Last Call Review of draft-ietf-paws-protocol-12
review-ietf-paws-protocol-12-secdir-lc-meadows-2014-07-10-00

I am resending this, because I got one of the email addresses wrong.

Cathy

Catherine Meadows

Naval Research Laboratory

Code 5543

4555 Overlook Ave., S.W.

Washington DC, 20375

phone: 202-767-3490

fax: 202-404-7942

email:

catherine.meadows at nrl.navy.mil

On Jul 8, 2014, at 12:10 PM, Catherine Meadows <

catherine.meadows at nrl.navy.mil

> wrote:

I have reviewed this document as part of the security directorate's

ongoing effort to review all IETF documents being processed by the

IESG.  These comments were written primarily for the benefit of the

security area directors.  Document editors and WG chairs should treat

these comments just like any other last call comments.

This ID describes a protocol, PAWS,  that allows wireless devices to access
currently unused portions of the radio spectrum.

The protocol works between a geospatial database and a device with geolocation
capabilities.  The device reports its location

and other relevant information to the database, which in turns gives it
information about which portions of the spectrum is available to it.

This removes the responsibility for managing the complex information about
spectrum available from the device and to the database,

which is better equipped to handle it.

The ID has a very thorough and well-written Security Considerations section,
which  covers the security threats against such a protocol.  They identify two

main threats

 By using the PAWS protocol, the Master Device and the Database expose

themselves to the following risks:

o Accuracy: The Master Device receives incorrect spectrum availability

information.

o Privacy: An unauthorized entity intercepts identifying data for

the Master Device or its Slave Devices, such as serial number and

location.

Note that core PAWS does not address client authentication, on the grounds that
unauthorized clients could find out the existence of white

space on their own without the help of PAWS, and in that case there would be
nothing preventing them from using it. The ID does point out though that client
authentication may be required by specific regulatory domains,

and so it is possible for the Database to require client authentication, e.g.
by TLS.  The authors appropriately point out the limitations of using TLS for
authentication, particularly

when the keys are trusted to small ubiquitous devices.

I believe this draft is ready.

Catherine Meadows

Naval Research Laboratory

Code 5543

4555 Overlook Ave., S.W.

Washington DC, 20375

phone: 202-767-3490

fax: 202-404-7942

email:

catherine.meadows at nrl.navy.mil