[Ndn-interest] NDN architecture in wireless sensor networks?

Muhammad Hosain Abdollahi Sabet M.AbdollahiSabet at mail.sbu.ac.ir
Tue Mar 3 19:18:27 PST 2015


Hi Jeff,

Exactly! Intermediate nodes with higher capacity in storage and energy would run a forwarder with cach, and as you pointed out, whole system (overall) whould enjoy its benefits. When I  said "why bother" I meant for tiny weak nodes in leaves which are connected only to one of those powerdul ones.

Thanks,
Sabet


Hi Sabet,

> What about the caching? In some star-like topologies most(if not all) of the sensors are not intermediate, and don't need content store, thus there is no need to check it, right? So why bother?

When you say "no need to check it" I assume you mean "no need to check if the cache already has a data packet for the interest." In your topoplogy, there would be a more powerful node that can run a forwarder with a cache. This should actually help the low-power sensors since they only have to answer the interest once since the returned data packet is cached in a forwarder. Future interests for the same sensor data would be answered from the cache in the forwarder, saving the low-power sensor from doing it.  (I hope I understood your question correctly.)

- Jeff T

From: Muhammad Hosain Abdollahi Sabet <M.AbdollahiSabet at Mail.sbu.ac.ir<mailto:M.AbdollahiSabet at Mail.sbu.ac.ir>>
Date: Tuesday, March 3, 2015 at 11:44
To: Anil Jangam <anilj.mailing at gmail.com<mailto:anilj.mailing at gmail.com>>, Jeff Thompson <jefft0 at remap.ucla.edu<mailto:jefft0 at remap.ucla.edu>>
Cc: "ndn-interest at lists.cs.ucla.edu<mailto:ndn-interest at lists.cs.ucla.edu>" <ndn-interest at lists.cs.ucla.edu<mailto:ndn-interest at lists.cs.ucla.edu>>
Subject: RE: [Ndn-interest] NDN architecture in wireless sensor networks?


Jeff,
Thanks for the reply. Actually we haven't experienced anything yet, not that I know of. I'm just the ndn-guy of the team, who is encouraging his colleagues to consider NDN for implementing their WSN scenario(I guess the reason why I'm here, is clear now:). My teammate will show up soon. She probably knows about the micro-controller specs. But in my humble assumption, it should be a kind of Arduino-based ones.
The sensors which are used in Anil's suggested paper are ATxmega128A1 microcontrollers.

What about the caching? In some star-like topologies most(if not all) of the sensors are not intermediate, and don't need content store, thus there is no need to check it, right? So why bother?

Anil,
Thank you so much. I will take a full look at that. Maybe we should consider some CCN-NDN differences then.

Thanks,
Sabet

-----Original Message-----
From: Anil Jangam [mailto:anilj.mailing at gmail.com]
Sent: Tue 3/3/2015 8:03 PM
To: Thompson, Jeff
Cc: ndn-interest at lists.cs.ucla.edu<mailto:ndn-interest at lists.cs.ucla.edu>; Muhammad Hosain Abdollahi Sabet
Subject: Re: [Ndn-interest] NDN architecture in wireless sensor networks?

Sabet,

You may want to take a look at this optimized ccn stack for WSN. I don't
know however what footprint it works on.

http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6529776&url=http%3A%2F%2Fieeexplore.ieee.org%2Fxpls%2Fabs_all.jsp%3Farnumber%3D6529776

/anil.
On Mar 3, 2015 8:17 AM, "Thompson, Jeff" <jefft0 at remap.ucla.edu<mailto:jefft0 at remap.ucla.edu>> wrote:

>  Hello Sabet,
>
>  A sensor can certainly receive an NDN interest and return a data packet
> using a client library. If the sensor's microcontroller is too small to
> sign the data packet with public key cryptography, it can use symmetric
> authentication, for example as described in section IV - D of "Securing
> Building Management Systems Using Named Data Networking".
> http://named-data.net/publications/securing_building_management_ndn/
>
>  Is it enough for your sensors to receive an interest and respond with a
> data packet?  May I ask what is the memory and program storage size of your
> target microcontroller? (We have some experience with Arduino-based
> microcontrollers.)
>
>  Thanks,
> - Jeff T
>
>   From: Muhammad Hosain Abdollahi Sabet <M.AbdollahiSabet at mail.sbu.ac.ir<mailto:M.AbdollahiSabet at mail.sbu.ac.ir>>
> Date: Tuesday, March 3, 2015 at 7:33
> To: "Ndn-interest at lists.cs.ucla.edu<mailto:Ndn-interest at lists.cs.ucla.edu>" <Ndn-interest at lists.cs.ucla.edu<mailto:Ndn-interest at lists.cs.ucla.edu>>
> Subject: [Ndn-interest] NDN architecture in wireless sensor networks?
>
>   Considering the fact that caching is an architectural attribute in NDN
> and again considering NDN trust model features, like cryptographic policies
> in command interest(public key-private key), is it ok for NDN to work in
> WSN scenarios? The need for designing a forwarding strategy seems clear.
> But having in mind that we have serious challenges for energy management in
> wireless sensors, couldn't we just ignore CS and PIT table for some
> applications, and have a simpler NFD?
> I've already seen the paper about securing the campus BMS on the website.
> It has said:
>
> "Our initial deployment collects sensing data from existing
> industry-standard sensors and
> gateways that employ legacy BMS protocols. Consequently, our design
> includes gateway devices that read data (through an internal IP network)
> from sensor devices running legacy
> BMS protocols and publish the resulting data in an NDN network."
>
> Well, I don't mean like that. I want to read data from sensors which
> running NDN protocols.
>
> I will be glad to read your ideas.
>
> Thanks
> Sabet
>
> _______________________________________________
> Ndn-interest mailing list
> Ndn-interest at lists.cs.ucla.edu<mailto:Ndn-interest at lists.cs.ucla.edu>
> http://www.lists.cs.ucla.edu/mailman/listinfo/ndn-interest
>
>


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.lists.cs.ucla.edu/pipermail/ndn-interest/attachments/20150304/3b54c268/attachment.html>


More information about the Ndn-interest mailing list