[SensorNetArch] Fragmentation & Reassembly

[Philip Levis]

On Sep 21, 2004, at 11:10 AM, Cheng Tien Ee wrote:

> I'm thinking about layering, especially the SP layer, which seems 
> necessary if we assume that the network may be heterogenous and 
> therefore can have different MAC layers. I think fragmentation 
> definitely implies that some sort of temporary storage is required. A 
> question would be: if the application passes a 100byte payload, that 
> is fragmented into 50byte-sized fragments at SP, and further split 
> into 20+byte ones at the MAC layer, does this mean that temporary 
> storage of fragments need to be present at each layer? Or perhaps we 
> should just leave the fragmentation to the application (it can choose 
> to use flash for storage for instance), bound the maximum packet size 
> across layers to be some number based on the worst case MAC layer, 
> then, as Joe mentioned, use packet aggregation if another MAC layer 
> can transmit at higher rates and therefore more able to support a 
> larger packet size?

I think it's unlikely that transparent fragmentation/defragmentation 
(strict layering) will be an effective mechanism. I also think we need 
to be careful to not conflate the idea of an architecture with that of 
a uniform API; I would assume that SP provides datagrams of a size 
corresponding to the underlying layer 2, as opposed to an a single (and 
inherently inefficient) constant size.

I suspect that discussing network protocols is going to start to become 
difficult if we constrain ourselves to the OSI model. We need to 
effectively distinguish the lowest level datagram format from media 
sensing from communication scheduling from SP. A strawman:

6: SP (e.g., AM, dispatch)
5: Scheduling Layer (e.g., S-MAC)
4: Datagram Layer (e.g, 802.15.4 format)
3: Carrier Layer (e.g. ChannelMonM, some of CC1000RadioIntM)
2: Encoding Layer (e.g., SecDed)
1: Physical Layer

The ordering isn't strict; although 4 requires some existence of 3, 
they really sit along side one another, as they can be totally 
independent.

Phil

-------

"We shall not cease from exploration
And the end of all our exploring
Will be to arrive where we started
And know the place for the first time."

- T. S. Eliot,  'Little Gidding'