[Lwip] draft-ietf-lwig-cellular-01

[Hannes Tschofenig <[e-mail]>]

Hi all,

I briefly read through this document.

Here is my summary of what I learned from the different sections.

* Section 2.  Goals for Low-Power Operation

This section high level summary about energy consumption and types of
power sources. It summarizes some of the statements from the LWIG
terminology draft.

* Section 3.  Link-Layer Assumptions

I learned that cellular networks are application agnostic (at least
agnostic to these types of IoT applications), that the common link model
is the point-to-point link and multicast discovery does not work
(whatever one would want to discover locally) and that listening as well
as sending on a radio requires energy.

* Section 4.  Scenarios

Two scenarios are being investigated,namely the sleepy device and the
real-time reachable device.

* Section 5.  Discovery and Registration

This section talks about the normal network attachment procedure and
then focuses on the CoAP protocol interactions (resource directory and
the not yet standardized mirror proxy). Since there is nothing in the
local cellular network to discover the document describes alternative
ways to discover these CoAP entities.

* Section 6.  Data Formats

This section highlights that there are different data formats and the
authors recommend the use of [I-D.jennings-senml].

* Section 7.  Real-Time Reachable Devices

This section discusses the use of CoAP servers to accomplish these
reachable devices. Needless to say that servers consume more energy.

* Section 8.  Sleepy Devices

This section speculates about the use of the CoAP mirror servers and
attempts to give recommendations for implementers.

* Section 9 Security Considerations

The description focuses on the delegation of queries in context of the
mirror server discovery.

My summary:

From the title of the document I was a bit surprised not to learn
anything about cellular radio technologies. I think the filename and the
title of the document is misleading.

I think the goals of the document have to be re-evaluated. There is a
lot of text in there but not much that I haven't read in other documents
already.

Ciao
Hannes