ASTC 8VSB not a bad choice for Canada, as 95% of the househoulds get TV via cable -- where 256 QAM is the standard...

On Wed, 7 Mar 2007 14:52:05 -0800, Mark Crispin
wrote:

On Wed, 7 Mar 2007, Bob Miller wrote:

The end game IMO is wireless broadband and broadcast. There is nothing else
IMO. No cell phones, no cable, no satellite including Sirius or XM merger or
not.

Speaking as someone who is actually working on wireless broadband
technology, you are full of ****. You vastly underestimate both costs and
feasibility, particularly in any sort of wide-area deployment.

For fixed reception the broadband fiber to the home will sooner or
later replace the cable network and the satellite feeds. Private base
stations at home and offices will provide connection to the fiber.

Wireless broadband in built up areas is not going to be a problem, but
in low population density areas the situation is going to be
problematic and most likely people living in these areas will have to
accept a lower level of service.

Cable has a huge advantage over OTA in that it has its own dedicated
ether. It doesn't have to share bandwidth with external entities.

The capacity growth potential of the coaxial cable is close to the
limit. Sure, it may be possible to add one or two bits to each QAM
symbol by using better amplifiers. Higher frequencies might also be
used, the cable loss increases and amplifiers have to be inserted at
closer intervals. Cascading amplifiers will worsen the noise and
intermodulation problems and making it hard to transfer even the
current number of bits/symbol. With coaxial "last mile" or "last 100
m" fed by a fiber, perhaps the capacity could be increased by only 2-3
times from the current situation.

Satellite gets its bandwidth by using GHz frequencies, but at the cost of
requiring line-of-sight from transmitter to receiver and carefully aimed
dish antennas.

At least the geostationary system is close to the capacity limit. You
can not go much higher in frequency due to atmospheric losses. The
spectral reuse cannot be increased in the geostationary satellite belt
by moving satellites closer to each other, since the receiver antenna
beamwidth would have to be narrower, i.e. the receiver antenna would
have to be larger and in many cases become too large for e.g. balcony
installations. A large antenna with a narrow beam is harder to aim,
may require active compensation for changes in atmospheric refraction
index, the satellite station keeping must be more accurate, consuming
more fuel and shortening satellite lifetime.

In order to reuse the satellite spectrum more efficiently will require
a larger angular separation between satellites, i.e. non-geostationary
orbits with large inclination would be required, but in this case,
each receiving antenna would have to be mechanically or electronically
steerable.

OTA has to suffer with a lot of limitations, including other users of the
ether. The reason why 802.11 is so severely power-limited is because it
uses frequencies owned by others on a "low-power" exemption.

When aiming for a large total system throughput, the cell size must be
small, making it possible to reuse the same frequency more often, thus
as small power levels should be used as possible. The cost of some
WLAN components are soon so low that it is feasible to integrate these
in every lamp-post and have cell sizes of 30-100 m in densely
populated areas and along main roads.

With such small cell sizes, the atmospheric losses are not an issue
even high in EHF bands, but sooner or later the attenuation in glass
(both in house and cars) will become the limiting factor. However, it
takes a few years, before the technology in 10-100 GHz is affordable
for such dense networks. The future throughput potential for future
wireless networks is huge, compared to current RF systems.

Paul

Paul Keinanen wrote:
On Wed, 7 Mar 2007 14:52:05 -0800, Mark Crispin
wrote:

On Wed, 7 Mar 2007, Bob Miller wrote:

The end game IMO is wireless broadband and broadcast. There is nothing else
IMO. No cell phones, no cable, no satellite including Sirius or XM merger or
not.
Speaking as someone who is actually working on wireless broadband
technology, you are full of ****. You vastly underestimate both costs and
feasibility, particularly in any sort of wide-area deployment.

For fixed reception the broadband fiber to the home will sooner or
later replace the cable network and the satellite feeds. Private base
stations at home and offices will provide connection to the fiber.

Wireless broadband in built up areas is not going to be a problem, but
in low population density areas the situation is going to be
problematic and most likely people living in these areas will have to
accept a lower level of service.

Cable has a huge advantage over OTA in that it has its own dedicated
ether. It doesn't have to share bandwidth with external entities.

The capacity growth potential of the coaxial cable is close to the
limit. Sure, it may be possible to add one or two bits to each QAM
symbol by using better amplifiers. Higher frequencies might also be
used, the cable loss increases and amplifiers have to be inserted at
closer intervals. Cascading amplifiers will worsen the noise and
intermodulation problems and making it hard to transfer even the
current number of bits/symbol. With coaxial "last mile" or "last 100
m" fed by a fiber, perhaps the capacity could be increased by only 2-3
times from the current situation.

Satellite gets its bandwidth by using GHz frequencies, but at the cost of
requiring line-of-sight from transmitter to receiver and carefully aimed
dish antennas.

At least the geostationary system is close to the capacity limit. You
can not go much higher in frequency due to atmospheric losses. The
spectral reuse cannot be increased in the geostationary satellite belt
by moving satellites closer to each other, since the receiver antenna
beamwidth would have to be narrower, i.e. the receiver antenna would
have to be larger and in many cases become too large for e.g. balcony
installations. A large antenna with a narrow beam is harder to aim,
may require active compensation for changes in atmospheric refraction
index, the satellite station keeping must be more accurate, consuming
more fuel and shortening satellite lifetime.

In order to reuse the satellite spectrum more efficiently will require
a larger angular separation between satellites, i.e. non-geostationary
orbits with large inclination would be required, but in this case,
each receiving antenna would have to be mechanically or electronically
steerable.

OTA has to suffer with a lot of limitations, including other users of the
ether. The reason why 802.11 is so severely power-limited is because it
uses frequencies owned by others on a "low-power" exemption.

When aiming for a large total system throughput, the cell size must be
small, making it possible to reuse the same frequency more often, thus
as small power levels should be used as possible. The cost of some
WLAN components are soon so low that it is feasible to integrate these
in every lamp-post and have cell sizes of 30-100 m in densely
populated areas and along main roads.

With such small cell sizes, the atmospheric losses are not an issue
even high in EHF bands, but sooner or later the attenuation in glass
(both in house and cars) will become the limiting factor. However, it
takes a few years, before the technology in 10-100 GHz is affordable
for such dense networks. The future throughput potential for future
wireless networks is huge, compared to current RF systems.

Paul

Insanely huge compared to current RF systems, approaching infinity and
incredibly inexpensive. 2.5 years or less before this will start to hit
the mainstream. Bluetooth prices for multi Gbps wireless in the medium term.

Super redundant networks that can and will be built and maintained by
members of the household. NO need for most of the current infrastructure
and its high cost, cable FIOS, satellite, cellular as it pertains to
individuals or households. Inter city and country fiber and satellite
connections to feed this but all last miles will be wireless.

Bob Miller

In article ,
Paul Keinanen wrote:

In order to reuse the satellite spectrum more efficiently will require
a larger angular separation between satellites, i.e. non-geostationary
orbits with large inclination would be required, but in this case,
each receiving antenna would have to be mechanically or electronically
steerable.

Or maybe statites http://en.wikipedia.org/wiki/Statite (a "satellite"
which uses a solar sail to stay in a fixed position relative to the
earth), but that's still in the realm of science fiction right now.

"Paul Keinanen" wrote:

When aiming for a large total system throughput, the cell size must be
small, making it possible to reuse the same frequency more often, thus
as small power levels should be used as possible. The cost of some
WLAN components are soon so low that it is feasible to integrate these
in every lamp-post and have cell sizes of 30-100 m in densely
populated areas and along main roads.

With such small cell sizes, the atmospheric losses are not an issue
even high in EHF bands, but sooner or later the attenuation in glass
(both in house and cars) will become the limiting factor. However, it
takes a few years, before the technology in 10-100 GHz is affordable
for such dense networks. The future throughput potential for future
wireless networks is huge, compared to current RF systems.

I agree with this completely. However, it is a fundamentally different
model of OTA TV. What you are describing is sort of one step removed
from a cable network to the home, with wifi distribution inside the
home. The foundation of such a system is still a cabled network.

Traditional terrestrial TV distribution has a few advantages over cable,
one of which is lower infrastructure cost. Your system, or the similar
LMDS concepts from a couple of decades ago, are sort of a hybrid between
cabled networks and more traditional OTA TV.

To allow more and more reuse of RF spectrum, the wireless link keeps
getting shorter and shorter, the supporting cabled network structure
more and more elaborate. As always, there is a tradeoff.

Bert

On Thu, 08 Mar 2007 14:45:59 GMT, Bob Miller wrote:

Paul Keinanen wrote:
On Wed, 7 Mar 2007 14:52:05 -0800, Mark Crispin
wrote:

For fixed reception the broadband fiber to the home will sooner or
later replace the cable network and the satellite feeds. Private base
stations at home and offices will provide connection to the fiber.

clip

[small cell wireless broadband systems]

With such small cell sizes, the atmospheric losses are not an issue
even high in EHF bands, but sooner or later the attenuation in glass
(both in house and cars) will become the limiting factor. However, it
takes a few years, before the technology in 10-100 GHz is affordable
for such dense networks. The future throughput potential for future
wireless networks is huge, compared to current RF systems.

Paul

Insanely huge compared to current RF systems, approaching infinity and
incredibly inexpensive. 2.5 years or less before this will start to hit
the mainstream. Bluetooth prices for multi Gbps wireless in the medium term.

Super redundant networks that can and will be built and maintained by
members of the household.

Unfortunately this does not work in the long run. While there might be
great enthusiasm in the beginning, but in the long run, when there are
user problems that needs to be solved, when equipment break down or
when outdated equipment needs to be replaced, is there really much
enthusiasm after a few years ?

Based on the experience with the amateur radio packet radio network in
the 1980/90's, I would say no.

There needs to be some organisation that will maintain the system
after a few years, even if the system can be started with a number of
enthusiast.

Paul

On Thu, 8 Mar 2007 15:53:07 GMT, "Albert Manfredi"
wrote:

"Paul Keinanen" wrote:

When aiming for a large total system throughput, the cell size must be
small, making it possible to reuse the same frequency more often, thus
as small power levels should be used as possible. The cost of some
WLAN components are soon so low that it is feasible to integrate these
in every lamp-post and have cell sizes of 30-100 m in densely
populated areas and along main roads.

With such small cell sizes, the atmospheric losses are not an issue
even high in EHF bands, but sooner or later the attenuation in glass
(both in house and cars) will become the limiting factor. However, it
takes a few years, before the technology in 10-100 GHz is affordable
for such dense networks. The future throughput potential for future
wireless networks is huge, compared to current RF systems.

I agree with this completely. However, it is a fundamentally different
model of OTA TV. What you are describing is sort of one step removed
from a cable network to the home, with wifi distribution inside the
home. The foundation of such a system is still a cabled network.

Traditional terrestrial TV distribution has a few advantages over cable,
one of which is lower infrastructure cost. Your system, or the similar
LMDS concepts from a couple of decades ago, are sort of a hybrid between
cabled networks and more traditional OTA TV.

You describe it as an extension to the cable network, I would describe
it as a very high throughput cellular phone network and still we are
talking about the same thing :-).

Anyway, the throughput of such system is so great that after all, the
broadcasting services only occupy a small fraction of the total
throughput, just like the VoIP traffic currently occupy only a small
part of the internet traffic. Only a few decades ago, the majority of
all landline activity was telephone conversations.

The same is going to happen to the broadcasting service, which once
was the major player in all kinds of radio communication, but in
future, it will just be one ordinary player in radio communication.

Paul

On Mar 8, 12:39 pm, Paul Keinanen wrote:

You describe it as an extension to the cable network, I would describe
it as a very high throughput cellular phone network and still we are
talking about the same thing :-).

Anyway, the throughput of such system is so great that after all, the
broadcasting services only occupy a small fraction of the total
throughput, just like the VoIP traffic currently occupy only a small
part of the internet traffic. Only a few decades ago, the majority of
all landline activity was telephone conversations.

The same is going to happen to the broadcasting service, which once
was the major player in all kinds of radio communication, but in
future, it will just be one ordinary player in radio communication.

I agree that in time the Internet will be able to support HDTV streams
with ease. A couple of years ago I figured it was about 10 years away
at least, and I'll stick to that prediction for now.

The only problem I have with your post is that it's not the RF mini-
links you're advocating that will solve the basic Internet capacity
problem by themselves. Your "high throughput cellular phone network"
will require a much faster basic infrastructure, a cabled
infrastructure, with possibly additional QoS mechanisms, compared with
what is deployed now in the Internet. If you want complete flexibility
in routing HDTV to whoever, whenever they choose, and avoid a walled
garden network architecture.

The RF last-few-meters link, or the fiber-to-the-home local loop, do
not resolve the system bottlenecks.

As of now, cable TV, DBS, or DTT, are a very good way to offload that
high capacity TV traffic from the interactive Internet backbones.

As a matter of fact, when the US telecom giant Verizon deployed their
FTTH system recently, for broadband Internet access, TV, and telephone
service, they also came to that conclusion. They use a DWDM scheme
over a passive optical network (PON). But instead of using IP to
deliver television, they assign their TV broadcast channels to one of
the 32 wavelengths, and simply broadcast the TV channels over that
frequency as a cable system would. They use IP for video on demand,
telephone, and Internet broadband connectivity.

This allows Verizon to use standard coax cable and cable STBs for
broadcast TV distribution in the home, and it keeps their internal
routers a lot cheaper than they would have to be if it were a true
IPTV network. I think of it as offloading the really fast stuff, which
is easily handled as a true broadcast. And, they say, they have the
flexibility to introduce IPTV in the future, once they've installed
the FTTH PON everywhere.

Bert

On Mar 8, 12:39 pm, Paul Keinanen wrote:

You describe it as an extension to the cable network, I would describe
it as a very high throughput cellular phone network and still we are
talking about the same thing :-).

In the previous reply, I figured your were using the Internet between
all these very small cells. I notice that maybe you are not. If you
are not, and are instead creating a broadcast system with very short
RF links, okay, but you are still depending on a much more labor-
intensive infrastructure than traditional OTA TV. Especially in the
sprawling suburbs of many US cities (and increasingly true in Europe
too).

Bert

Albert Manfredi wrote:
"Paul Keinanen" wrote:

When aiming for a large total system throughput, the cell size must be
small, making it possible to reuse the same frequency more often, thus
as small power levels should be used as possible. The cost of some
WLAN components are soon so low that it is feasible to integrate these
in every lamp-post and have cell sizes of 30-100 m in densely
populated areas and along main roads.

With such small cell sizes, the atmospheric losses are not an issue
even high in EHF bands, but sooner or later the attenuation in glass
(both in house and cars) will become the limiting factor. However, it
takes a few years, before the technology in 10-100 GHz is affordable
for such dense networks. The future throughput potential for future
wireless networks is huge, compared to current RF systems.

I agree with this completely. However, it is a fundamentally different
model of OTA TV. What you are describing is sort of one step removed
from a cable network to the home, with wifi distribution inside the
home. The foundation of such a system is still a cabled network.

Traditional terrestrial TV distribution has a few advantages over cable,
one of which is lower infrastructure cost. Your system, or the similar
LMDS concepts from a couple of decades ago, are sort of a hybrid between
cabled networks and more traditional OTA TV.

To allow more and more reuse of RF spectrum, the wireless link keeps
getting shorter and shorter, the supporting cabled network structure
more and more elaborate. As always, there is a tradeoff.

Bert

Not what I am talking about. The links don't get shorter necessarily. In
a densely populated area you may want to have a limited number of
viewers per NOC but any given link will be possible at up to 5 miles.
Much farther in many cases.

NO trade off. Just far lower cost and no need for most of the
infrastructure cost we have today. No satellite, no cable, no fiber
except between cities and countries and there is still a ton of dark
fiber there.

And who is talking about models of TV? We were talking about the long
term viability of high cost, high maintenance networks. If you can get
all the same content you can get today via the current TV model with a
new model at much lower cost then the old model dies.

And if it happens to come with a multi Gbps back channel and works
mobile all the better

Bob Miller

Paul Keinanen wrote:
On Thu, 08 Mar 2007 14:45:59 GMT, Bob Miller wrote:

Paul Keinanen wrote:
On Wed, 7 Mar 2007 14:52:05 -0800, Mark Crispin
wrote:

For fixed reception the broadband fiber to the home will sooner or
later replace the cable network and the satellite feeds. Private base
stations at home and offices will provide connection to the fiber.

clip

[small cell wireless broadband systems]

With such small cell sizes, the atmospheric losses are not an issue
even high in EHF bands, but sooner or later the attenuation in glass
(both in house and cars) will become the limiting factor. However, it
takes a few years, before the technology in 10-100 GHz is affordable
for such dense networks. The future throughput potential for future
wireless networks is huge, compared to current RF systems.

Paul

Insanely huge compared to current RF systems, approaching infinity and
incredibly inexpensive. 2.5 years or less before this will start to hit
the mainstream. Bluetooth prices for multi Gbps wireless in the medium term.

Super redundant networks that can and will be built and maintained by
members of the household.

Unfortunately this does not work in the long run. While there might be
great enthusiasm in the beginning, but in the long run, when there are
user problems that needs to be solved, when equipment break down or
when outdated equipment needs to be replaced, is there really much
enthusiasm after a few years ?

Based on the experience with the amateur radio packet radio network in
the 1980/90's, I would say no.

There needs to be some organisation that will maintain the system
after a few years, even if the system can be started with a number of
enthusiast.

Paul

Agreed but I don't think the network I am talking about is going to need
a lot of maintenance. People, me included, don't want to maintain my own
computer and home network but I do.

But you are right, long term I will find a way to farm out all the
problems I have with updating software, rebooting the network, buying
software, installing software and just keeping up with the technology.

But we have done it. Most of the population of the US has bought into
the computer algorithm. And that is far more complicated than what I am
talking about.

Basically we are talking about another connection on your home network,
a multi Gbps connection and the maintaining of that connection.

On one side you have the problem of keeping folks happy who might have
to maintain that network, something many are already doing, on the other
side you have the savings from doing it yourself.

And why not have a company that says we will maintain your network for a
fee,relax. If you want to have someone come out and switch out a server
that cost $50 or re point an antenna so be it.

Maybe the network will be so trouble free that a contract for this will
be very low cost. A lot less than the contract we now have for cable IMO.

Bob Miller