Videosignal in depth

A brief history of the video signal is required to understand today’s waveforms.

When television was originally developed the signal chain was designed to meet the constraints of several of the mechanisms used in the broadcast chain: the receiver technology and the transmission systems. Despite television production evolving, some of the basic concepts and constrains still apply.

A brief history of the video signal is required to understand today’s waveforms.

To capture and display an image, the light from an object is first passed through a lens and lands on a camera sensor. The sensor output produces a single varying signal which is in proportion to the light received. This signal can then be displayed on a simple screen and produce a grey scaled monochrome picture.

To broadcast colour, obviously, colour information must be obtained and transmitted.
If the camera takes the object information and split the light onto three separate sensors having colour filters, three brightness signals are obtained from separate parts of the spectrum. The signals can be transmitted and reproduced on a screen. With suitable colour filter choice and screen colour reproduction, the original captured image can be accurately displayed.

The three chosen colours are in the Red, Green and Blue areas of the spectrum and the signal level corresponds to the light intensity (brightness) of that colour. Mixing the RGB signals in relative proportions to how the eye perceives the brightness of colours, the original black and white signal can be produced.

When the original black and white transmission system was built, little thought was ever given to future changes such as colour. Colour transmission had to fit within the existing transmission system without causing issues, including affecting the domestic reception of black and white television.

As stated above, by carefully mixing the RGB signals together, a combined brightness signal can be generated for the black and white system, Y signal.

Y = 0.299 R + 0.587G + 0.114B

As the colour signals need to be transmitted (and it would be foolish to send three other signals as well as black and white) a mathematical solution was sought. As the Y signal is made up of a ratio of the R, G and B signals, by manipulation, only two other signals need to be sent.

There were three options (B-Y), (G-Y) and (R-Y). (B-Y) and (R-Y) were chosen for least potential received errors in transmission. These two signals were called the colour difference signals. So the three signals combined could serve black and white and colour delivery.

However this still resulted in three signals being transmitted to the home. A technical trick was created to encode the (B-Y) and (R-Y) signals onto one signal. The two signals were modulated onto a carrier signal through a process called “phase modulation”. This signal was the Colour Sub Carrier(CSC). The resulting CSC signal could be added onto the Y signal and all the information could be transmitted at once. To reference the CSC phase signal, an extra burst of known phase was added to the main Y signal in the image blanking periods of each line of the image.

Various tweaks and adjustments had to be made to make the signals fit.
To ensure that the combined signal did not damage transmitters, the values of B-Y and R-Y were scaled and the three signals became:

  • Y = 0.299 R + 0.587G + 0.114B
  • U = 0.493 (B-Y) or – 0.147R – 0.289G + 0.436B
  • V = 0.877 (R-Y) or    0.615R – 0.515G – 0.100B

The frequency bandwidth of the U and V signals was also limited compared to the Y Luminance. The eye fortunately does not see colour in as much detail as brightness and this enabled the U and V signals to be frequency restricted 1.3Mhz compared to the Y at 5Mhz for PAL transmissions. The frequency of the CSC signal was also chosen with care.

Its frequency was chosen so that it didn’t produce fixed patterns of interference of solid colour when viewed on B & W receivers and that information did not cause transmitter interference between the audio and video modulations.

Different Video Signal Types

RGB signals

RGB signals are full bandwidth and are device dependant signals, when working with cameras and vision mixers it is the configuration of the device to interpret the signal that is vital. Widespread movement of these signals is restricted due to the bandwidth constraints of moving these signals.

YUV Signals

The signals are derived from RGB. They are processed (frequency limited) and scaled to fit the PAL transmission system.

Ey,Ecr,Ecb / Y,Cr,Cb Signals

As systems progressed and technology improved it became sensible to keep the three signals separate through the production process. Although good, the CSC signal did cause post production issues with colour information affecting luminance and vice versa.

  • E’Y = 0.299 R + 0.587G + 0.114B
  • E'CR = 0.713 (E'R - E'Y) or 0.500 E'R - 0.419 E'G - 0.081 E'B
  • E'CB = 0.564 (E'B - E'Y) or - 0.169 E'R - 0.331 E’G + 0.500 E'B

The standard for Europe is defined in EBU N-10

This was used in recording formats such as Betacam SP.
The bandwidth for internal working was improved and local equipment had restrictions of 
Luminance: 5.75 MHz
chrominance signals: 2.75 MHz

Digital Component

It is important to note that the digital component standard ITU-R BT.601 was already specified at the time of the definition of the analogue components.

The specifications for the analogue component signals have the same definition for the bandwidths of the primary luminance and chrominance signals as for the digital components.

  • E’Y = 0.299 R + 0.587G + 0.114B
  • E'CR = 0.713 (E'R - E'Y) or 0.500 E'R - 0.419 E'G - 0.081 E'B
  • E'CB = 0.564 (E'B - E'Y) or - 0.169 E'R - 0.331 E’G + 0.500 E'B

Video signals for HDTV

An explanation of the differing HD resolutions and formats is provided elsewhere in the app. In principle, the relationship between the RGB signals and the Y, CR, CB signals is the same as the above-mentioned digital video signals for SDTV.
The following is a brief overview of the four HD formats relevant for Europe, as specified and listed in EBU Tech 3299:

  • System 1 (S1): 720p / 50
  • System 2 (S2): 1080i / 25
  • System 3 (S3): 1080p / 25
  • system 4 (S4): 1080p / 50

Thinking about the future, it would be best to transmit video signals with the greatest possible resolution (e.g., 8K), very high sampling rate (e.g., 60p or more), and maximum quantization (e.g., 14 bits). This would provide the viewer with a very high quality viewing experience and significantly improve the processing of the video signals, especially with effects.

The storage of such immense image signals today is difficult, but feasible, however the processing, and transmission costs would be substantial. Today we are looking at 12Gbit/s for some raw pictures and this will only increase with higher quality, and higher resolution formats.

In cinemas, a maximum of 250Mbit/s is played out, delivery to TV transmitters for HD is at a maximum of 100 Mbit / s and the transmission itself usually less than 10Mbit/s. With further development of codecs, it will be possible to increase the resolution and the sampling rate even further without having to significantly increase the data rate to be transmitted. Domestic Internet connection speeds are increasing and this will allow better user experience. It is highly likely that HD-SDI as we know it will disappear and it will be a series of compressed files and live IP streams.