[manet] Re: manet Digest, Vol 257, Issue 11

Hi Jim, I could go either way on the MANET router definition (simplest per -03 or move involved per -06) because the L2/L3 layering is discussed elsewhere in the document. But, I am curious to what extent your comments may be oriented toward intellectual property assertions of your company and/or yourself?

Thank you - Fred

From: Stevens, Jim <[email protected]>
Sent: Friday, August 14, 2026 10:29 AM
To: [email protected]
Subject: [EXTERNAL] [manet] Re: manet Digest, Vol 257, Issue 11

EXT email: be mindful of links/attachments.


Hi AB

Yes, I prefer the definition of MANET router mentioned in the draft -06  Section 2. Terminology over that of draft -03.

This is because people often make assumptions about MANET routers based upon layer 3 and ignore issues that arise from layer 2 where the MANET router is highly intertwined with the channel access protocol.  And yes, people have extended DLEP to try to solve these problems, but I have still seen layer 3 DLEP networking designs that are inefficient in operation because of being divorced from layer 2 channel access or else they recreate layer 2 functions in layer 3.

As an example, look at the email thread about MANET definition of a neighbor and the discussion that followed with many responders suggesting to just update routing tables faster to determine good neighbors
https://mailarchive.ietf.org/arch/msg/manet/P3f60tam6W9aCf-g5xiU7zyHw3o/

I have successfully designed and implemented layer 2 and layer 3 routers with over ten different fielded layer 2 or layer 3 MANET networking waveforms and found subtle differences in the design of the routing protocols:

  *   Layer 2 routers have to worry more how to determine if a node is a neighbor or not and what type of neighbor:  good neighbor, bad neighbor, interfering neighbor, not a neighbor, etc; and also can make better physical and channel access (layer 1 & 2) control decisions
  *   Layer 3 routers can often assume a node is either a neighbor or not a neighbor (easier to do with point to point RF links or above a Layer 2 MANET versus broadcast RF links without a layer 2 MANET); and also worry about interoperating with IP networking

Note that I use the term “waveform” as defined in DODI 4630.09, https://www.esd.whs.mil/Portals/54/Documents/DD/issuances/dodi/463009p.pdf page 22<https://www.esd.whs.mil/Portals/54/Documents/DD/issuances/dodi/463009p.pdf%20page%2022>:
“An electromagnetic signal-in-space, typically defined by Open Systems Interconnection model layers 1 through 3, along with the controls and processes for a desired function or application. These processes do not include the message content.”

Best regards,
Jim



Date: Fri, 14 Aug 2026 12:36:37 +0200

From: Abdussalam Baryun <[email protected]<mailto:[email protected]>>

Subject: [manet] Re: I support distinction draft manet inet gap

               analysi makes between L2/L3 MANETs

To: "Stevens, Jim" <[email protected]<mailto:[email protected]>>



Hi Jim,



do you agree with the definition of MANET router mentioned in the draft. My

comment also mentioned that the definition of 'MANET router' should make

distinction, because there are different routers in different layers.



I am still waiting for the draft author reply to my concern.



Best Regards

AB

_______________________________________________
manet mailing list -- [email protected]
To unsubscribe send an email to [email protected]