RE: [manet] Wi Max?

["Vitor Silva (Ext)" <[e-mail]>]

Hi all,

Exactly... I don't see where is the doubt. WiFi, Wimax, IrDA, ... are the L1/L2 technology that "glues" the nodes together. MANET is the *network*, with all its layers and planes. It's the same as discussing what is the relation between Ethernet and Internet. Ethernet is, or can be, an element of Internet. There's here a "include" relation -- and nothing more.

Actually, there's no reason, in theory, why you cannot build a MANET out of wired ethernet. From my point of view, it's just a practical issue: it would be silly to do it.

Finally, and probably it's where the confusion comes from, most current MANET approaches are trimmed to work with wireless RF and somehow (but easily understood...) strongly associated to 802.11. For that reason, all its proposed mechanisms are somewhat thought having 802.11 in mind, although it should be as independent as possible of the underlaying MAC/PHY technologies. But, then again, that's "only" a practical issue: 802.11 is, at least for now, the top enabler for MANETs.


--Vítor JS



-----Original Message-----
From: Carlos H. Rentel [mailto:[e-mail]] 
Sent: quinta-feira, 28 de Outubro de 2004 13:51
To: Rex Buddenberg
Cc: [e-mail]
Subject: RE: [manet] Wi Max?


Hello,

In my humble opinion though to say MANET is a layer 3 issue is mainly over
simplistic. IEEE 802.11 ("WiFi") or 802.16 (ver. a adapts better to the concept
of MANET, although more accurately speaking is a mesh concept) can not be the
ultimate (particullarly MAC) approach. MANETs is an *all layer* issue (and I am
not necessarily referring to cross-layer design :) ).

Best Regards,

Carlos Rentel



Quoting Rex Buddenberg <[e-mail]>:

> Parag,
>
> Giorgio is right.  MANET is a layer 3 issue and both WiFi and WiMax are
> layer 1-2 technologies.
>
> That said ...  Most experimental MANET stuff needs some kind of plumbing
> underneath and WiFi (IEEE 802.11) has been popular for that.  And a lot
> of the experimentation has been done in 'ad hoc' or 'infrastructureless'
> mode where two subscriber nodes can connect with each other.  In short,
> WiFi can be pretty promiscuous which happens to fit the MANET
> intentions.  And it's d.c. -- dirt cheap.
>
> Wi-MAX (IEEE 802.16) changes the MAC algorithm from a carrier sense one
> used in WiFi (and it's antecedent, wired ethernet) to a scheduling
> algorithm.  Scheduling MACs rely only on the assumption that everyone
> can hear the base station (BS); there's no requirement whatever that
> peers (SSs) hear each other so the hidden node problem goes away (so in
> v1, SS-to-SS is not well supported).  Many satellite comms algorithms
> (e.g. DAMA) fall in this general category, but are anemic by comparison
> to .16.  Scheduling MACs can also provide stability under overload /
> oversubscription, bandwidth efficiency, and ability to control QoS to
> the extent that we can provide deterministic service at layer 2 ... all
> of which are important in many situations.  ... but don't have much to
> do with MANET.
> 	What 802.16 does not do well yet is the Wi-Fi ad hoc mode.  We may have
> to modify this statement soon as real compliant products start appearing
> in the marketplace (the 802.16-2004 standard was ratified in June and
> that seems to have uncorked a lot of development).  There are two
> developments in IEEE 802 that are germane here:
> 	1) there is 'mesh' work going on within 802.16 but it's acknowledged as
> not being mature in the -2004 version.
> 	2) bit of bureaucratic dispute between 802.16, 802.20 and 802.22 about
> the terms 'fixed' and 'mobile' in the PARs (charters).  All three
> committees (.20 and .22 are much less mature) have both MAC and PHY
> development in their charter docs.  But it's not at all clear why either
> .20 or .22 need to develop a new MAC -- a little growth and maturity in
> the .16 MAC may be all that's needed.... but the 802.16 PAR has the term
> 'fixed' in it.  These two committees do indeed need to develop new PHY
> standards for their particular purposes (cellphone --> packet switch and
> TV spectrum reuse for rural, respectively).
>
> Wi-MAX is going to be interesting in the next few years.  At least some
> in the industry are planning to put Wi-Max chipsets in your laptop like
> Wi-Fi ones are now.  The shift from infrastructureless to a more
> hierarchical layer 2 structure does change some of the MANET tacit
> assumptions.
>
> Help?
>
>
> On Wed, 2004-10-27 at 02:36, Giorgio Mulas wrote:
> > Dear Parag,
> > WiFi is a Layer 1 and 2 technology and it is not related with routing
> > issues.
> > Ad hoc networking is a research topic considering routing issues (and,
> > of course, other important things)
> >
> > The main relation between WiFi and Ad hoc networks is that cheap
> > WiFi-enabled devices are usually used to build up testbeds for
> > MANETs...
> >
> > BR.
> >
> > Giorgio
> >
> >
> >
> >
> >         -----Original Message-----
> >         From: [e-mail] [mailto:[e-mail]]
> >         On Behalf Of Parag Goswami
> >         Sent: mercoledì 27 ottobre 2004 10.45
> >         To: [e-mail]
> >         Subject: [manet] Wi Max?
> >
> >
> >         Dear All,
> >         If anybody tells me the major differences between Ad hoc
> >         networking and WiFi.
> >         As we know both are used for the communication between mobile
> >         nodes.Or is it the case that WiFi is also a part of Ad Hoc
> >         networking ?
> >
> >         Best Regards,
> >         Parag Goswami
> >         http://www.geocities.com/paragboom2k
> >
> >
> >         ______________________________________________________________
> >          Do you Yahoo!?
> >         Yahoo! Mail Address AutoComplete - You start. We finish.
> >
> >
> > ______________________________________________________________________
> > _______________________________________________
> > manet mailing list
> > [e-mail]
> > https://www1.ietf.org/mailman/listinfo/manet
> --
>
> b
>
>
>
> _______________________________________________
> manet mailing list
> [e-mail]
> https://www1.ietf.org/mailman/listinfo/manet
>




_______________________________________________
manet mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/manet

_______________________________________________
manet mailing list
[e-mail]
https://www1.ietf.org/mailman/listinfo/manet