Digital TV uses the scanning concept to paint a picture on the CRT. So you can continue to think of the Digital TV screen in terms of scan lines, as you would think of the standard NTSC analog screen. However, you should also view the Digital TV screen as made up of thousands of tiny dots of light called pixels. Each pixel can be any of 256 colors. These pixels can be used to create any image. The greater the number of pixels on the screen, the greater the resolution and the finer the detail that can be represented. Each horizontal scan line is divided up into hundreds of pixels. The format of a Digital TV screen is described in terms of the numbers of pixels per horizontal line by the number of vertical pixels (which is the same as the number of horizontal scan lines).

One major difference between conventional NTSC analog TV and Digital TV is that Digital TV can use progressive line scanning rather than interlaced scanning. In progressive scanning each line is scanned one at a time from top to bottom. Since this format is compatible with computer video monitors, it is possible to display Digital TV on computer screens. Interlaced scanning can be used on one of the HDigital TV formats.

The 480p (the p stands for "progressive") standard offers perfon-nance comparable to that of the NTSC system. It, use.s a 4:3 aspect ratio for the screen. The scanning is progressive. The vertical scan rate is selectable to fit the type of video being transmitted. This format is fully compatible with modern VGA computer monitors.

The 720p format uses a larger aspect ratio of 16:9 (a 4:3 format is optional at this resolution also). This format is better for showing movies. Figure 1 shows the difference between the current and new Digital TV aspect ratios. The 1080i format uses the 16:9 aspect ratio but with more scan lines and more pixels per line. This format obviously gives the best resolution. The Digital TV set should be able to de­tect and receive any available format.

Digital TV Transmission Concepts

In Digital TV both the video and the audio signals must be digitized by A/D converters and transmitted serially to the receiver. Because of the very high frequency of video signals, special techniques must be used to transmit the video signal over a standard 6-MHz bandwidth TV channel. And because both video and audio must be transmitted over the same channel, multiplexing techniques must be used. The FCC's requirement is that all of this information be transmitted reliably over the standard 6-MHz TV channels now defined for NTSC TV.

Assume that the video to be transmitted con­tains frequencies up to 4.2 MHz. For this signal to be digitized, it must be sampled at least two times per cycle or at a minimum sampling rate of 8.4 MHz. If each sample is translated into an 8-bit word (byte) and the bytes trans­mitted serially, the data stream would have a rate of 8 X 8.4 MHz or 67.2 MHz. This does not include the audio, which must also be digitized and multiplexed with the digital data. The resulting digital data rate is very high. To permit this quantity of data to be transmitted over the 6-MHz channel, special encoding and .nodulation techniques are used.

Digital TV Transmitter

Figure 2 shows a block diagram of a Digital TV transmitter. The video from the camera consists of the R, G, and B signals that are converted into the luminance and chrominance signals. These are digitized by A/D converters. The luminance sampling rate is 14.3 MHz, and the chroma sampling rate is 7.15 MHz. The resulting signals are serialized and sent to a data compressor. The purpose of this device is to reduce the number of bits needed to represent the video data and therefore permit higher transmission rates in a limited-bandwidth channel. The data compression method used in Digital TV is referred to as MPEG2, which stands for the "Motion Picture Engineering Group-2" standard. This organization establishes standards for the compression of video and film pictures. The MPEG2 data compressor processes the data according to an algorithm that effectively reduces any redundancy in the video signal. For example, if the picture is one-half light blue sky, the pixel values will be the same for many lines. All of this data can be reduced to one pixel value transmitted for a known number of times. The algorithm also uses fewer bits to encode the color than to encode the brightness because the human eye is much more sensitive to brightness than to color. The MPEG2 encoder captures and compares successive frames of video and compares them to detect the redundancy so that only differences between successive frames are transmitted.

The signal is next sent to a data randomizer. The randomizer scrambles or randomizes the signal. This is done to ensure that randorn data is transmitted even when no video is present or when the video is a constant value for many scan lines. This permits clock recovery at the receiver.

Next, the random serial signal is passed through a Reed-Solomon (RS) error detection and correction circuit. This circuit adds extra bits to the data stream so that transmission errors can be detected at the receiver and corrected. This ensures high reliability in signal transmission even under severe noise conditions. In Digital TV, the RS encoder adds 20-parity bytes per block of data that can provide correction for up to 10 byte errors per block.

The signal is next fed to a trellis encoder. This circuit further modifies the data to permit error correction at the receiver. Trellis encoding is widely used in modems. Trellis coding is not used in the cable TV version of Digital TV. The audio portion of the Digital TV signal is also digital. It provides for compact disk (CD) quality audio. The audio system can accommodate up to six audio channels, permitting mono­phonic sound, stereo, and multichannel surround sound. The channel arrangement is flexible to permit different systems. For example, one channel could be used for a second language transmission or closed captioning.