• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

FAQ: MPEG-DVB-tech sat info


Honorary Admin
My Location
Cloud Cuckoo Land
I don't know if I ever posted this here.
Feel free to add or correct.

FAQ: MPEG-DVB-tech sat info

PIDs = Packet IDentifiers. They are parameters which identify particular packets (188-byte chunks) of the digital MPEG-2 data-stream... They are 4-digit integers represented in 13 bits. So max 8191 values. 0000 is reserved to identify the Program Association Table (PAT). 8191 is reserved to identify "null packets" used as fillers to keep the transport bit stream constant (I assume these are not used in the case of the less comon variable bit-rate streams). So the actual real values in use for audio, video, etc. are 0001-8190.

APID = Audio PID (a tag on the packet which says "this packet encodes some audio data")

VPID = Video PID (likewise for the video info)

PCRPID = Program Clock Reference PID (used to synchronise video & audio. Not commonly used).

TEXTID = the "PID" of any teletext info

SID = the "PID" of the SI (DVB service Information) packets. Such packets are used for various purposes like program info. Actually, SI specifies all the data required by the receiver to demultiplex and decode the programs and services in the
transport stream. Values x'0000' through x'0014'. It is metadata (data about data).

NID = the "PID" of the NIT (Network Information Table). The NIT typically contains pointers to other related transmitted data for the same network/provider.

There are a few common conventions with respect to PIDs but don't be misled. For example - the PCRPID is rarely used and defaults to the value of the VPID. The Audio PID is usually 1 higher than the VPID. Etc. But this is just convention. In theory they could all have arbitary values.
Often a provider retains the same frequency, symbol-rate, polarisation and FEC, but for reasons known only to themselves changes one or more PIDs. This will result in the receiver losing sound, picture - whatever - for the changed channel. The solution is to rescan the transponder, since the scan funciton usually pulls in the PIDs from the data stream. Some providers do not transmit the PIDs in the data stream. What I mean is; the PIDs are of course used (each packet has its appropriate identifier) but that the actual PID values are not provided in the data stream. So the scan functions will not be able to determine the PIDs and will not "find" the channel. An example of such a channel was the now defunct Irish "Tara" on Intel601. For those, one must manually create the channel in the reciever's settings - i.e. the receiver must have a "manual PID entry" function. Manaul PID entry can also be used to correct channels whose PIDs had changed thus avoiding a total re-scan of the transponder.

FEC = Forward Error Correction. It is the proportion (fraction) of data which is used for the actual data. The rest is redundant, used only for correcting any transmission errors. The lowest value is 1/2 which means that fully half of the data is there for error-correction purposes only. The highest value is 7/8 which measn that seven-eighths (nearly 84%) of the data is "real". IE only 16% is for error correction...
Possible values are 1/2, 2/3, 3/4, 4/5, 6/7. 7/8. If you see an "auto" function for this setting it means that your receiver is capable of automatically detecting the FEC in use.

Symbol-rate is a measure of the amount of data bits sent per unit time. The unit is Megasymbols per second... NOT megabits, but megasymbols. Actually, a "symbol" is equivalent to 2 bits. So for example, the commonly used symbolrate of "27.500" is actually 55 Mb/s. The lowest symbol rates used by satellite tv transmissions are around 1.000, below the threshold of most consumer receivers. The highest values are around 35.000. The Echostar receivers can process signals with symbol rates ranging from about 1.500 to 40.000 (formally, the lower limit is 2.000 but in practise they go lower).

This is a feature of the microwave wave-form.
Polarity can be linear or circular. Linear polarity is "Horizontal" or "Vertical". Circular is "Left" or "Right".
In fact, satellite receivers only ever work in horizontal or vertical settings. When you tune to a crircularly polarised signal, you must therefore choose either H or V. No receivers have L or R settings. But it works becuase the circularly polarized signals have both V and H components (ever studied vectors?). So tuning to "H" pulls out the H vector/component of the L or R signal, although with a 3db loss.
You can get dialectric plates for your LNB which "convert" the circularly polarised signals to linear signals, but they then cause 3db loss on the linear signals. It's one or the other... (see the separate FAQ I wrote).

The frequency range of signals is split arbitrarily into various bands. The C-band is the lower end (3.7 to 4.1 GHz) with the Ku-band higher (10.7 to 18 GHz, although only used to 12.75 in practise). Still higher is the Ka-band (18.3 to 1.2 GHz) used for data transmissions, not for TV signals.
Higher frequencies can be more tightly focussed than lower, so that the signal at a given location in its footprint is stronger and the dish required is smaller. So Ku-band dishes tend to be smaller than C-band dishes. The huge dishes one sees in America or Asia are usually that big because they have to be - those guys use a lot of C-band.

Hope this helps
See the following sites for more info:

http://www.dutchsat.com/begrippenindex.html (Dutch language)
http://www.broadcastpapers.com/sigdis/TektronixMPEGGuide - glossary.htm