RE: [802.21] HLSI
Hi, some comments below.
Stefano
________________________________
From: ext Yoshihiro Ohba [mailto:yohba@TARI.TOSHIBA.COM]
Sent: Tue 9/6/2005 9:23 AM
To: STDS-802-21@listserv.ieee.org
Subject: Re: [802.21] HLSI
On Tue, Sep 06, 2005 at 06:34:50AM -0700, Gupta, Vivek G wrote:
>
>
> One of the main purpose behind these IEs is to aid in handover decision
> making, i.e to help the policy engine
>
> in selecting the appropriate network and initiating handovers.
>
> Information about availability of high level services can influence this
> decision making, however
>
> a lot of these elements within HLSI as currently defined are
> questionable and it's not clear if we really need them.
I agree with you that the current HLSI is questionable.
>
>
>
> Also in terms of IEs we need to separate the ones which are media
> independent from the ones that are media dependent.
>
> The IEs that are media independent currently include:
>
> { Network_Operator, List of Networks Supported, Location,
> Roaming_list_Availabale, Neighbor_reports_Available }
List of Networks Supported, Roaming_list_Availabale,
Neighbor_reports_Available is questionable as well. Network_Operator
could be improved to contain an IANA-assigned, global unique SMI
enterprise number as well as the name of the operator. Location could
be improved to carry RFC 3825 Location Configuration Information (just
having latitude and longitude is insufficient to represent a
geo-coordinate).
>
>
>
> The media dependent IEs currently include:
>
> { Cipher_Suites, Authentication_Methods, Cost (free/not free),
> IP_Version, Data_Rates, QoS, Neighbor_Maps }
In my analysis, Cipher_Suites, Authentication_Methods, Data_Rates, QoS
are questionable. I agree on representing Cost as a binary value
(free or not free).
[Stefano] Authentication_Methods is needed. How does the UE know which methods are supported and allowed? The UE will have to try and fail to find out. Even if an AP implements WPA and WPA2, it does not mean the network enables/supports all authentication mechanisms.
>
> We could use a basic TLV format to represent these, along with a simple
> mechanism to query or set the values,
>
> of different IEs, so not sure why we really need a basic schema and an
> extended schema and all the other baggage along with it.
I am currently defining TLV format for IEs defined in basic set. I
think the IEs defined in basic set should be able to be queried using
not only XML but also TLV. On the other hand, I do believe schemas
(basic schema and extended schema) are needed to support extensibility
to deal with any link-layer technologies including the existing ones
and future ones as well as to make flexible and efficient information
queries. After the July meeting, I did qualitative analysis on the
applicability of XML/RDF to 802.21 information service and the
analysis result will be described in my another contribution prepared
for the September meeting, which is a white paper about 802.21
information service using XML/RDF technologies.
I really want to settle on all of those basic issues on information
service in the September meeting.
Best regards,
Yoshihiro Ohba
>
>
>
> Best Regards,
>
> -Vivek
>
>
>
> -----Original Message-----
> From: stds-802-21@ieee.org [mailto:stds-802-21@ieee.org] On Behalf Of
> Yoshihiro Ohba
> Sent: Monday, September 05, 2005 8:59 PM
> To: stds-802-21@listserv.ieee.org
> Subject: HLSI
>
>
>
> I am currently writing up a contribution to revise Information Service
>
> sections and I have the following questions:
>
>
>
> The HLSI IE defines several flags indicating the available higher-layer
>
> services including ISP, MMS, IMS, MIP, VPN, SIP and NAT.
>
>
>
> - What is the exact meaning of "VPN support"? Does it mean that if
>
> you connect to the PoA then all data traffic will be automatically
>
> forwarded to some remote network over a dedicated tunnel between the
>
> PoA and the remote network? Or does it mean that the network provides
>
> a VPN gateway? Or does it mean that the mobile terminal connected to
>
> the PoA can establish a VPN connection to any VPN gateway. Or
>
> something else? The first definition does not make sense because you
>
> will need additional information about the remote network to make a
>
> handover decision. The latter two definitions do not make sense
>
> either, because there are several different ways of establishing a VPN
>
> connection (i.e., IPsec, SSL, L2TP, PPTP, etc.) and you will need
>
> additional information as to which VPN method is used to make a
>
> handover decision.
>
>
>
> - What is the exact meaning of "SIP support"? Does it mean that the
>
> network has a SIP server or proxy, or something else?
>
>
>
> - Do we really need HLSI IE defined in the basic set? I think it can
>
> be defined in extended set. This is because we might need more
>
> detailed information about higher-layer (such as IP addresses and
>
> prefixes of access routers, supported IP mobility optimization
>
> mechanism, list of supported ISPs, etc.) to make a higher layer
>
> information and just defining a set of flags seems like a half-baked
>
> solution. Such detailed information can be provided via schema-based
>
> query by which various higher-layer (and lower-layer) MIB objects can
>
> be retrieved once converted to RDF data.
>
>
>
> Regards,
>
>
>
> Yoshihiro Ohba
>