[EFM] EFM Requirements
As a carrier, Qwest is in agreement that reach, self install, and
compatibility with home networks are all essential. Bandwidth
flexibility, spectral compatibility, and number of video streams are
also important.
Simultaneous Video Streams – We have found a much higher demand for
multiple video streams than indicated by Sherman. I attribute this to
the differences between how DBS markets their service relative to how we
have chosen to do it at Qwest. If you get charged for a second and a
third receiver you will be more conscience about the number of video
streams. Almost all our subscribers use two video streams, and probably
around 25% use three. So ideally, we would like to continue to support
3 streams.
Reach – Our current architecture for VDSL has a reach limitation of
about 4.5 kft at 21.5 Mbps. The 4.5 kft limitation in reach has created
a number of problems for us both in the number of locations in the
network where we have to deploy active electronics (which results in
higher capital and operational expense), and the ability to satisfy
franchise requirements. What we need is the ability to provide VDSL
data rates from a CSA rather than a DA. Translated, that means a 12 kft
reach (24 gauge) versus a 4.5 kft reach. We know that this is pushing
the limits. To help compensate for this, we may consider a solution
that supports 2 video streams, but can be aggregated with a second pair
to get to 3 streams for subscribers willing to pay to have the third
stream. If EFM is unable to provide this kind of reach, it will miss a
major piece of the market. We agree with Sherman that 90% of the
locations will be residential, and 10% business. We also anticipate
that binder fill will be high. Possibly greater than the 70% level
Sherman indicated.
Bandwidth Flexibility – Based on previous EFM meetings, there has been a
desire for 10 Mbps symmetric service. VDSL is an asymmetric. Different
markets will have different needs. Consequently, bandwidth flexibility
and the ability to manage the bandwidth will be important. Bandwidth
flexibility will also make it easier to develop new services (e.g.,
various levels of symmetric/asymmetric Ethernet access to the business
and residence). We also feel it important that the protocols be able to
not only support flexible Ethernet over copper, but to also support PON
so that there can be a potential migration path to FTTH.
Spectral Compatibility – Performance in the presence of other
technologies in the binder group is essential. May be willing to
consider deploying a single technology in a binder group if such an
approach (only having to deal with self interference of a like
technology) would translate into significantly longer reach. We then,
of course, would have to deal with application layer emulation of
services that may be displaced by deploying a single technology in a
binder. Note, however, life-line POTS services must be supported.
Self Install – The service must be deployable without a truck roll.
Truck rolls can make or break a business case. Most DSL deployments
consist of a subscriber plugging their modem into an existing telephone
jack. We want to make this as simple as possible to self install, and
we would prefer to be able to shrink wrap the customer-side EFM devices
and distribute them through retail channels.
Compatibility with Home Networks – Splitterless compatibility would be
ideal, and most DSL deployments are splitterless. While we would like
splitterless compatibility with all Home Networking systems, this may
not be possible.
Charles