LENA: issues with multiple NetDevices per eNB

[Hans Garry]

Hello,

I am trying to create a setting where I have two LTE netdevices per eNB, and where I have a set of UEs attached to one of the devices, and another set of UEs assigned to the other. I observe the following (I am building on top of using lena-dual-stripe example):

1. When I try to activate EPC (epc=true), and when I declare an epCHelper, I get the following error: 

msg="InternetStackHelper::Install (): Aggregating an InternetStack to a node with an existing Ipv4 object", file=../src/internet/helper/internet-stack-helper.cc, line=394

terminate called without an active exception. 

Could you please help me understand what I am missing? If I do not enable EPC, I get no errors. Shouldn't an eNB be allowed to have many devices (e.g. when there are multiple cell-IDs per eNB)? 

2. If I do not enable EPC and simply plot the REM, I observe that if I set both the eNB devices to the same EARFCN, the SINR increases. Is this normal? I believe that this should not be happening. 

Thanks a lot in advance for your help. 

G.

[Nicola Baldo]

On Wednesday, October 3, 2012 7:36:39 PM UTC+2, Garry wrote:

Hello,

I am trying to create a setting where I have two LTE netdevices per eNB, and where I have a set of UEs attached to one of the devices, and another set of UEs assigned to the other. I observe the following (I am building on top of using lena-dual-stripe example):

1. When I try to activate EPC (epc=true), and when I declare an epCHelper, I get the following error: 

msg="InternetStackHelper::Install (): Aggregating an InternetStack to a node with an existing Ipv4 object", file=../src/internet/helper/internet-stack-helper.cc, line=394

terminate called without an active exception. 

it's simply that LteHelper was not meant to be used for having multiple LteEnbNetDevices in the same node. You'll have to write your custom helper to do that.
 

Could you please help me understand what I am missing? If I do not enable EPC, I get no errors. Shouldn't an eNB be allowed to have many devices (e.g. when there are multiple cell-IDs per eNB)? 

the lena-dual-stripe example already features multi-sector eNB sites. This is achieved by having multiple eNB nodes at the same location. 

 

2. If I do not enable EPC and simply plot the REM, I observe that if I set both the eNB devices to the same EARFCN, the SINR increases. Is this normal? I believe that this should not be happening. 

I agree it should not happen. However, the EARCNs works properly in the lena-dual-stripe example. Are you sure it's not due to your particular sim program?

Regards,

Nicola

[Hans Garry]

Thanks a lot for the prompt response, Nicola. 

1. If you could provide some high-level directions on how to create my custon lteHelper to support 2 LteEnbNetDevices in the same eNB, I would really appreciate it, and also other researchers would find it useful.  

2. Regarding the 3-sector eNB, I agree that having 3 separate eNBs is the case in dual-stripe, but it is not the typical case in practice. Typically there are different cells per eNB, with one cell per sector. Currently, a UE would have to perform an inter-eNB handover in order to switch cell. It would be very useful to have support for such functionality in future ns3 releases. 

3. Regarding the REM, you are not observing the increase in SINR in dual-stripe, because there is only one LteEnbNetDevice per eNB. I am plotting the REM for the case where I have 2 such devices per eNB, and I set the same earfcn to both of them. 

Cheers and thanks a lot, 

G.

---

Date: Thu, 4 Oct 2012 03:08:56 -0700
From: nba...@cttc.es
To: ns-3-...@googlegroups.com
Subject: Re: LENA: issues with multiple NetDevices per eNB

--
You received this message because you are subscribed to the Google Groups "ns-3-users" group.
To view this discussion on the web visit https://groups.google.com/d/msg/ns-3-users/-/6OvHA-pAobYJ.
To post to this group, send email to ns-3-...@googlegroups.com.
To unsubscribe from this group, send email to ns-3-users+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/ns-3-users?hl=en.

[Hans Garry]

Hi Nicola, 

I re-wrote my setup, and instead of defining 2 devices per eNB, I simply place a 2nd eNB in the same location and set it to a different EARFCN. However, the REM problem still shows up: I observe that if I set all eNBs to the same EARFCN, the SINR is increased (!). 

G.

---

Date: Thu, 4 Oct 2012 03:08:56 -0700
From: nba...@cttc.es
To: ns-3-...@googlegroups.com
Subject: Re: LENA: issues with multiple NetDevices per eNB

[Nicola Baldo]

Hi Garry,

On Thursday, October 4, 2012 3:31:20 PM UTC+2, Garry wrote:

1. If you could provide some high-level directions on how to create my custon lteHelper to support 2 LteEnbNetDevices in the same eNB, I would really appreciate it, and also other researchers would find it useful.  

well, LteHelper might work more or less as-is, but EpcHelper will definitely need significant rework; to start with, you'll probably want to use the same S1-U NetDevice to talk to the SGW. But that's only the beginning. I am sorry but I don't think I can provide you with more information, it's just a use case I never considered. You have no choice but go through the code and get an idea of how to do it. At least, that's what I would do if I had to do it.

 

2. Regarding the 3-sector eNB, I agree that having 3 separate eNBs is the case in dual-stripe, but it is not the typical case in practice. Typically there are different cells per eNB, with one cell per sector. Currently, a UE would have to perform an inter-eNB handover in order to switch cell. It would be very useful to have support for such functionality in future ns3 releases. 

I am not sure what you are referring to. Could you please provide any reference for how a UE would switch from one sector to the other withouth invoking the handover procedure?
 

3. Regarding the REM, you are not observing the increase in SINR in dual-stripe, because there is only one LteEnbNetDevice per eNB. I am plotting the REM for the case where I have 2 such devices per eNB, and I set the same earfcn to both of them. 

On Wednesday, October 10, 2012 1:01:27 AM UTC+2, Garry wrote:

I re-wrote my setup, and instead of defining 2 devices per eNB, I simply place a 2nd eNB in the same location and set it to a different EARFCN. However, the REM problem still shows up: I observe that if I set all eNBs to the same EARFCN, the SINR is increased (!). 

I just filed bug 1515 to track this issue. Let's continue the discussion there.

Regards,

Nicola

[Hans Garry]

Nicola, 

I have reopened bug 1515 for some time now. Could you please have a look again? 

Thanks a lot,

Garry

---

Date: Wed, 17 Oct 2012 09:39:12 -0700

From: nba...@cttc.es
To: ns-3-...@googlegroups.com
Subject: Re: LENA: issues with multiple NetDevices per eNB

Hi Garry,

I just filed bug 1515 to track this issue. Let's continue the discussion there.

Regards,

Nicola

--
You received this message because you are subscribed to the Google Groups "ns-3-users" group.

To view this discussion on the web visit https://groups.google.com/d/msg/ns-3-users/-/XgXXW1U4o5gJ.

[何梦]

today，the same problem came to me, And at last I find that one node can only install the stack once time. My network structure is some simple, so I choose to install the node with stack one by one.