Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: [Phys-l] Digital TV signal delay



According to what I have found the minimum time delay may be 0.5s when
switching channels because:

<http://www.freshpatents.com/Image-display-delay-time-shortening-apparatus-d
uring-channel-change-method-thereof-and-broadcasting-receiving-apparatus-the
reof-dt20060817ptan20060184981.php>
---------------------------------------------------------------------------
[0007] However, when a digital TV receiver displays on a screen a digital
broadcasting signal by receiving the digital broadcasting signal according
to ATSC (Advanced Television Standards Committee) protocol, much more time
is consumed compared with displaying of an image on a screen by receiving
the analogue broadcasting signal because a moving picture data encoded by
MPEG (Moving Picture Experts Group) protocol should be decoded.

[0008] To be more specific, moving picture data of a digital broadcasting
signal differs in terms of resolutions, aspect ratios of pixels, and display
period of images. As a result, a digital broadcasting signal contains a
sequence header including information on resolutions and aspect ratios of
pixels. A digital TV receiver extracts a sequence header from a received
digital broadcasting signal and establishes an initial value of a video
decoder according to the extracted sequence header. The video decoder
decodes a video stream in response to the initial value.

[0009] The sequence header is stored in an elementary stream packet and
transferred. Therefore, the digital TV receiver uses a Program Specific
Information (PSI) included in the received digital broadcasting signal to
extract a Program Identifier (PI), and seeks the elementary stream packet
from the extracted PI to extract a sequence header.

[0010] The sequence header is recommended to be inserted into a digital
image signal per 0.5 second and transmitted in the ATSC digital television
standard document A/54, but it is just a recommendation. Accordingly, there
is a difference in an insertion time of the sequence header according to
broadcasting stations or broadcasting contents producers, and practically,
broadcasting stations or broadcasting contents producers insert a sequence
header into a digital image signal at a time interval longer than the
recommended time of 0.5 second.

[0011] When a channel of a digital TV receiver is changed, a video decoder
immediately implements a decoding operation to thereby enable to display an
image within an instant time if a sequence header is included in a digital
broadcasting signal received at the start of the channel change.

[0012] However, if the sequence header is not contained in the digital
broadcasting signal received at the start of channel change, an initial
value of the video decoder cannot be established resulting in
non-implementation of the decoding operation. Consequently, the video
decoder waits until the sequence header is received, and starts to display
the image on a screen once the sequence header is received.
----------------------------------------------------------

However when switching channels the bandwidth must be equalized which takes
time also. Some of these delays are circumvented by keeping tables for all
channels which allow immediate switching. But time delays are still there.
The decode time delay would appear to be a small fraction of a second, as
long as the signal is relatively clean.

The time delay in transmission is MUCH longer, and there is no reason for
hiding it. The station managers do need to take steps such as changing the
local clocks to read a delayed time. In the interim the analog signal can
be transmitted after being decoded from the digital feed so that analog
customers will not be given a wrong time.

Incidentally the ear is quite sensitive to time delay shifts, especially
when 2 ears are involved.

Incidentally, as usual the US had bought into a video system which is
completely different from the rest of the world. China and Brazil have also
bought into different systems. So it appears there might be a North
American, S. American, Chinese, and the rest of the world set of standards.
Our standard is criticized for being too complicated. At one time this made
sense because of the difference between the 50 and 60 Hz power grids, and
also the early adoption of TV, and color TV by the US. Now it appears to be
just nationalistic posturing, or perhaps the fact that the US does not have
a firm national standards organization. But with the falling price of
digital circuits, TVs might eventually be sold as compatible with all
standards.


John M. Clement
Houston, TX