Istanbul traffic sure you're

getting close to Manisa

now you're at this base station number

four you will essentially have the same

architecture as we had before where you

have a bunch of base stations that are

connected to each other they have an

interface x2 you may or may not have an

extra interface between all base

stations in most cases these interfaces

are present between all base stations

that you could expect to have an

interface like this but I showed a

general case where this interface does

not exist and your Mme is the one the

mobility management entity which is

handling the mobility between these base

stations and it also has an interface to

the HSS now when you move from this base

station to this base station you need it

to basically do a handover and for that

hand over there is yet another interface

that is needed and that's between these

EMA means themselves and that is the

interface called s10 interface so that

is used when you go from one subset of

base stations that are controlled by one

Mme to another Mme which is controlling

another set of base stations and this

interface over here is used for

providing signaling messages that the

hue is coming from this base station to

the next base station so please handle

it and also authentication information

about the mobile so this includes both

signaling for handover so this is both

for handover and also for passing the

user context okay there we go

I could write that here and you can

almost read it and that is what your

architecture is in general the

instantiation of an LTE architecture now

just for a few terminology that you will

see quite a bit

there are two terms which are used one

is what's called the EU trap and that is

your radio access network and this

includes your Eno's be another element

that we hadn't talked about is the

mobile itself and the mobile code is

called a user equipment you he and this

is a term you will see in specifications

being used and there is a the interface

between the mobile and the base station

which is called by the letter LTE you

view which is both includes a data plane

that carries the user traffic and also a

signaling for all the signaling about

what is it my power I need to handover

please connect me please give me more

please hand me over to the next base

station here on my measurements so these

are a bunch of signaling messages that

go between the mobile and the e note B

so you have the Evolve EU Trond which is

a bunch of base stations and also

includes your interface to to your

mobile and then the rest of the

architecture we just called your Evolved

Packet core which is including all of

your mme your serving gateway key

gateway and HSS and that's called EPC so

whenever you see EPC that's your core

over here and whenever you see EU Tron

that's your radio access network now

what is LTE so now where is LD the thing

is LT is actually a commercial term

which is typically not used in the

standards specifications themselves to

most people the entire LTE system means

this entire system but but as I said

this term is not really used in

specifications what you use is either

the or packet core or the award Neutron

which is evolved um Universal

terrestrial radio access networks we

love acronyms I'm sorry about that and

that is your architecture for your

Evolved Packet system as it is called

the typical diagram that you will see in

specification actually looks like this

which on the first look doesn't give you

any intuition as to how this network

works but for specifying the system this

is the the terminology which which gets

used and these diagrams are where you

have your essentially your network

elements as they call your base station

your serving gateway your P gateway your

Mme

the number of these base stations that

are there are not shown in this

architecture diagram what is shown is

that is that you could have an interface

between two e node B's and that's called

the x2 interface and we saw that you

could also have an interface between two

mm E's and that's the s10 interface so

you do see this kind of a line drawn

back to the same node itself and that is

specifying an interface between two of

these instantiations of your mobility

management entity the lines between

these network elements are called

interfaces or reference points and they

can either be a single protocol this one

just has a data plane or it could

include both a user plane and also a

control plane but they are still called

s5 sometimes in order for people to

refer to the user plane they will call

this the user plane as s5u and for the

control plane they will refer to it as

s5c but the interface itself both of

these together are just called s5 and

then you have interface out to the

internet with this SGI operators also

have their own private network and where

they offer you services and here what

the service means that things like volte

so this is this will host any of your

voltage servers that also interconnect

through your packet gateway down to to

your mobile which is on this side

okay at the high level the network

architecture in the specification

actually consists of two main elements

which is your mobile on the left-hand

side and your network on the right-hand

side the mobile on the left-hand side is

actually broken up into two parts one

which is your SIM card or which is also

called you sim which includes your

identity and your password your main

identity of who you are is in your SIM

card so if you do not have a SIM card

you cannot connect to the network well

you can get some connectivity only for

emergency and you also have your

password which is your called your main

key so that is built into your SIM card

and then the other part which is your

mobile equipment your phone that is

called the mobile equipment and these

two together are using the term user

equipment and that's what you will see

everywhere in specification the term you

eat together which which means both your

sim and your mobile equipment on the

network side you essentially have a

radio access network which is your in

ode B and your core network which we saw

was a combination of Mme serving gateway

P Gateway and HSS and this is how this

is what at the high level your brand as

people keep referred to and your core or

core network and ran a radio access

network the key concept of the LT

architecture is roaming and you will see

this roaming support in specifications

quite a lot well what is roaming and you

all know the roaming is essentially you

can take your mobile phone if you are in

subscriber of a Turkish operator let the

tour cell and you go visiting your

parents in the home and in your home

country for example if that's India

you're able to make phone calls you're

able to browse the internet though

sometimes that are much more expensive

right even when you're in India without

having to do something special on your

side and this support for roaming has

been built into the 3G

the architectures right from the gsm

days and similar concepts have also been

extended to LTE what what does roaming

really require it requires that when you

are in India that you're the operator is

able to figure out which operator you

belong to so who is your home operator

in one way and in the internet days you

always kind of say yes what is your

domain name and in a way so what's your

domain name of your identity so the way

the identity is created of each

individual of each subscriber it's in a

way that it contains the domain name of

the home operator and in the in DLP

attached lecture we look at what your

identity that your network knows you buy

and and and that contains your home

operators identity also in it apart from

just identifying your home operator also

the visited operator or the visited

country where your N and visited

operator needs to be able to

authenticate you so it should be able to

get to where your subscription data is

in your home country the subscription

data your keys

never leave your home country so your

your passwords are always in your home

network and and the most important thing

from an operator point of view is

obviously they want to be able to

sharing the revenues and what a lot of

times it's that expensive bill that you

see on your phone when you're like oh my

god is this so expensive sometimes you

don't make calls but nowadays the

roaming charges are coming down so the

architecture kind of allows these things

to happen or enable these things to

happen without without a human

intervention and this has kind of

impacts on the architecture itself let's

take a look at what our own on roaming

and roaming architecture would look like

and the non roaming architecture is what

we have seen so far it's it's the usual

concept that you are in your home

network and your

enter all your network entities are in

your home network so this isn't Turkey

this your internal operator all of these

entities are in in Turkey in in touch

cells

data centers of course these are out in

the field and these are out in various

data centers in in Turkey now you're out

visiting India for example you are in

the visited network it's called so

that's out in India the your HSS your

subscription information is actually is

still in Turkey it never moves out of

here so the Mme which is out in India

will have to they'll have to locate the

your home domain from from your identity

so when you attach you will have to

provide your identity from your identity

D visited Network figures out that you

are actually a roamer

as it's called you are Romer and it will

find out what your home HSS is and

authenticate you from the home addresses

and and also one of the most common

architectures and this is an

architecture which is called roaming

with home routed traffic is that your IP

address and your P gateway is actually

in the hole so this is roaming with home

routed traffic it it's also called it so

that's that's what this architectures

primarily and even though the

architecture kind of when you look at it

you say this looks very similar to this

it's just where things are located you

may be right but when we did develop the

architecture the whole concept that you

needed to have a local anchor in the bid

as an escape way was essentially also

driven a lot from a roaming architecture

so the need for this entity becomes much

more apparent when you say that you

don't want to show or you don't want to

be able to handle each and every base

station change out in

the whole network there are also things

that when you look into the details

there are ways in which charging or at

least charging not may not be at a user

specific level but at least at an

aggregation level also occurs in the

serving gateway in addition to the P

gateway so the home operator is doing

your home charging obviously and and and

it's going to create your your bill the

visited operator is also finding out oh

how many megabytes of traffic is

actually going out to this home operator

so that I can build a home operator so

that is inter operator billing so this

visited operator bills the home operator

saying of 17,000 of your subscriber came

to my network they used 45 megabytes

gigabytes hand me over the money so that

that's those are the key things there

there's also one more part in the

roaming architecture which we just kind

of said the first part was this this

home router traffic where your T

gateways in your home network and there

is a similar architecture where your

traffic is actually being going out the

internet from the visited operator and

that's what's called roaming with local

break up you can see the term LBO being

used in in the specification that's

local break up this architecture is

primarily used for voice because you

don't have to go back all the way to

your home network I mean India and

Turkey is not that bad suppose you're in

the United States and and your traffic

has to come all the way back to Turkey

and then be able to go out to the

internet most operators actually

deployed this architecture today and the

main main reason is that operators are

are not trusting of the non trusting

kind they want to make sure that they

are the ones who are owning your traffic

building your traffic making sure that

they they know exactly how many

megabytes you've used when you are in

the visited network so that when you

call up your operation say hey

got that huge bill I didn't use the

internet the way the way the bill says

and the operators they know you didn't

or you didn't and it has it's able to

get directed rather than depend on the

visited operator to provide the record

but if you're doing voice over IP kind

of a traffic then you want to make sure

that you're able to get out to the

Internet and be able to talk to your

correspondent node right away and that's

where the ping gateway is in the is in

the visited Network so the key concepts

of the lp architectures are the

following one is that all radio related

functionality is pushed out the base

station and that's all there in the he

node be for those people who are who are

knowledgeable about the 3G architecture

we had another entity which is called

the RNC which was separated from the e

node B and we had two nodes but now we

are down to just one node and all the

functionality is done in the e note B in

the control plane there is control plane

and user plane separation the Mme is the

control plane entity which does all the

transaction related work and then the S

gateway P gateway our user plane

entities and they carried the users

traffic there is a little bit of a

control plane signaling between the

escapee gateway but for most practical

purposes these two nodes are essentially

handling just a user plane traffic one

interesting thing is that LTE is a

packaged switch domain only and and

there is a funny video that you can see

on YouTube which talks about LTE not

supporting voice it's about Hitler kind

of being made aware that LT is not

supporting voice and he gets all furious

so it's got got some time to a Google

look at it and it's a hilarious video so

there was there was set at the time when

LT had come out there were the people

who can affect them out with this sign

essentially creating something with

something like this voice is not

supported and and which is not

correct in the sense that voice is

supported as voiceover LTE but what's

called the CF domain that is the wave

voices today carried in 3gpp networks

it's include G networks is not supported

and it's what what is going to be

supported as voiceover LTE and that's

all voltage