Cisco Prisma II Platform 기술 참조
2
Recommended Levels of Digital (QAM) Signals Relative to Analog Video Levels 4000333 Rev C
Assumptions
Assumption 1: Amplitude Distributions of Analog and Digital Signals
The recommended QAM level is determined by first assuming that it is desirable to
set it equal to the average power of an analog video signal. However, even when this
is done, the two signals will have significantly different amplitude distributions; that
is to say, the probability of measuring any specified amplitude at any given instant
of time will be markedly different for the two signals. (Viewing an analog video
signal and a digital QAM signal alternately on an oscilloscope will highlight these
differences. The analog signal has a predictable, repetitive element due to the line
and field blanking pulses, whereas the digital signal is essentially random). In
statistical terms, it is said that the Probability Density Functions (pdf) of the two
signals are dissimilar, with the digital signal having an almost Gaussian shape.
However, when a sufficiently large number of independent signals are combined,
However, when a sufficiently large number of independent signals are combined,
the amplitude distribution of the overall signal tends toward a Gaussian
distribution. This can be proved by the Central Limit Theorem, although an
exhaustive discussion of the mathematics is beyond the scope of this bulletin. (For
further information, the attention of the reader is drawn to the References at the
conclusion of this document). The practical consequence of this is that the 'peakiness'
of the two signal types tends to become more and more similar, as the number of the
signals is increased, and therefore the effect on optical transmitters, which might
respond to signal peaks by 'clipping', will be the same. In this context, a 'sufficiently
large' number of analog video signals are taken to be approximately 20.
Assumption 2: Difference Between Peak Envelope Power and Average Power of an Analog Video
Signal
The peak envelope power of an analog video signal is equal to the power of the
unmodulated RF carrier, because, in the interval of the horizontal synchronising
pulse, the modulation depth is zero. The reduction in power when the carrier is
modulated will be determined by the maximum modulation depth, corresponding to
peak white level, and the average picture level (APL).
For a single signal, the APL will vary with scene changes and may have a
For a single signal, the APL will vary with scene changes and may have a
significantly different value during different types of programming, but when a
number of signals are combined, the APL will not be subject to such a large
variation. A suitable value for the APL must therefore be selected, and the figure of
50% has been chosen for this analysis. This corresponds to an average picture
luminosity mid-way between black and white, and seems to be a reasonable
assumption when, again, a 'sufficiently large' number of signals is considered.
The calculations that follow are therefore largely concerned with the determination
The calculations that follow are therefore largely concerned with the determination
of the average power of an analog video signal having an APL of 50%.
Continued on next page