How to simulate the delay increasing in LTE networks

[Rafael Fernandes Lopes]

Hello,

I am doing some simulations in LTE networks using some specific protocols. However, one of our evaluations is the throughput and goodput measurements as a function of the channel delay... The LTE module supports the change of channel delay in any way? There is some way to simulate this?

Best regards,

Rafael

[Tommaso Pecorella]

Hi,

the channel delay depends on the nodes mutual position, and the models already include this.

However, i can ensure you that the channel delay variation is negligible. You could have a variation on the packet's delay due to queues, scheduling, and retransmissions, but the "pure" channel delay is negligible.

Have fun,

T.

[Rafael Fernandes Lopes]

Thanks for the response, Tommaso!

I will consider it... In which component do you think we should simulate the queue delay? PGW? (We really want to avoid to change the scheduling parameters).

And how can we simulate this in this scenario?

Regards,

Rafael.

--
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/FrUurqwmM2E/unsubscribe.
To unsubscribe from this group and all its topics, send an email to ns-3-users+...@googlegroups.com.
To post to this group, send email to ns-3-...@googlegroups.com.
Visit this group at http://groups.google.com/group/ns-3-users.
For more options, visit https://groups.google.com/d/optout.

[Tommaso Pecorella]

Hi Rafael,

no idea tbh, it depends on your system. Moreover, queues are usually due to (temporary) congestion events, so in any network the delay shouldn't just increase. Unless, of course, you have an evil ISP that is voluntarily delaying your packets just because it can. They call it traffic shaping... but it's just a way to say "we know shit about how our network works".

To simulate this, I'd simply add a node adding a fake delay to all the packets. A sort of DelayBox, it shouldn't be hard to build one from scratch.

Cheers,

T.

[Rafael Fernandes Lopes]

I see it! The idea of this simulation is just to show how the protocol responds (in terms of the throughput) to the presence of some sort of delay, although we know that the delay is not constant (as well as it is not monotonically increasing! ;)).

I will try some approaches to introduce this artificial delay (maybe including it in the p2p channel between eNB and SGW/PGW).

Thank you very much for your attention, Tommaso!

Best regards,

Rafael.