Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02508463 2005-05-27
METHOD TO COMPENSATE FOR RAIN FADE
IN A DIGITAL VIDEO BROADCAST SATELLITE SYSTEM
FIELD OF THE INVENTION
The present invention relates to satellite communications and to earth
stations
(terminals) for geostationary satellites using "bent pipe" (analog repeater)
transponders.
More particularly, the invention relates to large networks of receive only
stations and one
transmit station. This is usually the case for broadcast of television signals
over the
satellite medium.
BACKGROUND OF THE INVENTION
This invention is a method to solve a long standing problem of broadcasting
digital
video signals by satellite. Older methods of broadcasting video signals by
satellite used
frequency modulation which where analog in nature and are not covered by this
invention.
The problem occurs mostly on broadcast frequencies in Ku and Ka band.
The propagation characteristics of those bands are such that moisture present
on
the propagation path absorbs the signal and thus causes impairments in the
reception. The
impairments can cause the signal to noise ratio of the signal to go below the
receiver
threshold, thus causing complete signal loss. This loss results in loss of
video content for
the duration of the fade, which causes viewer dissatisfaction. The effect of
signal
attenuation is three fold:
1. A first effect is to reduce signal strength while the noise caused by the
receivers
low noise amplifier remains constant.
2. A second effect is to increase slightly the noise floor of the station
because the
moisture is warmer than the background space. This effect is small for the low
cost, small sized parabolic dish.
-1-
CA 02508463 2005-05-27
3. Depolarization of the received signal: part of the signals energy appears
in the
orthogonal polarization. Experimental results reported indicate this effect to
be
negligible compared to the first item.
Several methods exist to control the uplink part of the transmission to ensure
that
the broadcast signal from the satellite is essentially immune to fade.
Downlink fades are a
different problem. Because the plurality of receivers are dispersed in wide
areas that can
experience fade events that are specific to the location, conventional means
to control
uplink power do not apply.
It is well known from communication theory the relationship between the signal
to
noise expressed as the ratio Eb/No, the bit rate and the forward error coding
(fec) rate. It
is assumed that the transmit power out of the satellite transponder to the
multiple
receivers is constant. The only 2 other variables to improve the received
signal to noise
are the bit rate and the fec rate. A bit rate reduction of a factor of 2
results in a signal to
noise improvement of 3 dB.
The receiver threshold is usually defined as the worse bit error rate that can
be
tolerated for a given Eb/No. The bit error rate is a function of the Eb/No,
modulation and
type of forward error coding. Thus a threshold improvement of 3 dB is obtained
by
reducing the bit rate by a factor of 2. An decrease in fec rate from %Z ( 1
coding bit for each
data bit) to 1/3 (2 coding bits for each data bit) will produce an improvement
in bit error
rate performance and thus reduced threshold that depends on the particular
scheme used.
There are many combinations of bit rate and fec rates that can be used to
improve the
receiver threshold.
-2-
CA 02508463 2005-05-27
The new scheme proposed in this invention multiplexes into a single signal
digital
streams of at least 2 different bit rates. These alternate bit rates can also
use varying
coding rates. Several multiplexing scheme can be devised and some are given in
a
following example. In all cases the different choices of combination bit rates
and coding
rates are known in advance by the receiver. The receiver decodes the different
streams
and selects the one that has the best signal to noise.
During clear sky reception conditions, the receiver would select to display to
the
viewer the digital stream of the highest bit rate that corresponds to the best
audio and
video quality. During fade conditions, the receiver would select to display
one of the
lower bit rate streams that correspond to a lower audio and video quality.
As an example, assume a high quality broadcast signal that has a channel bit
rate
of 10 Mb/s and a high compression lower quality broadcast signal of 1 Mb/s.
Assume that
both have the same forward error coding method and the same modulation: qpsk.
Assume
further that the link budget (sizing of antenna size, power from the
satellite, noise
contribution from low noise amplifier) is designed (under clear sky
conditions) to provide
negligible bit error rate for the 10 Mb/s broadcast signal at an Eb/No of 5 dB
(see Figure 8
for the definition of this parameter) . The bit error rate performance in
presence of
additive noise is given by articles published in learned journals. These
results are given
for a specific modulation method and forward error coding scheme.
When a received signal is attenuated by 10 dB by a fade event (received power
is
1/10 the clear sky value), the signal power is decreased by 10 dB while the
noise density
remains approximately constant. From Figure 8, the Eb/No is decreased by 10 dB
for the
Mb/s signal. This worsens the bit error rate and renders the high quality
transmission
-3-
CA 02508463 2005-05-27
unusable because of the high number of bit errors. Note however that in Figure
8, the
lower bit rate link has a bit period that is 10 times longer than the high bit
rate. The result
is that the Eb/No for the low bit rate is the same SdB under the fade
condition. In this
condition, the bit error rate is negligible for this stream. The appropriate
decompression
scheme is selected by selecting the best output stream and a reduced quality
broadcast is
provided to the viewer instead of no reception at all.
While the signal strength is reduced by one tenth, those normally skilled in
designing wireless demodulators can design signal processing schemes that can
handle
this requirement.
The previous example is independent of the exact method used to multiplex the
two streams. Those normally skilled in the design of modems will select the
method most
appropriate.
Furthermore, the method can be used to provide preview of broadcast material
using the lower resolution stream for the receivers that do not have access to
the better
quality material. When the receiver does not have the authorization from the
service
provider to view the broadcast material, the receiver can operate in low
resolution mode
to provide a preview of the broadcast.
BRIEF DESCRIPTION OF THE DRAWINGS
Having thus generally described the invention, reference will be made to the
accompanying drawings illustrating embodiments thereof, in which:
Figure 1 is an overall block diagram of a satellite video broadcast systems;
Figure 2 is a block diagram of an improved transmitter scheme
Figure 3 is a block diagram of variable transmission streams;
-4-
CA 02508463 2005-05-27
Figure 4 is a block diagram illustrating two streams being multiplexed;
Figure 5 is a block diagram providing an example of a multiplexing scheme
where
three streams are multiplexed into a single carrier;
Figure 6 is a block diagram providing a time domain view of Figure 5;
Figure 7 is a block diagram of the receiver; and
Figure 8 illustrated the definition of E,,//No in a data transmission system
Referring to Figure l, for an overall block diagram of a satellite video
broadcast
system, a transmitter emits a signal to the satellite. This signal is often
controlled by an
uplink power control scheme that compensates for fades in the transmitter to
satellite
path. The output of the satellite transponder has a constant output power on
the path from
satellite to the receivers, irrespective of fades in the uplink path when such
a scheme is
used.
The receivers can however be in a clear sky reception area or individually
affected
by fades that are specific to their geographic location. The uncompressed
broadcast
material supplied by the producer of the material is fed to at least 2
compressors shown in
Figure 3 that use the same compression scheme but with different compression
ratios.
These compressors perform both the voice and image compression. Furthermore,
they
multiplex the compressed voice and image into a single data stream. Each
compressor has
a different output bit rate corresponding to the different broadcast quality
required. To
compensate for the delay difference in the variable ratio compression, a delay
equalizer in
Figure 3 is introduced to simplify the task of changing dynamically between
compressors
at the receiver. Examples of compression scheme are mpeg 2 and mpeg 4.
-5-
CA 02508463 2005-05-27
Referring to Figure 2, several streams of data corresponding to the different
broadcast quality are multiplexed together into a single signal split between
the in phase
and quadrature channel of a conventional IQ modulator.
Several schemes are possible to perform this multiplexing. One example is
provided in Figure 4 for the case where only 2 streams are multiplexed. In
such a scheme,
often called unbalanced qpsk, the 2 unequal bit rate streams are simply
filtered with a
filter to ensure inter symbol interference (isi) free reception with a Nyquist
type filter.
After conversion to an analog format and harmonics removed, the signal is fed
to a
conventional modulator.
A well known alternate format for IQ modulator that converts to analog after
the
modulation is equivalent to the implementation in the diagram Figures 4 and 5.
Figure 6 shows an I channel time domain representation the scheme in Figure 5.
In
that case, two bursts of different data rates are shown. In the event the
signal to noise is
too low for the highest data rate burst to synchronize and decode correctly,
the following
lower rate burst can receive the data. The preamble is used for initial
synchronization of
the Garner phase, symbol timing and signal amplitude. Figure 5 shows a
preamble
insertion block that formats the incoming 2 streams into separate data bursts.
For
synchronization purposes, each burst is preceded by a preamble. A guard time
is included
between each burst to prevent interference between successive bursts.
Numerous variations of multiplexing schemes on the transmitter are possible
and
are not shown.
A block diagram of the receiver is shown in Figure 7. After processing by the
receive electronics and the IQ demodulator the signal is converted to a
digital form and
-6-
CA 02508463 2005-05-27
filtered with the appropriate matched filter. Each stream has a signal to
noise estimator.
The stream that has the best signal to noise is selected for decompression.
It will be understood that the above described embodiment is for purposes of
illustration only and that changes and modifications may be made thereto
without
departing from the spirit and scope of the invention.
