Re: Dynamic Multipath


"Bob Miller" <robmx@xxxxxxxxxxxxx> wrote in message
news:FYdtf.305$%W1.133@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
> Neil - Salem, MA USA wrote:
>> "Bob Miller" <robmx@xxxxxxxxxxxxx> wrote in message
>> news:VG4tf.2144$M%4.2041@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
>>
>>>Here is a poster on AVSForum who tells of dynamic multipath problems.
>>>Since an airplane can fly over virtually any spot in the US then anyone
>>>can suffer from this unnecessary interference. Why does the US still have
>>>this junk 8-VSB? Congress and the FCC both knew of this problem in 1998
>>>and ignored it.
>>>
>>>http://www.avsforum.com/avs-vb/showthread.php?p=6807808&&#post6807808
>>>
>>>"I can answer that one.
>>>
>>>I live between Dulles airport in Northern Virginia and the broadcast
>>>towers in DC and Maryland. My Winegard UHF/VHF rooftop antenna works very
>>>well. The problem that I have is that every time and airplane flies
>>>between my antenna and the broadcast towers, I get multipath interference
>>>and my picture pixellates for a second or two and then comes back. Mind
>>>you, I have to be able to see and hear the planes for this to happen.
>>>
>>>If I lived on the other side of the flightpath, there would be no
>>>problem.
>>>
>>>The same thing happens on analog broadcasts, except the picture will
>>>wobble instead of pixellate. "
>>>
>>>[snip]
>>>
>>>Bob Miller
>>
>>
>> Bob,
>>
>> First let me say that I read almost all of your posts because you do have
>> quite a bit of knowledge about ATSC and COFDM, and I respect such
>> knowledge. However, in this case I must take issue.
>>
>> On the same page that you quote is another post, a reply to the post you
>> quote:
>>
>> http://www.avsforum.com/avs-vb/showthread.php?p=6808344&&#post6808344
>>
>> In this reply the author says, "Hey Cawgijoe, I am your neighbor right in
>> Centreville as well.... It seems like the flights come in from all
>> directions here so its hard to avoid. Sometimes i really wonder if they
>> need to fly so low since the airport is still a good 15-20 miles away.
>>
>> "As far as multipath interference goes, I used to have an old generation
>> OTA receiver (Toshiba DST3000 HD/Directv box) and during prime time hours
>> things were pretty much unwatchable.
>>
>> "I have since upgraded to a newer generation receiver (Hughes HD Tivo)
>> and very rarely get picture or sound freezes now.
>>
>> "I read that some of the upgrades to the tuners specifically were done to
>> address multipath reception issues."
>>

Let's get our understanding of multipath addressed first and then we will
understand why faster ATSC tuners have less problems with multipath.

Ghosts are multipath but they are generally referring to stable non-changing
reflections from buildings. Any ATSC tuner can deal with ghosts as long as
either the reflection or the original signal is at least 10 db stronger than
the other. The ATSC tuner tries to lock on the strongest signal first,
checks for a magic number (a simple analogy) and if received with no errors
locks at that signal level. IF corrupted it tries a lower signal level
until it gets an uncorrupted number. This requires time to find the
equalization (magic number) data stream that is broadcast three times
randomly in each frame and decode the entire stream until it finds the
equalization data. With severe ghosting and or weak signal that can take
several seconds. There is a proposal out by Samsung to change the random
equalization pulses to fixed timed pulses so that the tuner has less trouble
finding the equalization pulse and can lock faster ( min time at least 1/100
of a second to 1/30th second) vs. sometimes 3 seconds on third generation
ATSC tuners.

Multipath is for instance the changing signal levels of an airplane ghost or
the second signal (ghost) that's delayed by travel time through the lower
atmosphere where temperature causes varying densities of the air thus
varying signal and phase levels. These cause equalization problems for the
ATSC tuner but if the tuner is fast enough it can keep up with atmospheric
density multipath signal level changes which are relatively slow 5 db signal
level changes over 1/2 sec. 5th generation tuners can lock on weak signals
that have multi-path issues easier than older slower tuners. The airplane
multipath generally changes faster than atmospheric multipath so that's a
bigger problem.

Portable digital TV is not possible with the current standards, that's why
the Samsung proposal is so important. Right now it's impossible to have a
fast enough ATSC tuner for mobile applications, besides the processing power
involved, a tuner and digital processor that is that fast will be a battery
hog. If the equalization pulse is sent at a predetermined time after the
carrier tone, equalization will not require processing power and slow
inexpensive tuners can reliably be designed for portable applications.

The random nature of the equalization pulse was not a planned feature just
an oversight by the engineers of the atsc standard. They wanted a random
data stream or one that appears to be random so that the energy of the
transmission is spread throughout the entire transmission band to reduce
adjacent channel interference and extend the range of the transmission. The
equalization pulse just got randomized with the rest of the data being
transmitted.

This problem with equalization is the biggest disadvantage 8-VSB has when
compared to CODFM. That CODFM is designed for the urban market with
multiple transmitters in a coverage area is an advantage for portable
applications as line of sight is usually easier for CODFM than for 8-VSM
that only has one tower in a coverage area.

For me in my coverage area situated 50 miles south of a large city with 10
stations and 40 miles north of an area with 4 stations I'd be out of luck
with CODFM but with 8-vsb I get all these stations and only have problems
with co-channel when conditions allow skip of stations that are as far away
as 300 miles.



.

Relevant Pages

  • Re: Dynamic Multipath
    ... The problem that I have is that every time and airplane flies between my antenna and the broadcast towers, I get multipath interference and my picture pixellates for a second or two and then comes back. ... The ATSC tuner tries to lock on the strongest signal first, checks for a magic number and if received with no errors locks at that signal level. ... There is a proposal out by Samsung to change the random equalization pulses to fixed timed pulses so that the tuner has less trouble finding the equalization pulse and can lock faster vs. sometimes 3 seconds on third generation ATSC tuners. ... These cause equalization problems for the ATSC tuner but if the tuner is fast enough it can keep up with atmospheric density multipath signal level changes which are relatively slow 5 db signal level changes over 1/2 sec. 5th generation tuners can lock on weak signals that have multi-path issues easier than older slower tuners. ...
    (alt.tv.tech.hdtv)
  • Re: HDTV ghosting and brightness flicker
    ... analog TV caused by a multipath problem. ... >> generation receiver ... > introduced on the same transmission path used by an analogue signal -they ... > overlapping digital signals, which just cause major errors in the digital ...
    (microsoft.public.windows.mediacenter)
  • Re: Problem with dropped calls
    ... Bottom line - CDMA compensates and works WITH multipath. ... >> MULTIPATH AND RAKE RECEIVERS ... Due to its wide bandwidth and rake receivers, CDMA uses the multipath signals and combines them to make an ...
    (alt.cellular.verizon)
  • Re: Multipath Delay
    ... Rune Allnor skrev: ... >> multipath fading channel and 8PSK modulation. ... > Assume a reflection coefficient R at the reflecting surface ... > For broadband signals, there might be lots of nulls in ...
    (comp.soft-sys.matlab)
  • Re: Friends of Radio 3
    ... The CT7000 has a meter to display multipath. ... I noticed in a review in a recent mag that one tuner ... orchestral concert. ... Audio Misc http://www.st-and.demon.co.uk/AudioMisc/index.html ...
    (uk.rec.audio)