We all know perfectly well that the world of technology around is digital, or strives for it. Digital broadcasting is far from new, but if you have not been specifically interested in it, the technologies inherent in it may be unexpected for you.
Contents of the article series
- Part 4: Digital component of the signal
- Part 5: Coaxial Distribution Network
- Part 6: RF Amplifiers
- Part 7: Optical Receivers
- Part 8: Optical backbone network
- Part 9: Headend
- Part 10: Troubleshooting on the cable TV network
Composition of a digital television signal
A digital television signal is a transport stream of different versions of MPEG (sometimes other codecs), transmitted by a radio signal using QAM of varying degrees. For any signalman, these words should be clear as day, so I will give only a gif from , which, I hope, will give an understanding of what it is for those who are simply not interested yet:
Such modulation in one form or another is used not only for "teleanachronism", but also for all data transmission systems that are at the peak of technology. The speed of the digital stream in the "antenna" cable is hundreds of megabits!
Digital signal parameters
Using the device Deviser DS2400T in the mode of displaying the parameters of a digital signal, we can see how it really happens:
In our network, there are signals of three standards at once: these are DVB-T, DVB-T2 and DVB-C. Let's consider them in turn.
DVB-T
This standard did not become the main one in our country, giving way to the second version, but it is quite suitable for use by the operator for the reason that DVB-T2 receivers are backward compatible with the first generation standard, which means that the subscriber can receive such a signal on almost any digital TV without additional attachments. In addition, the standard intended for transmission over the air (the letter T stands for Terrestrial, ether) has such good noise immunity and redundancy that it sometimes works where, for some reason, an analog signal does not creep through.
On the screen of the device, we can observe how the 64QAM constellation is being built (the standard supports QPSK, 16QAM, 64QAM). It can be seen that in real conditions the points do not add up at all into one, but come with some expansion. This is normal as long as the decoder can determine which square the arrived point belongs to, but even in the above image, areas are visible where they are located on the border or close to it. From this picture, you can quickly "by eye" determine the quality of the signal: if the amplifier does not work well, for example, the dots are arranged randomly, and the TV cannot assemble a picture from the received data: it "pixelates", or even completely freezes. There are times when the amplifier processor "forgets" to add one of the components (amplitude or phase) to the signal. In such cases, on the screen of the device, you can see a circle or a ring the size of the entire field. Two points outside the main field are reference points for the receiver and carry no information.
On the left side of the screen, under the channel number, we see quantitative parameters:
Signal level (P) in the same dBμV as for the analogue, however, for a digital signal, GOST already regulates only 50dBμV at the input to the receiver. That is, in areas with greater attenuation, the “digit” will work better than the analogue.
The value of modulation errors (MER) shows how distorted the signal we receive, that is, how far the arriving point can be from the center of the square. This parameter is similar to the signal-to-noise ratio from an analog system, the normal value for 64QAM is from 28dB. It is clearly seen here that significant deviations in the above image correspond to a quality above the norm: this is the noise immunity of the digital signal.
The number of errors in the received signal (CBER) is the number of errors in the signal before processing by any correction algorithms.
The number of errors after the operation of the Viterbi decoder (VBER extension) is the result of the work of the decoder, which uses redundant information to recover errors in the signal. Both of these parameters are measured in "pieces per number taken." In order for the device to show the number of errors less than one in a hundred thousand or ten million (as in the above image), it needs to accept these ten million bits, which takes some time on one channel, so the measurement result does not appear immediately, and may even be bad at first (E -03, for example), but after a couple of seconds, get to a great setting.
DVB-T2
The digital broadcasting standard adopted in Russia can also be transmitted via cable. The shape of the constellation at first glance may be somewhat surprising:
Such a rotation further increases noise immunity, since the receiver knows that the constellation should be rotated by a given angle, which means that it is possible to filter what comes without a built-in shift. It can be seen here that for this standard, the bit error rates are an order of magnitude higher and the errors in the signal before processing no longer go beyond the measurement limit, but are quite real 8,6 per million. A decoder is used to correct them. LDPC, so the parameter is called LBER.
Due to the increased noise immunity, this standard supports the modulation level of 256QAM, but at the moment only 64QAM is used in broadcasting.
DVB-C
This standard was originally created for transmission over cable (C - Cable) - the environment is much more stable than air, therefore it allows you to use a higher degree of modulation than DVB-T, and therefore transmit more information without using complex coding.
Here we see the constellation 256QAM. There are more squares, their size has become smaller. The probability of error has increased, which means that a more reliable medium (or more complex coding, as in DVB-T2) is needed to transmit such a signal. Such a signal can “scatter” where analogue and DVB-T / T2 work, however, it also has a margin of noise immunity and error correction algorithms.
Due to the greater error probability, the MER parameter for 256-QAM is already normalized to 32dB.
The counter of erroneous bits has risen an order of magnitude and already calculates one erroneous bit per billion, but even if there are hundreds of millions (PRE-BER ~ E-07-8), the Reed-Solomon decoder used in this standard will eliminate all errors.
Source: habr.com