stds-802-16: std-802-16: SYSREQ: Followup Comments on 802.16s0-99/1

[Scott Marin <smarin@boschtelecominc.com>]

[Notice: It is the policy of 802.16 to treat messages posted here as non-confidential.]

Brian,

Thanks for the opportunity, at the Montreal meeting, to discuss comments
on the subject document. 

The following are elaborations on some comments in which you asked for
specific text and additional comments that we did not have time to get
to during the meeting.

2.1 regarding the sentence about "fixed (also portable), not mobile."
How
does this sound? "802.16 systems are in the fixed radio-communication
service. 802.16 equipment may be re-located to new station locations but
the stations are in a fixed location when in operation. 802.16 systems
are not in the mobile service."

2.1 definition of system. (see Figure in attachment sent separately to
Brian and Roger due to size limitations of reflector) The 802.16 system,
for the
purposes of the air-interface standards (802.16.1), consists of an array
of cells operating over a block of frequencies. Cells contain hub
stations, which may employ sectored antennas, and subscribers stations.
Interference from transmitters within the array of cells to receivers
within the array of cells is self-interference. Interference from
outside the array, either geographically or spectrally, is external
interference and is dealt with in the coexistence project (802.16.2).

Note, the use of the terms hub station and subscriber station is
consistent with FCC Part 101.3. The longer I think about using terms
like BTS and STS, the more I'm inclined to suggest using Hub and Sub.
Can we get some discussion going? The T in BTS and STS implies a
transceiver which in not always the case. 

The definition of system for co-existence should match the definition
for interoperability. The coexistence PAR implies that interference
within a system is the responsibility of the systems design. Unless the
definition of system in the interoperability PAR includes multiple cells
rather than a single radio link, there will be a functionality gap
between interoperability and coexistence. Allowing such a functionality
gap in our definitions can become a fatal flaw in our project even
though it will take a while to realize the mistake.

Note the proposed definition does not try to deal with the issues of
network interfaces or subscriber interfaces, but only with definitions
that influence the air interface. Self-interference definitely
influences the air interface, but is readily managed and mitigated
within the proposed system definition.

2.1 replace the sentence containing "out-side plant" with "802.16
systems are intended use in the out-door portion of an access network.
Other systems such as 802.11 and 802.15 are intended for wireless
in-door communications."

Para 5.2, delete last sentence. The comments about cost are negative and
potentially misleading to those observing the 802.16 effort. Cost
statements probably don't belong in the systems spec. If we must discuss
cost, I suggest a carefully worded section that deals only cost topics
at a very high level. Costs will come down as deployments occur,
competition increases, and standards emerge. As the market grows, costs
will come down which will open up more markets.

5.3 last sentence. too negative. delete sentence or change wording "low
degree of flexibility". The paragraph seems to be discussing the
relative
rate of provisioning service. Provisioning rates need to ranging from
the slow and
infrequent "nailed-up" service to the fast and per virtual circuit
service.

5.4 rename title to Link Availability. Total availability depends on
link availability, equipment availability, and other unavailability
factors. The link
availability requirements vary from about 0.999 to 0.99999 depending on
the type of service and local market requirements. Specific links within
a system may be optimized for specific link availability (high or low).

    Note that multiple links into a given subscriber site may be using
to increase the effective link availability.

5.4 3rd paragraph. After to sentence ending in "rain fall" and "or other
causes of excess attenuation". Add additional sentence, "Note that
automatic transmit power control (ATPC) algorithms, which normally
adjust power based some form of S/(N+I), must detect excessive
interference
and not increase power substantially due to interference." Note that the
above requirement is extremely important when negotiating sharing
criteria with other services and for assuring system stability.

5.5 after "(BER)" add "requirement"

5.5 2nd paragraph, last sentence. after "bandwidth" add "or less
delay."

5.5 3rd paragraph, replace "relaxed" with "different"

5.6 delay categories. none of the definitions appear to be the steady-
state delay that payload information receives during the main body of
the message; therefore, add "transport delay, media access delay
excluding set-up time."

6 2nd paragraph, after "For instance, the MAC layer" add "in connection
oriented mode" 

6.1 after last paragraph, add "Some classes of service, e.g. SDH/PDH,
require distribution of a network clock."

after 6.3 add new major paragraph "6.4 Stability Considerations"

The 802.16 system must remain stable, controllable, an observable in the
presence of:

1) steady-state and transient interference
2) restart from outages such as a BTS failure.
3) sudden loss of signal (airplane flies through beam)
4) failed component (e.g. all BTSs should not escalate to max transmit
power if a BTS receiver fails).
5) frequency drift of transmitters.

7.2 after current sentence, add "An STS should not transmit unless
authorized by the BTS." and add "The transmit power of the STS is
controlled by the BTS and the BTS should prevent the STS from
transmitting at maximum power during clear sky conditions." add "The STS
can elevate transmit power to overcome attenuation in excess of clear
sky path loss only under the direction of the BTS."

7.3 Accounting and Auditing

after last sentence, add "The 802.16 systems support the Communications
Assistance for Law Enforcement Act (CAECO, "wire tap law") in the
following ways: [tbd]".

7.4 Transmit power

On any authorized frequency, the average power delivered to an antenna
must be the minimum amount of power necessary to carry out the
communications desired. (FCC Part 101.113)

8. 3rd paragraph,
after "employed" add "as an option for authentication, authorization, or
privacy, independently"

Note, some markets may warrant a "give it away or let'em steal it"
approach to increase market demand and to sell other services such as
advertising. In such cases, no security is necessary and the logistics
cost for
managing keys or other crypto information is unnecessary. The minimum
requirement should be no security. On the other hand, security for
authentication and authorization will probably be necessary for
deployment of many systems.

8.3 Privacy.

The minimum requirement should be no security. As security of data
transmission is normally a function of higher level protocols, its
duplicating effort, adding cost, and increasing latency to require
security of the air-way information.

Again, you work as editor has been outstanding. Thanks for considering
these suggestions.

Regards,
Scott Marin