UNAOHM
EP2500 (Rev. 0)
53
The post Viterbi BER (pV BER) is a measurement that is taken on the far side of the correction stage
where a signal is already less severely errored that is, where it yields readings approximately from one
to one million times smaller than the Channel BER (CH BER). Because fewer errors occur at the point
where pV BER measurement is taken, its results can be less accurate than what the CH BER results can
be, given the same impaired signal. In fact, statistically the pV BER has a much more difficult
measuring task to perform than the CH BER which, thanks to its position on the measuring path, is
intrinsically more accurate for the same job. To all practical purposes, therefore, the CH BER is highly
recommended.
If, on the other hand, the value of the pV BER is, for example, 1E-12, this means that it is necessary for
the measuring system to wait for at least 1000 billion bits to pass through, to be able to measure it in a
statistically effective way.. This is equivalent to a waiting time of about a few days.
The DVB standard dictates that when a pV BER value is better than the threshold at 1E-4 (one bit in
error over 10000 bits total) the system can be considered good (at delivering its payable load of data)
and that the 1E-4 threshold can be considered Quasi Error Free. For comparison purposes, therefore, the
pV BER is highly recommended.
However:
especially in the case of communal installations ( the most critical for the QPSK Sat signal), the
pV BER may not be used because the values supplied for this parameter do not cover the
measurement range wanted and do not, therefore, provide information useful to establish whether
or not the installation is correct as far as digital signal reception is concerned.
In fact, to guarantee, in the worst hypothesis, that the pV BER is at least 1E-4, from the LNB
output (or headend) it should measure a pV BER of at least 1E-12.
However, for the above-mentioned reasons, it is impossible to measure such a value.
Fortunately, a much more significant indication is offered by the CH BER which:
• Is immediately measurable with all the signals present in the installation (from best to bad).
• Always supplies an indication of the margin of the reception threshold (obviously, if the
installation allows for an even minimal reception).
• Is a very sensitive index regarding ambient and meteorological conditions (temperature,
humidity, cloudiness, cable efficiency, amplifiers, etc.).
The diagram in Figure 27 supplies, instead, a meaningful comparison between the “CH BER”, and the
“pV BER” over publications or diagrams dealing with digital QPSK transmissions if such parameter is
not directly measurable by the instrument.
For example, the line shows that for a “CH BER” of 1E-3 (10
-3
or 1 out of 1000) the corresponding “pV
BER” lies between 1E-8 and 1E-11 approximately, depending on the programmed “CODE RATE”.
Technical literature supplies, in such diagrams, the “X” axis indicated as Eb/No which is the ratio
between the energy per bit and the noise per unit of bandwidth (Hz). The Eb/No is easily calculable,
once the C/N ratio is known, by means of the following formula :
Eb/No (in dB) = C/N – 10-log[(Rs k)/Bw] where:
Eb = Energy per bit.
No = Noise energy in a 1 Hz band.
C/N = Carrier/noise ratio measured in dB.
Rs = Symbol rate in M symb/sec.
Bw = Signal band.
k = Number of bits transmitted per symbol (2 for QPSK, 4 for QAM 16, 6 for QAM 64,
7 for QAM 128, 8 for QAM 256).
Example:
To Next Page