headers
Ruffino Simone | 1 Nov 2004 10:29

New draft on connectivity scenarios and autoconfiguration issues

Hi to all,

a new draft was submitted as individual submission, titled
"Autoconfiguration in a MANET: connectivity scenarios and technical
issues". 

Goal of the draft is to describe the scenarios, in which a MANET can be
connected to an external network (by means of either wired of wireless
links) and to list some of the technical issues, which are related to
autoconfiguration and global connectivity, that arise in such scenarios.

Comments welcome and appreciated !

Thanks in advance,
Simone Ruffino

> A New Internet-Draft is available from the on-line Internet-Drafts
directories.
>
>
> Title: Autoconfiguration in a MANET: connectivity scenarios and 
>   technical issues
> Author(s): S. Ruffino, et al.
> Filename: draft-ruffino-manet-autoconf-scenarios-00.txt
> Pages: 26
> Date: 2004-10-19
>
> MANET interconnection with external networks enables a number of
>    usage scenarios, but generates also a number of technical issues,
>    mainly related with node autoconfiguration and global connectivity.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Ali Hamidian | 1 Nov 2004 12:02
Picon
Picon

Re: Another questions about MAC protocols

What Herb says is correct of course, but AODV will retransmit the 
broadcasted RREQ message if no RREP is received. After some number of 
retries AODV will give up. See the AODV RFC for more details.

Hope it helps
Ali

Herbert Rubens wrote:

>This is no ACK or retransmission of broadcast packets, only unicast
>packets. So the RREQ will have no retransmissions by the lower layer but
>the RREP will since it is unicast back to the source.
>
>Hope this helps.
>
>-Herb
>
>
>
>-----Original Message-----
>From: manet-bounces <at> ietf.org [mailto:manet-bounces <at> ietf.org] On Behalf
>Of Yau P
>Sent: Saturday, October 30, 2004 10:20 AM
>To: manet <at> ietf.org
>Subject: [manet] Another questions about MAC protocols
>
>Dear all,
>
>I was wondering if you help me answer a simple question.  Assuming that
>802.11 or some other similar MAC protocol is being used, and the ad hoc
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bart Braem | 1 Nov 2004 11:47
Picon

Re: AODV: previous hop for RREP

> Well basicaly, it means that when you receive a RREP, you just forward it
> as it would be expected: sending it in the direction of the node which
> originated the RREQ. You have a route to that node, because that route
> has been reated when you received the RREQ (in the old AODV drafts
> terminology: the reverse route (the route that goes from the destination
> till the originator that is created when RREQ is being forwarded).
>
> But you don't send it to the originator node directly. You just
> forward it to the next hop of the route to the originator.
> And that next hop will do the same. And so on, till it
> reaches the node that originated the RREQ.

Okay I understand the principle. So that previous hop mentioned in the first 
paragraph is the next node which will send the RREP to the originator of the 
RREQ.
But what happens when no valid route to the previous hop is found? It says "If 
needed, a route is created for the previous hop, but without a valid sequence 
number." I can create a new route table entry but then the RREP is stuck with 
me... A little bit further: "...  the node consults its route table entry    
for the originating node to determine the next hop for the RREP packet, and 
then forwards the RREP towards the originator using the information in that 
route table entry." So a route to the next hop of presumed?

Thanks for your help
Bart
Permalink | Reply | 0 comments |
headers
McLanahan.Stevens | 1 Nov 2004 16:54

RE: New draft on connectivity scenarios and autoconfigura tion issues

First of all some comments (nothing personal of course). 

Overall I think this will help to serve as an architectural starting point, a 10,000 foot view if you will. It
does need to be cleaned up from an editorial perspective (grammar and spelling errors mostly). Since the
group probably does not want to read my spelling nitpicks, I will leave those in a private follow up email.
Besides, I'm an engineer, not an English professor.

Technical Discussion
--------------------

Section 2. Node: Do we want to tie ourselves to IPv4/IPv6 as a node definition. Perhaps IP-Based would be
best here. After all, we never know if someone will come up with some sort of IPvWireless in the distant future.

Section 3.2.1 Surveillance networks: I do not think the generalization that nodes will remain stationary
is a good one. Surveillance (or preferably called Sensor networks), can have a dynamic nature to them
depending on the application. Take, for example, a gas monitoring van (often used by gas companies) with
several sensors on it. The van is in constant motion throughout the network. Perhaps a 3rd case should be
added here. In addition a "Heavyweight" node scenario should probably be added where each node can act as a
gateway/router if necessary.

Section 4.1 I think a scenario needs to be addressed in this section that discusses the transient/isolated
MANET issue. Take for example a MANET that periodically comes into contact with a gateway. That MANET will
transition between an isolated MANET and a connected MANET. The potential for this happening in a mobile
environment is high.

Section 4.1.3 Volatile network scenario: What happens when the MANET is volatile? Let's take this case
into consideration... 2 MANETs join with say x nodes that have the same address. The MANETs separate into
two different networks with one MANET getting several of the duplicate addresses. Capability will have
to exist in each MANET "sub-network" to broker network addresses.
(Continue reading)

Permalink | Reply | 0 comments |
headers
Pankaj Godbole | 1 Nov 2004 18:11
Picon

Re: AODV kernel module problem

Thanks Luke,

The module is working now. The reason for the error was that I was
passing the wrong interface to $use_dev in start.sh.

Regards,
Pankaj.

On Mon, 1 Nov 2004 10:02:07 -0500, Luke Klein-Berndt <kleinb <at> nist.gov> wrote:
> Hi Pankaj,
>         I think what might be happening is that additional error
> messages are being generated by Kernel AODV. I would check the kernel
> log to see if there are any messages there. I would also make sure the
> interface you specify in the start.sh is properly configured and has an
> IP address.
> 
> Try: "insmod kernel_aodv.o use_dev=wlan0(or the interface you wish to
> use)"
> 
> Also if you would like to use modprobe, make sure you do "make install"
> to place the module into the correct directory.
> 
> Hope this helps, send me an email if it doesn't.
> 
> -Luke Klein-Berndt
> 
> 
> 
> > -----Original Message-----
> > From: manet-bounces <at> ietf.org [mailto:manet-bounces <at> ietf.org]
(Continue reading)

Permalink | Reply | 0 comments |
headers
Kais Mnif | 1 Nov 2004 21:37
Picon
Favicon

how to build a benchmark for experimentation for ad hoc network

Dear patriots

In our educational laboratory, we like to build an ad hoc network for 
experimentation
we just starting, we like to know if someone can help us to choose hardware 
(product such as switchs routers wireless cards etc. ) and software ! we 
anticipate to use Linux as an operating system.

Any suggestion and help will be appreciated
Permalink | Reply | 1 comment |
headers
Christopher P. Rigano | 1 Nov 2004 23:30
Picon

RE: Are there protocols that are like SIP or H.323 that couldprovide session or peer-peer sceesions in MANETS for Voice or Multimedia?

Greetings Daniel,
In your MANET architecture, do all the nodes act as Location Services,
Proxies, User Agent Clients (UACs) and User Agent Server (UASs)?

Thanks and regards,

Chris

-----Original Message-----
From: manet-bounces <at> ietf.org [mailto:manet-bounces <at> ietf.org] On Behalf
Of Daniel Henkel
Sent: Saturday, October 30, 2004 11:13 AM
To: manet <at> ietf.org
Subject: Re: [manet] Are there protocols that are like SIP or H.323 that
couldprovide session or peer-peer sceesions in MANETS for Voice or
Multimedia?

Hi Christopher,

We have successfully used SIP to initiate VoIP sessions over several
hops in our testbed. Using Linphone on both laptops and iPAQs we were
able to have very reliable communication with our DSR implementation.
You can possibly use a much more stripped down version of SIP and
achieve same results if footprint is an issue; try oSIP:
http://www.gnu.org/software/osip/osip.html

Regards,
Daniel Henkel
University of Colorado at Boulder
http://AUGNet.colorado.edu
(Continue reading)

Permalink | Reply | 0 comments |
headers
Joo-Han Song | 2 Nov 2004 06:49

layer of manet routing?

Hi all,

Is there anyone who can clarify the layer of manet routing in real
implememtation?
In many real papers, they are implemented as an application daemon.
Does that mean the manet routing header should be included in the
application layer header?

Another question is about routing header of AODV. In AODV's both RREQ or
RREP messages, I can find a destination address field. Since the IP header
has the same field, I am wondering why we need to keep the same destination
address field in both IP and AODV header.

Regards,

JH 
_______________________________________________
manet mailing list
manet <at> ietf.org
https://www1.ietf.org/mailman/listinfo/manet
Permalink | Reply | 0 comments |
headers
srgopal | 2 Nov 2004 07:20
Picon
Favicon

RE: Are there protocols that are like SIP or H.323 that couldprovide session or peer-peer sceesions in MANETS for Voice or Multimedia?

Dear Daniel and christopher,

Are you saying that this SIP protocol can be used to
generate VoIP packets like packet voice in Mobile  Ad Hoc
networks to see the performance of Ad Hoc networks when
supporting VoIp and/or multimedia? I mean can it be used
with ns-2 simulator and see its affect on ad hoc network
performance. Please clarify this. If not, how can one
generate packet voice while simulating Ad Hoc network using
ns-2 simulator?

I would appreciate your time and patience in this regard.

Regards,
Raja.

> Greetings Daniel,
> In your MANET architecture, do all the nodes act as
> Location Services, Proxies, User Agent Clients (UACs) and
> User Agent Server (UASs)?
>
> Thanks and regards,
>
> Chris
>
> -----Original Message-----
> From: manet-bounces <at> ietf.org
> [mailto:manet-bounces <at> ietf.org] On Behalf Of Daniel Henkel
> Sent: Saturday, October 30, 2004 11:13 AM
> To: manet <at> ietf.org
> Subject: Re: [manet] Are there protocols that are like SIP
> or H.323 that couldprovide session or peer-peer sceesions
> in MANETS for Voice or Multimedia?
>
> Hi Christopher,
>
> We have successfully used SIP to initiate VoIP sessions
> over several hops in our testbed. Using Linphone on both
> laptops and iPAQs we were able to have very reliable
> communication with our DSR implementation. You can
> possibly use a much more stripped down version of SIP and
> achieve same results if footprint is an issue; try oSIP:
> http://www.gnu.org/software/osip/osip.html
>
> Regards,
> Daniel Henkel
> University of Colorado at Boulder
> http://AUGNet.colorado.edu
>
>
> ---- Original Message ----
> From: Christopher P. Rigano
> To: <manet <at> ietf.org>
> Subject: [manet] Are there protocols that are like SIP or
> H.323 that could provide session or peer-peer sceesions in
> MANETS for Voice or Multimedia?
>
> Friday, October 29, 2004, 1:59:34 PM, you wrote:
>
> CPR>  
>
>
>
>
>
>
>
>
>
> _______________________________________________
> manet mailing list
> manet <at> ietf.org
> https://www1.ietf.org/mailman/listinfo/manet
>
>
> _______________________________________________
> manet mailing list
> manet <at> ietf.org
> https://www1.ietf.org/mailman/listinfo/manet
>
>
Permalink | Reply | 501 comments |
headers
Li Li | 2 Nov 2004 13:58
Picon

RE: Are there protocols that are like SIP or H.323 thatcouldprovide session or peer-peer sceesions in MANETS for Voiceor Multimedia?

We did the SIP over MANET simulation using OPNET, generating SIP voice
packets over an OLSR MANET to evaluate service performance. It should be
possible with ns-2 simulator. 

You can find our work and paper at
http://www.crc.ca/en/html/manetsensor/home/research_area/support

li li

At 12:20 AM 11/2/2004 -0600, srgopal wrote:
>Dear Daniel and christopher,
>
>Are you saying that this SIP protocol can be used to
>generate VoIP packets like packet voice in Mobile  Ad Hoc
>networks to see the performance of Ad Hoc networks when
>supporting VoIp and/or multimedia? I mean can it be used
>with ns-2 simulator and see its affect on ad hoc network
>performance. Please clarify this. If not, how can one
>generate packet voice while simulating Ad Hoc network using
>ns-2 simulator?
>
>I would appreciate your time and patience in this regard.
>
>Regards,
>Raja.
>
>
>
>> Greetings Daniel,
>> In your MANET architecture, do all the nodes act as
>> Location Services, Proxies, User Agent Clients (UACs) and
>> User Agent Server (UASs)?
>>
>> Thanks and regards,
>>
>> Chris
>>
>> -----Original Message-----
>> From: manet-bounces <at> ietf.org
>> [mailto:manet-bounces <at> ietf.org] On Behalf Of Daniel Henkel
>> Sent: Saturday, October 30, 2004 11:13 AM
>> To: manet <at> ietf.org
>> Subject: Re: [manet] Are there protocols that are like SIP
>> or H.323 that couldprovide session or peer-peer sceesions
>> in MANETS for Voice or Multimedia?
>>
>> Hi Christopher,
>>
>> We have successfully used SIP to initiate VoIP sessions
>> over several hops in our testbed. Using Linphone on both
>> laptops and iPAQs we were able to have very reliable
>> communication with our DSR implementation. You can
>> possibly use a much more stripped down version of SIP and
>> achieve same results if footprint is an issue; try oSIP:
>> http://www.gnu.org/software/osip/osip.html
>>
>> Regards,
>> Daniel Henkel
>> University of Colorado at Boulder
>> http://AUGNet.colorado.edu
>>
>>
>> ---- Original Message ----
>> From: Christopher P. Rigano
>> To: <manet <at> ietf.org>
>> Subject: [manet] Are there protocols that are like SIP or
>> H.323 that could provide session or peer-peer sceesions in
>> MANETS for Voice or Multimedia?
>>
>> Friday, October 29, 2004, 1:59:34 PM, you wrote:
>>
>> CPR>  
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> manet mailing list
>> manet <at> ietf.org
>> https://www1.ietf.org/mailman/listinfo/manet
>>
>>
>> _______________________________________________
>> manet mailing list
>> manet <at> ietf.org
>> https://www1.ietf.org/mailman/listinfo/manet
>>
>>
>
>_______________________________________________
>manet mailing list
>manet <at> ietf.org
>https://www1.ietf.org/mailman/listinfo/manet
> 
Li Li, Ph.D
Communications Research Centre, Industry Canada
3701 Carling avenue 
P.O. box 11490 
Station H, Ottawa, 
Ontario, Canada K2H 8S2 
Tel: (613) 990-5246 
Email: li.li <at> crc.ca
Permalink | Reply | 501 comments |

Gmane