Ntsc and Beyond

166 IEEE Transactions on Consumer Electronics, Vol. 34, No. 1, FEBRUARY 1988 NTSC AND BEYOND Yves Charles Faroudja, President Faroudja Laboratories, Inc. Sunnyvale, California ABSTRACT tant details of this proposal for a SuperNTSC system that goes well beyond current NTSC per- Two recent technological developments of consi- formance characteristics. derable significance have led to the critical reevaluation of the NTSC standard as an adequate public television service. These two developments were: (a) the rapid proliferation of a INTRODUCTION variety of HDTV proposals, and (b) the emer- gence of low-cost frame memories for home The major technical impact of the frequent de- receivers. This critical reevaluation, which the monstrations of various high definition systems author conducted in the course of his own re- at television conferences and equipment shows, search work, led him to the unavoidable conclu- all over the world, has been to motivate research sion that NTSC is indeed capable of rivalling and development engineers to more critically re- HDTV image quality under certain conditions, examine the existing color television systems and these are: now serving the public. In the NTSC areas of the world, where over 200 million home receivers 1. That today's much better understanding of represent almost 60% of the global color receiver the fundamental principles of color tele- population, the widespread adoption of an in- vision be put to optimal use, and compatible high definition television system is inconceivable, and fully precluded by social con- 2. That the advanced technology tools deve- straints. loped primarily for HDTV systems be applied in an equally innovative manner to That fact alone has led to a variety of research the NTSC environment. efforts to explore the ways in which the basic NTSC system can be improved to its maximum The proposal being made by the author is that potential, particularly when some of the new this improved NTSC which he calls SuperNTSC technology developed for HDTV has a beneficial (Trademark pending) can fulfill the need for a effect when it is tailored to NTSC requirements. superior television system that does not require extra bandwidth for transmission, a new subcar- NTSC has indeed been both a commercial and rier for additional information, or any new technical success when one looks at it in a his- signals "buried" or interleaved in the existing torical perspective. It has, and continues to NTSC spectrum. What is even more important, provide the least costly color receivers for the this system is fully forward and reverse com- largest number of users, and that with the the patible with NTSC today and tomorrow. smallest transmission spectrum (6 MHz) of all existing color systems. The author also recommends that other HDTV proposals that do use wider transmission chan- While a number of new techniques have sprung nels and/or extra subcarriers be considered for up for handling TV signals within the studio or broadcasting only after all of the potential production center (analog component, digital, improvement options for a compatible NTSC etc.), the final product that goes on air or over have been fully examined and exploited, espe- cable is still the classical encoded NTSC. Con- cially those that take into account emerging tinued international adherence to a set of well "smart" receivers with built-in flexible archi- defined rules about the NTSC signal, in a man- tecture which can greatly enhance the home dis- ner that maintains full reverse compatibility play side of the NTSC of the future. with the existing NTSC receivers, should be the first goal of any future system dedicated to deli- This paper deals with some of the more impor- vering better pictures into the home. Contributed Paper 0098 3063/88/0200 0166$01.00 0 1988 IEEE Manuscript received December 8, 1987. Authorized licensed use limited to: University of Washington Libraries. Downloaded on September 26,2020 at 03:00:15 UTC from IEEE Xplore. Restrictions apply. Faroudja: NTSC and Beyond 167 STRENGTHS & WEAKNESSES OF THE NTSC STANDARD The basic NTSC standard that we have been using for more than three decades has three LUMINANCE I fundamental advantages: I SUBCARRIER 7 \ 1. There is a vast distribution network already in place covering 32 countries with over 200 million receivers. 2. It uses the least of our most precious I resource, spectrum space, needing only a 6 Figure 1 Typical quadrature modulaled color television specburn (NTSC). MHz channel. 3. The basic benefits of the NTSC encoding process are almost self-evident. It makes the most efficient use of the available spectrum through the interleaving of the luminance and chrominance signals (Figs. 1 and 2). It was, and continues to be, fully compatible with monochrome television. It is the simplest of the three color television systems (NTSC, PAL and SECAM) in use today, because it does nFh I (n+i)Fh (n+Z)Fh (n+S)Fh (n+h)Fh (n+5)Fh not use phase-line alternation or multiple sub- 2ntl Fh carriers. This inherent simplicity makes it the 7 most cost-effective system in use as well. Hun- Figure 2 Chroma/lumrnance Speclrum In the vrcmfly 01 lhe subcamer dreds of millions of viewers, who live in those countries where NTSC is the national color standard, already receive acceptable quality color images in their homes via NTSC trans- 1. LINE STRUCTURE VISIBILITY & missions. RESOLUTION However, the present NTSC system also has 3 The inherent weakness of a 525 line, 21 inter- basic weaknesses which must be corrected, if an lace television system is the line structure visi- image quality comparable to HDTV is to be bility. Many other researchers have proposed achieved. These deficiencies are: means to alleviate this problem, while improving the vertical resolution as well, and these propo- 1. Monochrome limitations, visible line struc- sals can be found in the literature (Ref. 11, 17, ture (525 lines, 2:l interlace), and poor 21, 22). vertical and horizontal resolutions, The combination of some of the processes listed 2. NTSC encoding limitations, especially those below can lead to a practical compromise where relating to intermodulation between line structure and vertical aliases are no longer luminance and chrominance, detectable. These are: 3. Gamma problems that usurp the constant 1. 525 lines, 30 Hz progressive scan at the luminance principle for perfect image camera, rendition. 2. 30 Hz progressive scan to 2:l interlace, 60 The techniques described in this article greatly Hz conversion, reduce the impact of these problems, to the ex- tent that the final NTSC image will appear on 3. Use of a frame store in the receiver to an improved home-viewer set as a full band- transcode a 525 line, 2:l interlaced image width, 1050 lines RGB image, emulating an into a 60 Hz display with 1050 lines and 2:l HDTV display. interlace. Authorized licensed use limited to: University of Washington Libraries. Downloaded on September 26,2020 at 03:00:15 UTC from IEEE Xplore. Restrictions apply. 168 IEEE Transactions on Consumer Electronics, Vol. 34, No. 1, FEBRUARY 1988 If progressive scan is used in the camera, verti- chrominance components also imposes no burden cal interpolation is made easier in the line of spurious contamination between these two doubler, and an apparent vertical resolution of channels, and, in fact, if the signals are kept in the order of 440 lines is observed without visible analog component form for distribution and artifacts (ragged or "stepped" diagonal transi- dematrixed for display, the results would look tion). The subjective results are close to those virtually identical to the RGB source. obtained with HDTV systems. However, to accommodate the need for a single- In the horizontal domain a frequency response channel transmission path, conventional NTSC of 4.2 MHz is certainly not satisfactory. The makes use of the band-sharing principle in subjective sharpness, however, may be signif i- which the luminance and chrominance compo- cantly improved by the combination of 2 tech- nents are multiplexed or encoded into a single niques: detail processing in the encoder lumi- signal, and where the separation of the two nance path, and spectrum expansion in the principal components is maintained by the fre- decoder. quency interleaving process (Fig. 2). The NTSC composite signal, which emerges from a standard Detail Processing encoder (Fig. 3), has three distinct deficiencies in descending order of importance, as follows: If small detail levels are increased (Ref. 25) without modifying large transitions, the broad- 1. Cross color cast information will appear to be of a wider bandwidth than 4.2 MHz, even though measure- 2. Cross luminance ments at 0 to 100°/o transitional levels will not exhibit an increase in resolution. 3. Limited chroma bandwidth Bandwidth ExDansion If all signal transitions in the image were either horizontal or vertical, the interleaving process In the receiver, it is desirable to shorten the rise would adequately avoid any unwanted interfer- time of large horizontal transitions without in- ence between the two components. Vertical- troducing pre-shoots and overshoots in order to domain transitions have no high-f requency lumi- simulate a wider bandwidth. This is accomplish- nance information in them, so they do not inter- ed through multiplicative enhancement (Ref. 4). fere with the chrominance spectrum. Horizon- tal-domain transitions can be easily separated on The combined result of these 2 techniques leads a monitor with a comb filter, because the lumi- to an apparent bandwidth increase in a ratio of nance information is at even multiples of half 1.8, as has been proven by numerous subjective the line frequency, while the chrominance tests. This is equivalent to a 7.5 MHz bandwidth at 525 lines, or 15 MHz at 1050 lines, and is not too far from results obtained with wide- band HDTV systems, particularly if one takes into consideration the different aspect ratios involved.

Load more