Spectrum Compatible-HDTV Data Transmission System

This invention relates generally to video signal transimission systems and specifically to a low power data transmission system for use in conjunction with a HDTV system.

The spectrum compatible high definition television (HDTV) system recently proposed by Zenith Electronics Corporation has the capability of transmitting television signals without objectionable interference with NTSC co-channels and adjacent channels (the so-called "Taboo" channels). This is accomplished in part by minimizing the power of the transmitted signals by removing the low frequency portion of the video signal, which requires high transmission power, developing a digital representation thereof and transmittig the digital data with the high frequency portion of the video signal. The digital data is transmitted during the vertical blanking interval (VBI) of the television signal. This approach, when correlated with "frame locking" to the co-channel or adjacent channel signal helps by "hiding" the data in the VBI of the other signal. Thus any interference caused will be much less visible.

Major considerations in transmitting digital data as part of a television signal include the bandwidth required for its transmission and the potential of interference into co-channels and adjacent channels. Therefore the data in the VBI is preferably transmitted at as low a power level as possible, consistent with good recovery, to maintain the noninterfering characteristics of the transmission system. Additionally, it is desirable to use data compression techniques to minimize the required transmission bandwidth for the data.

A principal object of the invention is to provide a novel transmission system and method for a video signal.

Another object of the invention is to provide a system for transmitting data in a low power, low bandwidth format for use in a television transmission system.

A further object of the invention is to provide an improved HDTV television system.

These and other objects and advantages of the invention will be apparent upon reading the following description in conjuction with the drawings, in which:

FIG. 1 is a partial block diagram of a transmission system constructed in accordance with the invention;

FIG. 2 is a corresponding block diagram of a compatible receiver;

FIG. 3 is a diagram indicating a 12-bit packet of a series of three samples of data in the form of delta modulation comonents; and

FIG. 4 is a depiction of the constellation states of a quadrature modulator arranged for use with the inventive system.

In FIG. 1, a source of video information 10, which may comprises a wide band signals of about 30 MHz, is coupled to a low frequency removal circuit 12 where the low frequency video information in the signal is subtracted, preferably on a line-by-line basis. The removal low frequency information is sampled and digitized in a sampling and adaptive delta modulator circuit 16, which may be of the type disclosed in Ser. No. 453,525, filed Dec. 20, 1989. The delta modulated slope and amplitude components are applied to a reconstruction filter 15, the output of which is substracted from the video signal from video source 10 in a summation circuit 13. This process assures that the reconstructed low frequency signal added in the receiver has the same resolution as that taken out at the transmitter. The remaining high frequency video signal in the output of summation circuit 13 is supplied to a video compressor 18 where encoding of the high frequency portion is accomplished to enable transmission of the 30 MHz wide video signal in a 6 MHz bandwidth. The low frequency removal and video comprssion techniques are described and claimed in copending applications Ser. Nos. 238,956 and 239,155, both filed Aug. 31,1988, and assigned to Zenith Electronics Corporation. The output of modulator 16 is supplied to a state table compressor 20 where the coded representation of the removed low frequency video information (in the form of delta modulated sample components) is compressed and supplied to a constellation table 21. The outputs of video compressor 18 and constellation table 21 are in I and Q form and are applied to a multiplexer 23 which supplies I and Q output signals to a quadrature modulator 24. The output of modulator 24, which is at intermediate frequency, is applied to a transmitter 25 where the signal is converted for transmission at RF frequency. A sync separator circuit 14 is coupled to video source 10 and supplies a timing circuit 26 for controlling the operation of the various functional blocks of FIG. 1.

In FIG. 2, a receiver for receiving and decoding the transmitted signal is shown. The receiver input circuit 28 receives the transmitted signal and supplies it to a video demodulator 30 and to a quadrature demodulator 32. The video demodulator 30 recovers the compresses high frequency portion of the transmitted video signal and applies it to a video expander 31. Quadrature demodulator 32 recovers the transmitted compressed data, including the data for reconstituting the remove low frequency video information. The I and Q outputs of quadrature demodulator 32 are supplied to a table 36 for conversion to appropriate four-bit compressed data words that are expanded to 12 bits in a data expander circuit 38. The output of data expander 38 is applied to an inverse delta modulator 40 where the direction and slope components of the data are recovered and supplied to a reconstruction filter 42 where the the removed low frequency video information is reconstituted. The reconstituted low frequency data is applied to a summer 34 where it is recombined with the high frequency video information from video expander 31. The output of summer 34 is the demodulated video signal which is utilized by apparatus (not shown) in a conventional way.

Referring again to FIG. 1, the removed low frequency video information is digitized by suitable apparatus in block 16 to develop a plurality of samples. Each sample is encoded in a delta modulation format and expressed as four bits of data, with two bits representing the direction of the sample (i.e. whether it is increasing or decreasing) anid two bits representing the slope or amount of any change in magnitude. As illustrated in FIG. 3, the bits are identified as D0 and D1, for direction, and SO and S1, for slope. The direction and slope bits corresponding to three sucessive samples are grouped together into a 12-bit packet where sample 1, sample 2 and sample 3 each include two bits corresponding to D0 and D1 and two bits corresponding to S0 and S1.

Each 12-bit packet represents a particular state of the three corresponding samples. The packets are entropy encoded to reflect their probability of occurence. Entropy encoding is predictive and is based upon the statistical likelihood of occurrence of each of the 12-bit delta modulation data packets in typical video scene information. Thus, by analyzing a large number of such 12-bit packets derived from a variety of video signals, a classification or hierarchy, based upon the statistical likelihood of occurrence of each packets is developed. Entropy Encoded Packets and State Table Compressor 20 (State Table 20 ) has the various combinations of 12-bit packets arranged in order of their statistical likelihood of occurrence in a typical video signal. There are 4096 unique combinations of the 12-bit packets. State table 20 maps each 12-bit packet into a compressed four-bit output code, with the first 15 output codes corresponding to the 15 most likely to occur 12-bit packets. These 15 four-bit output codes are effectively supplied directly to the modulator 24, via multiplexer 23. The remaining 12-bit packets are all mapped to the last four-bit output code, referred to herein as a linking code, of state table 20. The linking code and remaining 12-bit packets are in turn supplied to constellation table 21. (In practice the tables 20 and 21 may be provided in the same memory and are separately shown here for purposes of explanation). In response to the linking code, constellation table 21 further maps the 255 most likely to occur of the remaining 4080 12-bit packets (i.e. 4096-16) into a unique eight-bit output code. All of the remaining 12-bit packets (combinations 273 -4096) are mapped to the common remaining state of the eight-bit output codes. The four-bit output codes and the four-bit linking code, and the eight-bit output codes are all supplied to quadrature modulator 24, via multiplexer 23.

As shown in FIG. 4, the quadrature modulator operates along I and Q axes. The various circles indicate the constellation states of the modulator. Constellation state "0" occurs at the origin and can be advantageously used to represent a signal requiring the least amount of transmission power. The remaining constellation states are sequentially numbered from 1-15, in the order of increasing transmission power requirements, for signals identified by corresponding I and Q components of the modulator. It will be appreciated that the 0 state which requires the least amount of transmission power is assigned to the most likely or occur of the four-bit output codes. In most video signals the most likely occurrence (from point to point) will be no change. Consequently, the 0 constellation state is ideal for the four-bit output code corresponding to no change. As mentioned, the likelihood of occurrence of the other four-bit output codes is determined empirically with the codes less likely to occur being transmitted by the constellation states requiring increased transimission power. The 15 four-bit output codes of FIG. 4 correspond to constellation states 0-14 with constellation state 15 being used for the linking code (the highest transmission power state). The eight-bit output codes are transmitted using this constellation state. In this way, the data in the vertical blanking interval is compressed and requires minimum transmission power. The reduced transmission power enables the data to be sent with low power television signals corresponding to Taboo co-channel and adjacent channels.

It is recognized that the number of packets selected is arbitrary. In the preferred embodiment, the 12-bit packets are believed to be satisfactory in view of the conjoint demands of resolution, transmission power and system cost. Variations in the preferred embodiment will occur to those skilled in the art without departing from the true spirit and scope of the invention which is to be limited only as defined in the claims.