Wonk,
>> To my knowledge, none are "down" on the technology. Up to
now, the reasons hypothesized for not stepping up to the plate<<
Agreed, I don't know of an instance where public statements came
from any of these players downing the technology, per se, except
perhaps where specific terrain conditions played into the equation
for a particular form of wireless technology on a case specific
basis like LMDS or MMDS. Then again, I once knew of a company
known as CAI and a couple of BOCs who... but I still don't
think that one was a fair deal all around. Seem more on politics
and folk lore later, below.
Although, many of them, including the RBOCS/LECs, have
managed to have mis-fired several times in making it work. I
happen to think that one of the unspoken barriers to the successful
trial-ing and deployment of wireless data services to the residence
and to the business lies in the cultural preoccupation with what the
telcos themselves are regarding as "their" next gen contribution to
the history of telecomm. That is, the Full Service Area Network or
FSAN which is predicated on ATM over DSL. Above the ATM
Layer, anything goes.
And yes, it's largely founded on financial arguments, ROIs, etc., and architectural rationalizations, but don't ever kid yourself: There is no shortage of folk lore in this equation of the future, and this folk lore, too, adds enormously as a factor in determining the final directions these folks will take.
I'm not saying that there is anything intrinsically better or worse
about ATM in the ground than it, or any other Layer 2 or 3 protocol,
over Wireless, or vice versa. But what I am saying is that perhaps
the FUD factor in some subliminal way is behind the slow
acceptance of the ideas which would foster greater acceptance of
over-the-air interfaces. In other words, "preordained directions will prevail, in any event, so why bother?" is how I think many BOCsters feel. And who more than the BOCs could add impetus to the wireless regimen, in such a way that all others would be forced to follow?
I first become better acquainted with the FSAN principles (it
wasn't called FSAN at the time, but that's what it was) when
Alcatel won the RBOCs' RFP for ADSLs almost two years ago. I
studied the Alcatel framework, which was largely a make over
of the Bellcore model, as I recall, and lately I have come to
understand it a lot better, as their version of the FSAN
Initiative. For more on this, you can go to the Telephony Mag
site...
internettelephony.com
...which should bring you into the "In Focus" section where you
will find the FSAN article in the July 6th slot.
I'll post some excerpts of the main article here for a flavor of what
I'm referring to, because it will probably scroll at the home site
in a week's time. If you go there and read the entire article, be sure
to go to the other links referenced therein, some of which are from
the Telephony archives.
I'll get to your other points later, or tomorrow night. Hope you all enjoy the article. Gotta get some sleep now!
Regards, Frank C.
==================
From the July 6th 1998 Telephony Magazine, In Focus Section:
internettelephony.com
"The FSAN Initiative" By RANDY SHARPE
For most of this century, telephone companies have deployed access networks
primarily for the delivery of plain old telephone service. Now, on the eve of the new
millennium, there is explosive growth in the demand for Internet access and other
high-speed data services--and increasing competitive pressure.
In response to this environment, leading network operators and equipment suppliers
from around the world are working together to define a global specification for
access systems supporting a full range of narrowband and broadband services. This
activity is has been dubbed the FSAN-Gx initiative.
The FSAN initiative began in July of 1995 to expedite the introduction of full-service
networks. The leaders are12 of the world's largest telephone companies. These
companies combined serve more than 300 million access lines and deliver a host of
telephony services to end users worldwide.
The FSAN Initiative
Participating network operators
Supporting equipment
manufacturers
Bell Canada
Alcatel
BellSouth
Ascom
British Telecom
BroadBand Technologies
CNET-France Telecom
Ericsson
CSELT-Telecom Italia
Fujitsu
Deutsche Telekom
Italtel
GTE
NEC
NTT
Siemens
SBC
Bosch
Swisscom
Lucent
Telefonica
Nortel
Telstra
SAT
FSAN Organization
FSAN-Gx is not a standards body. Rather, it is a group of network operators
dedicated to creating a common set of requirement specifications for full-service
access systems.
Although the network operators have differing regulatory and demographic
environments, and different business objectives, the organization believes there is
sufficient commonality that common technology, components and network elements
can be used to help network operators increase volume while simultaneously
decreasing costs. FSAN-Gx builds on and contributes to existing and developing
standards and specifications from bodies such as the ANSI, ETSI, ITU, IETF,
DAVIC, ATM Forum and ADSL Forum (see Additional Resources on the Net).
Phases
FSAN-Gx has progressed in three phases. The first phase, from the inception of the
initiative in July 1995 until June 1996, identified technical and economic barriers to
broadband access network introduction and defined a common FSAN architecture
and key components that could support a wide range of access network
architectures.
The second phase, from July 1996 to February 1997, concentrated on devising a
common set of requirement specifications. The third and current phase focuses on
completing the specifications and driving toward field trials. To accomplish this last
phase, two chapters were set up:
Fiber-to-the-Home, which addresses fiber to the premises (home, building or
business).
Fiber-to-the-Cabinet, focusing on networks involving a final copper xDSL link
to the customer.
The FSAN Architecture
The challenge of the FSAN-Gx is to create a service-independent architecture that
provides demanding business customers with voice and data connectivity that
combines high capacity and Quality of Service (QoS); residential customers with
affordable voice, data and video services with a minimum of network construction;
and future-proof, high-capacity, low operations cost, all-dielectric residential access
networks.
The solution requires service independence, flexibility and scalability. The two key
elements are:
A service-independent access network providing Layer 2 connectivity
between service specific service nodes and network terminating devices.
An ATM passive optical network (PON).
The FSAN architecture is based on an arrangement in which an access network
provides Layer 2 connectivity between home or business terminating devices and
one or more service nodes. In FSAN, the Layer 2 connection is based on ATM
virtual paths (VPs).
FSAN-Gx selected ATM multiplexing because it supports various service classes
such as constant bit rate (CBR), variable bit rate (VBR) and available bit rate (ABR)
with a rich set of QoS parameters. In addition, ATM muliplexing provides intelligent
queuing and scheduling mechanisms to ensure the negotiated QoS.
Standards have been defined for the transmission of principle FSAN services such as
IP, DS-1/E-1 and MPEG 2. The Universal ADSL Working Group (UAWG) has
also selected ATM as the underlying technology for U-ADSL.
Service nodes and customer premise equipment (CPE) are service-specific while the
access network is service-independent. The access network does not interpret user
signaling and has no knowledge of the ongoing services.
New services can be introduced across the access network by the introduction or
enhancement of a service node and the attachment of compatible CPE such as
personal computers, set-top boxes, PBXs and IP routers.
The network termination can be as simple as a cell-forwarding device for network
demarcation, a VP multiplexer supporting multiple ATM CPE devices, or integrated
with adaptation or interworking functions such as conversion to IP on Ethernet or
Universal Serial Bus (USB). The ATM VPs across the access network contain one
or more ATM virtual channels (VC).
What gets carried on the VCs is determined by the attachments to the access
network, the SN and the CPE. To more efficiently use the resources of the access
network, the access network may include an ATM VC multiplexing capability. The
VC multiplexer in the access network is controlled by the service node. Using this
mechanism, VCs can get established on-demand within pre-established VPs.
Additional capabilities have been defined in support of switched digital broadcast
services such as digital television. When the user requests a channel change a
message is sent transparently across the access network to the SDV service node.
If the user is permitted to view the channel, a control message is sent to the access
network from the service node requesting the access network to forward the
selected channel to the set-top box. If another set-top box on the same access
network is viewing the same channel then the access network can replicate the
channel for the new viewer.
An ATM PON
After careful investigation of component technologies, trends and their uses, the
FSAN participants determined that an ATM passive optical network (PON) was the
best approach that could serve the diverse needs of the network.
An all-optical network consisting of optical fiber and optical couplers carries ATM
cells from the head of the ATM PON called the optical line termination (OLT) on the
left to one or more subtending terminals called optical network units (ONU) on the
right. Cells transmitted by the OLT are received by all ONUs and either terminated,
forwarded or discarded based on the cell header. Return cell transmissions from the
ONUs are synchronized such that cells from one ONU do not collide with cells from
other ONUs upon their arrival at the OLT.
A ranging protocol is used to measure the optical distance from the ONU to the
OLT to set a delay compensating for differential distances between the various
ONUs and the OLT. The ATM PON format used by FSAN has now been
accepted as an ITU standard (ITU-T Rec. G.983). The essential features of this
standard are:
It is ATM based.
It supports symmetric (155.52 Mb/s both ways) and asymmetric operation
(622.08 Mb/s downstream, 155.52 Mb/s upstream).
It supports a range of up to 20 km and supports up to a 32-way split (limited
by optical loss budget).
It supports a single fiber for both directions by using wave division multiplexing
(WDM) or dual fiber operation, one for each direction (as shown in Figure
4).
Access Network Topologies
The technologies cited above can be applied in a number of ways based on the
requirements of the network operators. They can be used to provide Internet
telephony service, frame relay and cell relay services, telephony service, digital
broadcast service or video-on-demand.
Fiber-to-the-exchange (FTTExch) offers the flexibility and scalability of FSAN with
the cost benefits of ADSL. Fiber-to-the-cabinet (FTTCab) offers higher speed
access (up to 26 Mb/s) to users from an existing convenient access point in the
network.
This is sufficient capacity for very high-speed data services or multiple digital video
channels. It's interesting to note the cabinets are typically within one km of the user,
which requires network operators deploying FTTCab to install fiber roughly as deep
into the network as cable companies upgrading to hybrid fiber/coax networks.
Fiber-to-the-curb (FTTC) offers even higher speed access (up to 52 Mb/s) and is
most attractive for new builds and high subscriber-density applications.
Fiber-to-the-building or -business (FTTB) uses a highly reliable all dielectric
distribution network to deliver up to 600 Mb/s to a group of users.
Some network operators are looking at FTTB to deliver broadband service to
apartment buildings and other multi-dwelling units. At the same time, other operators
envision FTTB as an alternative to Sonet transmission systems for cost-sensitive
business applications. Fiber-to-the-home (FTTH) is the holy grail of some network
operators. The all-passive network is highly reliable, has low operations costs and
can deliver tremendous amounts of bandwidth.
Benefits to Carriers
The architecture devised by FSAN-Gx allows a common access network platform
to deliver voice, data, video and future services. Elements of the system will be
common to several topologies, helping reduce the cost and simplifying management
for network operators deploying to more than one market segment or migrate from
business to residential customers, or from high-speed access to very high-speed
access customers. A unified specification for full service access networks between
major network operators will increase the worldwide volume of network elements
and lead to cost savings.
FSAN-Gx has made significant progress toward defining a global system
specification for FSANs. They have organized a group of leading telephone
companies and vendors with a common purpose. They have established a flexible
architecture based on ATM and PON. Looking to the future, the FSAN group is
well positioned to bring the promise of the full-service network to reality for millions
of end users worldwide.
Randy Sharpe (rbs@bbt.com) is Director of Advanced Technologies for
BroadBand Technologies Inc., Research Triangle Park, N.C. |
