Journal of the Society of Motion Picture and Television Engineers (1950-1954)

even harmonics of one-half the line frequency. Row B shows a color-difference signal as generated, again with solidly drawn components representing even harmonics of one-half the line frequency. Row C shows this same signal appearing as modulation on a subcarrier fte. The spacings of the components are still the same but the subcarrier itself being an odd harmonic, all of the other components must themselves be odd harmonics also. Finally, Row D of the drawings shows the solid lines of the first section and the dotted lines of the third section interleaved as they are in a normal transmission. Some other characteristics of the signal may be noted by further reference to Figure 6. For example, since the red signal and the signal EY' are equal on white, the output of the adder in the red channel, the red color-difference signal, must equal zero on white. It will have an amplitude which we may call positive when the scene is red and an amplitude which we may call negative when the scene is the complement of red, or cyan. Now, if the modulator of Fig. 6 is a balanced modulator, it can be arranged so that its output will also vanish on white; it is advantageous to design the system in this fashion, since this reduces still further the residual perceptibility of the low-visibility components represented by the dotted lines of Figs. 4 and 7, especially in picture highlights which are generally white or relatively unsaturated color. We find, however, that a transmission whose output vanishes when the modulating signal vanishes, is a transmission in which the carrier frequency has been suppressed, leaving only the modulation sidebands. For correct detection of such a transmission the carrier must be resupplied at the detector of the receiver, and we shall provide a synchronizing pulse in the transmission to enable receivers to generate a suitable carrier. The resupplying of a properly synchronized carrier in the receiver offers us another advantage: it permits us to distinguish successfully between phase modulation and amplitude modulation of a single subcarrier; alternatively, we may say that it permits us to modulate two subcarriers at the same frequency 90° apart in phase and transmit their sidebands over the same circuit and yet distinguish each set of sidebands from the other in the receiver. It will be seen that we make use of exactly this property to permit the transmission of the blue color-difference signal as modulation sidebands on the very same subcarrier frequency as the red colordifference signal, with the two subcarriers differing merely by 90° in phase at the subcarrier frequency. Finally, the two sets of subcarrier sidebands representing respectively red colordifference information and blue colordifference information, are combined and are then added to the luminance signal as illustrated frequency-wise in Fig. 7. The resulting signal is then applied to the transmitter. Let us note again that this signal is in all respects a perfectly normal monochrome television signal to which there has been added, in a fashion which makes it essentially invisible on normal monochrome receivers, the color-difference information which can be used in a color television receiver to reconstruct the original image in full color. The transmission of two independent sets of modulation sidebands based on the same subcarrier frequency over the same circuit requires that the transmission be on a double sideband basis if the two sets of modulation components are to be separated at the receiver. Consequently, the maximum frequency which we may choose for the subcarrier is a frequency enough lower than the top frequency of the expected passband of the system to satisfy this requirement for double sideband transmission. Now, if one of the modulation components has a bandwidth of a half megacycle, while the other had a bandwidth of a megacycle or greater, the necessity for double 330 April 1953 Journal of the SMPTE Vol. 60