Tuesday, November 24, 2009

LTE: APN, PCO & Initial Attach

This post is out of an interesting conversation I had with one of my readers.

In LTE, UE gets attached to a network in a single shot. That is when the UE is switched on it will send a NAS message, Attach Request along with PDN connectivity request. Look here.

3GPP TS 24.301 has a little different story to tell though. According to the spec UE shall not include APN and PCO in the PDN connectivity request when the same is sent along with attach request. The spec says UE shall send the PDN connectivity request with a flag "ESM Information transfer" on and no APN or PCO shall be included. Once MME receives the Attach Request+PDN connectivity request, it shall go ahead and accept the attach but it shall not establish the EPS bearers just yet.

MME now goes ahead with establishing security context. Look here. After the security context is established MME will send a NAS message "ESM Information Request" asking UE for APN and PCO. Now UE shall send "ESM Information Response" with APN and PCO, encrypted(?). Once MME receives this response it will go ahed with establishing the EPS bearers. If the response doesn't include APN then default APN shall be used by MME.

Why this? I guess this is for security. We just dont want to reveal the user name passwords to any network that asks for it, right?. And also above is only valid if Attach Request and PDN connectivity request messages are sent together. The story is different if the two messages are sent separately. More thoughts?

Saturday, November 21, 2009

LTE S1 Handover: Indirect tunnel

I wrote about LTE S1 handover in the white paper. However I dint cover this interesting concept. I am working on next revision of white paper and will make it public soon.

X2 based handover is used when there is X2 link between source and target eNB's. This also means that EPC should be just informed about change in eNB FTEID's for downlink. If we look deep, the downlink packets are still sent to source eNB until EPC receives a handover complete notification and the target eNB FTEID's. The buffered downlink packets are sent to target eNB by source eNB later which are then sent to UE. These packets are directly sent over the X2 link. This means there is direct tunnel formed between source and target eNB for data forwarding.

Now I started looking at a S1 based handover. I saw something called Indirect Tunnel.


Now in S1 handover we assume that there is no X2 link between source and target eNB. Also I am assuming that SGW has changed during the handover. So the downlink packets buffered at the source eNB during handover execution should be sent all the way to Source SGW, then to target SGW and then to target eNB. To do this target MME sends a GTP based message Create Indirect Data Forwarding Tunnel Request to target SGW. In this message the target eNB FTEID for downlink are sent. In the same way the message is sent from source MME to source SGW. This will create a indirect tunnel between source and target eNB. After this the buffered packets are sent from source eNB to target eNB which are later sent to UE.

Sounds good? Anything to add? Please feel free.

Thursday, November 19, 2009

An observation!

There are bunch of office boys at the place I work. Since one year I have noticed them using big fat mobile phones while I carry a small GSM phone which can do nothing but voice and sms. So I talk to the boys and they tell me that these phones are made in China and apparently are very cheap. They also told me that these phones might stop working anytime or they might just keep working for years.

Later I observed them using GPRS and accessing WAP sites of service providers. They download a ringtone or a photo of an actress everyday. Now this made me wonder. People who have never used internet on a computer, or rather who have never used computer, have started using internet on mobile phone. So what happens when these guys realize the beauty of youtube. If 3G comes in and service providers start advertising youtube service on television, these guys are going to have a blast. Remember its the adds on television that made facebook so popular in India (Aircell advertisement?). This means India is directly moving to mobile internet, which is jumping over a step, skipping the internet revolution on PC. India is adding 10 million subscribers every month. Now with 3G coming and youtube like services rolling out just imagine how congested the networks are going to get!

Any thoughts?

Sunday, November 15, 2009

IMS is the way to go?

I have been strong supporter of VoLGA, but I have recently realized that industry should move towards one long term solution. I believe IMS is the way to go in future.

3GPP clarifies the LTE Myths - Strong support for IMS :-)

Friday, November 13, 2009

LTE Dedicated Bearers: The Big Question!

I wrote about LTE dedicated bearers some time back(here , here and here ). Just when I thought I got every thing right I realized that I am back to square 1.

We know that dedicated bearers are network initiated. But the big question is when does the network decide to initiate the dedicated bearer. Many people have asked me about this so I thought of putting down what I knew.

In LTE there is a bearer which is always established until UE is shut down and it is called default bearer. Well UE can ask for a dedicated bearer by sending out a bearer allocation request to the network. Once network receives the UE request then we will have dedicated bearer. This is one case.

What we noticed is there is a trigger for dedicated bearer. Now if we look at the spec for dedicated bearer creation, it immediately starts with Create Bearer Request. Many are confused with what could be the trigger for this message. Next para should help.

Second case is:- default bearer is established. Network now wants to have all the http traffic put in separate dedicated bearer. So once the default bearer is established, network might ask UE to run the http traffic on dedicated bearer. Note here that there is no trigger for dedicated bearer, which means network is configured in such a way that all http shall run on dedicated bearers with a particular qos. Even if UE is running http of default bearer it may be asked to switch to dedicated bearer. If UE is not capable enough to run a http then it may reject the dedicated bearer request also it may reject the request if it thinks default bearer is fine for http.

Now the confusion here is IMS calls. Many say how will network initiate a dedicated bearer for IMS calls. The big confusion is with reception of IMS call. Well, this is what I think. IMS needs an application to be running UE, a special app. If UE is intelligent enough then it will request a dedicated bearer for IMS call while it is registering with network or when the IMS app on UE starts running. Else PGW may look at the packets that are being sent by UE (deep packet inspection?) and initiate a dedicated bearer by it self. The point here I want to make is we shouldnt think that network will create a dedicated bearer only when UE receives an IMS call, the bearer is created much ahead of it.

If we consider that network is modifying or creating a new bearer for already running traffic, then what about the UE application's connection state. Will the connection be reset as the bearer has to be switched. I dont know, an UE engineer should answer this. How ever this dedicated bearer is still pretty confusing, in fact many are confused with what could be triggers for a dedicated bearer creation. Any ideas?

Sunday, November 8, 2009

Evaluating LTE SGW

We know that Serving Gateway is one of the most important entities in LTE network. Say if XYZ service provider is evaluating a SGW, what are the factors that XYZ would be looking at. Lets avoid price and technical support offered by the equipment vendor here.  

Serving Gateway is supposed to handle both control plane and user plane traffic. Just to increase the complexity, lets consider that S5 interface is PMIP based. So we have GTPv2 communication on S11 interface, GTP-U encapsulated traffic on S1-U interface, PMIP based control plane and GRE encapsulated user plane traffic on S5 interface. So the technical requirements are revolving around control and user plane.

Control Plane: Input to the SGW are GTPv2 messages and output are PMIP message towards PGW. In between SGW needs to communicate with PCRF/Radius servers for pulling other information. So first thing XYZ wants to see is a successful session established. That means we need to get the basic functionality correct. More over XYZ would also want to see how easy it is to configure the SGW. This may not be one of the criteria for evaluating the box but its good have a easy to use interface. Getting the basic functionality right means there needs to be a successful interop. That is the SGW should work with other vendors MME or eNB or PGW or etc. This also means that SGW vendor should get the specs right :-).

Next, XYZ wants to see how many sessions can be activated per second. Also XYZ would be very interested to see how many active sessions can the SGW handle. May be a million users per box? That means SGW should have million contexts activated and data base integrity should be smooth. XYZ might also want to see how many dedicated bearers can be activated for single default bearer. More over XYZ might fancy to see how many default/dedicated bearers can be activated and deactivated per second. This test might prove how strong is the box is.

Handovers is another functionality which XYZ wants to see. Does SGW support various handovers mentioned in the spec. If so XYZ would prefer a number of users that can have successful handover per second.

Roaming too is important. There is typical requirement which arise with roaming scenarios. MME of visiting network might want to contact home network SGW for authentication, charging policy etc. Which means the GTP message is hopping for one router to other in the internet. Which also means that there needs to be some security involved. IPSec comes into picture. MME can encrypt the GTPv2 control messages in IPSec and pass it on to the home network SGW. This means SGW should also be to handle several IPSec sessions. I dont want to go below this layer. SGW might also be able to do BGP/OSPF/MPLS etc at the network layer. These could be some of the control plane requirements.

User Plane: First thing that comes into my mind is throughput. XYZ wants to see how may user plane session can be handled at what throughput. Usually the packet size is fixed in S1_U interface, so there is need to know the maximum throughput that box can achieve. I strongly believe that this is the most critical factors.

Finally, XYZ wants combination tests, thousands of sessions established along with thousands of user plane tunnels with different kinds of traffic. These are system level tests. If everything matches XYZ's requirement then they are ready to spend heck loads of money on the gateway.

These were few thoughts running through my mind. Feel free to add more!