[ndnSIM] Mobile package

[Adhy Satya]

Hi,


I am looking on the TODOs from the code and have a few questions.



1. In the first TODO in the `doSend() ` -here
<https://github.com/4th-ndn-hackathon/ndnSIM-Mobile-Simulation-Package/blob/master/model/ndn-v2v-net-device-transport.cpp#L92>-
 where we need to remove the previous tag if it exists, does
`lpPacket.get<::ndn::lp::GeoTagField>();` work like the PeekPacketTag()?
i.e., does it search for a matching tag and calls deserialize()? I don't
understand why line 95 is commented out.



2. What is the reason of setting isLocal to true if geo tag field exists?
(Line #98)


3. On Line #112, what the reason for checking the HopCount after computing
the WaitingTime? There's no reference to hopCount in the Rapid Traffic
Dissemination... paper.


4. How do I associate the v2v-net-device-transport to my link type? Or how
do I hardcode my simulation to use the v2v-net-device?



I hope I can finish this code and contribute to the project :)



Best,

Adhy



On Tue, Aug 29, 2017 at 6:11 PM, Spyridon (Spyros) Mastorakis <
mastorakis at cs.ucla.edu> wrote:

> Hi,
>
> the strategy is explained (more or less) in this paper:
>
> https://lasr.cs.ucla.edu/afanasyev/data/files/Wang/nom.pdf
>
> To finish the v2v net device transport implementation, you can take a look
> at the .c file. There is a number of TODOs there. I think that if you
> finish those TODOs and do some debugging, you should be good to go.
>
> Thanks,
>
> Spyridon (Spyros) Mastorakis
> Personal Website: http://cs.ucla.edu/~mastorakis/
> Internet Research Laboratory
> Computer Science Department
> UCLA
>
> On Aug 28, 2017, at 6:30 AM, Adhy Satya <adhysatya820 at gmail.com> wrote:
>
> Hi Spyros,
>
> Thank you for your answer.
> If I understood correctly, the broadcast strategy uses the exponential
> Interest suppression, which is the only mechanism preventing flooding
> (since the v2v-transport is not being use).
> What is needed to be able to use the v2v-net-device-transport? Is it a
> feature implemented in the new version of NFD that is not implemented in
> the 0.4 version?
>
>
>
> On Fri, Aug 25, 2017 at 6:45 PM, Spyridon (Spyros) Mastorakis <
> mastorakis at cs.ucla.edu> wrote:
>
>> Hi,
>>
>> if I recall correctly, we did not get to use the v2v-net-device
>> transport. We directly modified the net device transport instead (
>> https://github.com/4th-ndn-hackathon/ndnSIM-Mobile-Simulati
>> on-Package/commit/d6bce17032111bdf8c170b9c694dced7f0796b14).
>>
>> From the perspective of long term design planning, a Face of NFD will be
>> associated with a link of specific type. This link can be an ad-hoc or a
>> point-to-point link, etc, so this could determine how a strategy makes its
>> decisions. I do not think that an application will be able to know the type
>> of the underlying link (unless it sends a query to NFD somehow — this could
>> be implemented as a hack in ndnSIM if needed).
>>
>> We are planning to integrate the latest version of NFD with ndnSIM this
>> fall to be able to use the link type feature in simulations, so please stay
>> tuned.
>>
>> Thank you,
>>
>> Spyridon (Spyros) Mastorakis
>> Personal Website: http://cs.ucla.edu/~mastorakis/
>> Internet Research Laboratory
>> Computer Science Department
>> UCLA
>>
>> On Aug 24, 2017, at 9:29 PM, Adhy Satya <adhysatya820 at gmail.com> wrote:
>>
>> Hi Spyros,
>>
>> I downloaded and tested the mobile simulation package code. It's very
>> helpful to me, thanks
>> My question is how does the application/strategy knows which
>> net-device-face to use, since there are the "original"
>> net-device-transport" and the v2v-net-device-transport.
>>
>> Thanks
>>
>>
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndnsim/attachments/20171009/b8b207dc/attachment-0001.html>