[opendtv] Re: IEEE Ericsson article on use of LTE for TV

  • From: Craig Birkmaier <craig@xxxxxxxxx>
  • To: opendtv@xxxxxxxxxxxxx
  • Date: Tue, 19 Jun 2012 07:36:59 -0400

At 8:04 PM -0700 6/18/12, Ron Economos wrote:

Your calculator must have one of those fuzzy math chips
in it. The current broadcast television spectrum is 300 MHz
(50 channels numbered 2 - 51). Even less if you subtract
channel 37 and channel 51 (that the FCC is trying to clear
for lower 700 MHz A block users).

If 120 MHz is taken away, that leaves 180 MHz total
and only 108 MHz at UHF.
Not certain how I came up with this number - somethin fuzzy goin on, probably in my head.
Still not certain what the real number is. According to this chart the number should be 138 MHz. 

http://www.ntia.doc.gov/files/ntia/publications/2003-allochrt.pdf

But let's not quibble. For ease of calculation, lets assume 100 MHz.

According to the following paper LTE channel bandwidths are specified as:

http://www.radio-electronics.com/info/cellulartelecomms/lte-long-term-evolution/lte-ofdm-ofdma-scfdma.php

The channel bandwidths that have been chosen for LTE are:
        1.      1.4 MHz
        2.      3 MHz
        3.      5 MHz
        4.      10 MHz
        5.      15 MHz
        6.      20 MHz
The following paper provides some useful information about how the Broadcast LTE infrastructure might be implemented. 

http://world2011.itu.int/sites/default/files/pdf/TS07%20-%20Huschke_Presentation.pdf

There are two important tidbits here:
Channel bandwidths for LTE downlinks (OFDM) can be many sizes. This paper talks about 5 MHz and 8 MHz. 

The effective Bits per Hz is 2.5 or 2.5 Mbps per MHz.

So lets assume 20 MHz channels to minimize losses to guard bands.
Each 20 MHz channel should be able to deliver 50 Mbps, or a total of 250 Mbps if a market has five 20 MHz channels. With five channels it should be possible to checkerboard transmission sites effectively to allow for some of these channels to be differentiated in sub markets.
Using modern codecs, it should be safe to assume that an HD service would be 10 Mbps, a 480P SD service about 2 Mbps, and a variety of lower resolutions could be supported at less than 1 Mbps.
This looks like it would be more than adequate to support all existing broadcast TV services in most U.S. markets, and a LARGE number of new mobile channels. 

But maybe my math is still a bit fuzzy...

Comments?

Regards
Craig
P.S. here is another paper with way too much detail for me, which addresses the issue of site density... 

http://lib.tkk.fi/Dipl/2009/urn100056.pdf


----------------------------------------------------------------------
You can UNSUBSCRIBE from the OpenDTV list in two ways:
- Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org 
- By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word 
unsubscribe in the subject line.

Other related posts:

  1. Home
  2. opendtv
  3. Archive
  4. 06-2012