Unrealistic LTE Handover Delay (How to calculate)

[iliya Alavy]

Hello NS-3 community,

   I am simulating a LTE scenario in which Ue moves between several eNB nodes with constant speed of 40kmph. The distance between the Ue and eNB nodes are not the same (different rsrp/rsrq for different handovers) in different handovers. I try to calculate the handover delay by logging the Simulator::Now() value at the start and the end of the handover process. 

   Surprisingly, the calculated handover delay is constant in all of the handovers and has an unrealistic value of 4.2ms (unreal!). Also handover preparation time (Time needed to send the request handover packet and receive the ack) is 0ms which is quite unlikely. 

Am I using a wrong method to calculate the handover delay? 

[Zoraze Ali]

Hi iliya,

I am bit unclear how you determine the start of the handover process but i think the delay can be calculated through the traces mentioned in LTE user documentation,

https://www.nsnam.org/docs/release/3.25/models/html/lte-user.html#x2-based-handover

Additionally, could you please tell me which value you consider as realistic for this delay?  It will also depend on the configuration of a handover algorithm and the RRC implementation(real or ideal) you are using.

P.S. it will be helpful if you could attach your script.

BR,
Zoraze

[iliya Alavy]

Thanks for the quick response Zoraze,

I have also tried the trace method described in the documentation you linked but results are the same. 

About the realistic delay, I gotta say, I am no expert in NS-3 or LTE networks but logically considering that my Ue has a speed of 40kmph and is in different distances from eNBs while doing different handoffs, 4ms constant handover delay seems quite unlikely. I expected different handover delays and something in the range of 20-100ms (end to end handover including the preparation time which is not included in the documentation's trace method and is surprisingly 0ms for me). 

I'm using a2a4 handover algorithm with threshold of 30 and I have tried both ideal and real RRC implementation and got the same result.

[iliya Alavy]

Ok guys, I found the problem. 

The rrc configuration was not at the top of my code. I changed it's position in code and set it to real and everything is now fine. Other configurations actually take effect now. 

[Zoraze Ali]

Hi iliya,

Thanks for updating and good job for figuring it out on your own.

BR,
Zoraze

[Konstantinos]

Hi

Please see the location of the code that this error comes from

With basic programming understanding you can clearly see that this line is a comment.

Regards,

K

On Wednesday, December 7, 2016 at 1:01:18 PM UTC, zhangl...@gmail.com wrote:

hello，

       I want to test your lte_handover.cpp code，so I copy your code to ns3/scratch and save it lte_handover.cc,then i ran this .cc file,but come to a error,i don't know how to solve it.this result show:

  ../scratch/lte-handover.cc:366:3: error: ‘connect’ was not declared in this scope

   connect custom trace sinks for RRC connection establishment and handover notification

   ^

could you tell me this question?

在 2016年10月18日星期二 UTC+8下午8:21:15，iliya Alavy写道：

[iliya Alavy]

I believe I have added some custom traces to NS source codes and that's causing this error. Try the code by removing traces. In case you needed further clearance, I'll be happy to assist.

--
Posting to this group should follow these guidelines https://www.nsnam.org/wiki/Ns-3-users-guidelines-for-posting
---
You received this message because you are subscribed to a topic in the Google Groups "ns-3-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/ns-3-users/NuU0R9LrChc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ns-3-users+unsubscribe@googlegroups.com.
To post to this group, send email to ns-3-...@googlegroups.com.
Visit this group at https://groups.google.com/group/ns-3-users.
For more options, visit https://groups.google.com/d/optout.