NO159900B - PROCEDURE FOR TRANSFER OF PROGRAM INFORMATION SIGNALS IN A TELEVISION SYSTEM AND SYSTEM FOR EXECUTION OF PROCEDURE. - Google Patents

Description

The present invention relates to a method for transfer

of program information signals in a television system of the type specified in the introduction to claim 1 as well as a system for carrying out the method of the type specified in the introduction to claim 8.

Transmission and reception of information in the frequency spectrum allocated to television has developed to a point where viewers are provided with information of an audio or video nature in addition to normal audio and video signals associated with a specific television programme. The normal television video signal is ordinarily transmitted with an associated audio signal within an allocated frequency band of about 6 MHz. In recent times, however, additional information has been transmitted and received by some viewers with normal audio - video signals, especially with subscription television (STV).

Most STV systems include e.g. a video signal which is scrambled for certain subscription programs, an audio signal associated with the scrambled video signal and a further audio channel generally referred to as an announcer channel, also called announcer channel in the description. The announcer channel is transmitted at normal audio center frequency within the band allocated to a television channel. The announcer is generally used for messages regarding descrambled video signals available to all viewers, but not in relation to subscription television programs that are scrambled. Audio signal related to scrambled subscription television programs is transmitted at a center frequency, shifted from the normal audio center frequency and thus cannot be detected by ordinary television receivers. A subscription television system may therefore include a video signal at its normal location in the assigned frequency band (but scrambled) and two audio signals, one at the normal audio frequency position in the television channel frequency band and the other offset from the normal television audio frequency location.

in the television channel frequency band and the other offset from the normal television audio frequency location.

The normal television bandwidth allocated to a particular channel

can also be used to transmit additional information related or unrelated to television program video where subtitles or other textual data are provided. With subtitling and teletext systems, e.g. information be transmitted at vertical intervals to a normal television video signal or at other suitable locations in the signal. The additional information is broadcast by normal television band assigned to a particular channel and is detected at the television receiver by a special adapter or decoder which then produces the presentation of the subtitle or other text material on the television screen (CRT).

Other systems have been proposed where, of course, additional information was transmitted, although not over normal television frequency bands, to a particular channel to increase the amount of information available to a television viewer during the broadcast of a program. Television and FM stations often cooperate by simultaneously broadcasting a television program over a normal allocated television channel and by broadcasting its associated audio over an allocated FM radio channel. The viewer can thus watch a program on a normal television

device while at the same time listening to the sound in stereo using an FM receiver. A similar system has been proposed and has recently been used to produce multi-lingual audio for a particular television program video. E.g.

in a system described in the July 1980 issue of TW World page 34, an ordinary television signal is obviously late in the frequency band normally allocated for a television channel and an additional audio signal, in another language, is transmitted over an FM radio band. The viewer can choose either the FM radio band or the normal television sound and thereby choose between two different languages when viewing the programme.

One purpose of the present invention is thus to provide information in addition to normal television audio and video from which a television viewer can choose. A primary purpose, however, is to provide such information within the normal frequency band allocated to a television channel. Moreover, it is an aim of the present invention to provide a greater variety of information than has previously been possible without exceeding the allocated frequency bandwidth of a single television channel.

A further object of the present invention is

to provide information in addition to normal television audio and video into a frequency band allocated to a single television channel in a way that allows the selection of a desired type of information by only those who are subscribers and thus qualified to receive this information.

It is therefore a further object of the invention to provide a new method and system for the transmission and reception of television signals within a normal frequency band allocated to a television channel where a television program video signal and at least three other signals are transmitted and can be selectively received.

Another object is to provide a new multi-signal scrambling and transmission technique in which multiple signals are transmitted at a minimum frequency bandwidth in a format secure from reception in an intelligible form by a normal receiver, but is available on a selective basis to qualified persons having decoders .

A further object of the present invention is to provide a new television transmission and reception method and system, especially for subscription operation where the subscribers can choose between several multilingual audio tracks belonging to the television video program material.

Yet another object of the present invention is to provide a subscription television system and a method in which highly secure transmission of the STV program material is effected by low level, relatively low security scrambling of the video part of the television program and high level, very secure scrambling of the audio part of the program and in which a of several audio signals may be selected by qualified subscribers.

The above-mentioned purpose is achieved by means of a method of the nature mentioned at the outset, the characteristic features of which appear in claim 1, as well as by means of a system for carrying out the method and the characteristics of which appear in claim 8.

Further features of the method and system appear in the other requirements.

The above-mentioned and other objects and advantages of the present invention will become clearer to a person skilled in the art from the following detailed description with reference to accompanying drawings, where: Fig. 1 shows a functional block diagram of a television transmitting and receiving system in accordance with the present invention.

fig. 2 shows a functional block diagram of the audio encoder and the combining device of fig. 1 in more detail,

fig. 3 shows a curve of the baseband audio spectrum produced in accordance with the embodiment shown in fig. 1-2,

fig. 4 showed a functional block diagram, which shows in more detail an embodiment of an audio encoder or scrambler in accordance with the present invention,

fig. 5 shows a curve of the scrambled audio signals produced by the encoder of fig. 5

fig. 6 shows a functional block diagram of another embodiment of an audio coder or scrambler according to the present invention,

fig. 7 shows a curve of scrambled audio signals

brought by the encoder in fig. 6,

fig. 8 shows a functional block diagram of the audio decoder and selector of FIG. 1 in greater simplicity,

fig. 9 shows a functional block diagram in greater simplicity of an embodiment of the decoder and selector controller of

fig. 1, and

fig. 10 shows a functional block diagram of an embodiment

form of a receiver turning device and code category storage device according to the present invention.

A commercial broadcast television channel is generally allocated a frequency band of around 6 MHz within which audio and video signals must be transmitted at allocated center frequencies. The majority of the bandwidth is taken up by the video signal with around 200-250 KHz at the upper end of the band remaining for audio information. In accordance with the present invention which is functionally shown in fig. 1 and described in more detail later, the television video signal is modulated onto a video carrier wave at a normal video center frequency for the particular television channel. This video signal may or may not be scrambled, but its location in the frequency band and its transmission may be normal in its entirety. The transmitted video signal can therefore, in accordance with the present invention, be received by any ordinary television video receiver, even if it is scrambled or contains additional information, the video may be uninteresting or even impossible to view and additional information may be lost without further signal processing by the receiving location.

At least one audio signal is simultaneously transmitted at normal television audio carrier center frequency and can be received on a normal television audio receiver. Because of its "normal" location in the frequency band of the television channel, this audio signal is hereafter referred to as the normal audio signal. As with the other "audio" signals described here, the information content of the normal audio signal can vary from normal program audio or sound to information unrelated to the video. However, each "audio" signal referred to herein will be limited in the frequency range of those audio signals (eg, from 20 to 20,000 Hz and preferably from 50 to 15,000 Hz). The term audio signal with regard to the present invention can therefore refer to a signal limited in frequency to the audio frequency range. Such signals may include audio information associated with a program video signal (e.g. soundtrack of a film either mono or stereo), additional audio information associated with video (e.g. one or more foreign language versions of a film soundtrack), and information unrelated to the program video signal (e.g. control signals, fax information, etc).

As appears from the description later, several audio signals other than normal audio are provided in the composite television signal within the band portion allocated for audio transmission. These other audio signals are shifted in carrier frequency from the normal audio carrier signal frequency, but the composite audio signal is within the 200-250 KHz band allocated for audio transmission from the frequency band to a single television channel.

Fig. 1 shows an embodiment of a system according to the present invention where signals produced at a television station or at other central locations are transmitted to a subscription location (or a viewing location in the case of a no-subscription service). Transmission from the television station to the subscription location may be over broadcast as shown or by other suitable common means such as cable.

However, it should be noted that certain features of the present invention are applicable to any type of transmission, while others, particularly the scrambling technique and the frequency spectrum conversion feature described later, are more applicable to a broadcast system. While the invention is described in connection with the US commercial transmission standards, it should of course be clear that the principle of the present invention can be applied to different television signal transmission formats including those to other countries and those used by satellites or MDS transmissions.

In fig. 1, a normal audio signal source 10 produces a normal audio signal N.AUDIO to an audio encoder and combiner 12, described in more detail later. Similar audio signal sources 10A, 10B, . A combined audio signal CAUD is supplied from the encoder and combiner 12 to a conventional exciter (modulator) and transmission circuits 16. In the case of conventional television systems, the audio signal is supplied to an FM exciter and transmitter as the signal is sent as a frequency modulated signal.

The video equipment at the television station includes a video signal source 18 which, in the case of a subscription system, has video scrambling of some program material, produces video signal VID to a video encoder 20. In the case of a scrambled video system, a video code control circuit 22 produces a video control signal VCTL to the video encoder 20 and either an unscrambled or a scrambled video signal VID or SVID is supplied from the video encoder 20 to the exciter and transmitter 16. For video signal transmission in the normal television system, the video signal VID (SVID) is supplied to the AM exciter and subsequently transmits information in an amplitude modulated form. In the case of a system without video scrambling such as a commercial TV system or a subscription service where the subscriber pays only for non-video service, a video encoder and encoder control circuit is naturally not necessary.

At the viewing location, ordinary AM and FM receivers receive and detect respective video and audio signals transmitted from the television station. The AM receiver produces the VID (SVID) signal to a video decoder 26 which in turn produces the video output signal VIDEO OUT to common modulators 2 8 for re-modulation at a particular television channel frequency. FM

the receiver produces the combined audio signal CAUD

for an audio decoder and selector 30 which in turn produces an audio output signal AUDIO OUT to the modulators 28. The combined audio and video modulator output signal is supplied to the television antenna terminals of a conventional television set.

Video and audio encoders are provided with different control signals as described below in more detail by means of a decoder and selector controller 32. In a subscription system, the signals from the decoder and the selector controller can

32 also be provided with a suitable accounting mechanism 34

of suitable ordinary construction. With such a billing mechanism shown in US patent no. 4,163,254, the billing information is e.g. stored in the accounting mechanism 32 for periodic polling over telephone lines. As will become apparent later, such a billing mechanism is easily adaptable for use in connection with the present system when information sent by a subscription is to be paid for on the basis of the selection and use of this information.

In operation, the normal audio signal and several additional audio signals are combined by the audio coder and combiner 12 to form the composite audio signal CAUD during the control of the audio code control signal ACTL. In a preferred embodiment of the invention, free audio signals are combined in addition to the normal audio signal to form the composite audio signal. However, it should be noted that other numbers of audio signals in addition to the normal audio signal may be used to form the composite audio signal.

As will be seen later, audio codes and the combiner combine

12 the multiple audio signals over a frequency baseband spectrum that is not greater than one half of the bandwidth allocated to the audio transmission of a normal television channel. In this way, the total bandwidth of the frequency-modulated signal transmitted by the FM exciters and the transmitter does not exceed this allocated frequency band. The different audio signals are also placed within the baseband audio spectrum in shapes

which are easily separable with relatively cheap hardware.

These audio signals are e.g. shifted in frequency from

the other sufficiently so that they can be separated using relatively cheap conventional filtration techniques.

It should also be noted that when using in-phase and cross-phase components for a single carrier to produce different forms of audio signals, they can be easily separated and thus detected.

The video signal VID from the video signal source 18 is used for

to modulate a standard AM exciter and transmit when any other standard television video signal would be used. However, as previously described, it may be desirable to scramble the video signal before the transmission and thus a video coder can be placed for this purpose. The video encoder 22 can determine in any suitable way whether or not the video encoder 20 scrambles the video signal and, in addition, in what manner the scrambling is performed. The encoder 22 cannot of course be required if relatively simple scramblings such as suppressed synchro or alternating current signal input techniques are used.

The composite television signal, i.e. transmitted AM video and FM audio signals are received respectively by AM and FM receivers in a suitable and known manner in order to produce a composite audio signal and video signal at desired intermediate frequencies. In case the video signal is scrambled (ie the SVID signal

is transmitted and received) the video decoder 2 6 descrambles the received video signal). Depending on the type of scrambler used, the video decoder 26 demodulates the total video signal to produce the baseband video signal, reconstructs synchro or polarity, if synchro or polarity scrambling is used and then produces a reconstructed video output signal VIDEO OUT at the baseband. This video output signal is then remodulated at a desired television channel frequency which may or may not correspond to the frequency band of the originally transmitted television signal. In accordance with a feature of the present invention, i

the actual audio and video output modulators are fixed at a specific frequency corresponding to a television channel (e.g., channel 3) and the AM and FM receivers 24 include an all-channel tuning circuit which enables all channels to be received and viewed on a television set. without changing the channel selector on the device (see fig. 10). To select a different television broadcast band, i.e. a different television channel, the tuning circuit of the AM and FM receiver 24 is changed to change the desired channel without the need for any change in the channel selector of the television set. In this way, enemy control and other features described later can be provided in a simple way on existing enemy vision devices.

The composite or combined audio signal CAUD received

by means of the FM receiver is supplied to the audio decoder and selector 30 and a single audio signal is selected as the audio output signal AUDIO OUT. The selection of the audio output signal is controlled by the viewer through the decoder and the selector controls 32 described in more detail later. The decoder and the selector controller 32 can also produce the selection of one or more additional audio signals for input at an output contact or other equipment than the normal remote control receiver. Selection of a specific audio signal or signals and/or acceptance of a scrambled video signal for descrambling and viewing can be registered by the billing mechanism 34 for printing a bill as previously mentioned. In this regard, information contained in the video program signal such as program identification may be fed to the calculation mechanism as shown dashed to provide a complete record of the viewer's selection.

Fig. 2 shows an embodiment of audio codes and the combining device 12 in Fig. 1 where a normal audio signal N.AUDIO and four additional audio signals AUDIO A - AUDIO D are coded and combined for transmission as composite audio signal CAUD. In the embodiment in fig. 2 are the audio signals

not necessarily scrambled, although, as it appears, all with

exception of normal audio N.AUDIO will be shifted in the audio frequency band from the ordinary television audio mid frequency so that special equipment will be necessary to detect additional audio signals. It should also be noted later that in connection with fig. 4-7 by further adding a scrambling circuit at the front end of the audio coder and combiner, additional audio signals can be safely scrambled so that even with equipment that can receive and detect additional audio signals, it cannot be used if the subscription equipment does not have a suitable descrambling option.

Fig. 2 shows that each of the normal and additional audio signals is filtered by an associated bandpass filter 40-40D. Each filter 40-40D is a regular filter that limits the frequency range of the audio signals between 50 and 15,000 Hz,

an area well within the audio frequency range and even sufficient to transmit hi-fi sound information.

The output signals from each filter 40-40D are supplied to respective pre-emphasis circuits 42-42D. These pre-emphasis circuits are ordinary circuits which have about a 75y time constant at the frequencies involved. The lower frequency part of the audio signals consequently has a relatively constant amplitude with a frequency of around 2.1 KHz. At the 2.1 KHz cutoff, the amplitude of the audio signal becomes "emphasized" and increases in a generally linear fashion (see Fig. 3).

The output signal from the pre-emphasis circuit 42 is supplied to a conventional summing or combining circuit 44 while the output signals from the pre-emphasis circuits 42A-42D are supplied to respective mixers 46A-46D of a conventional type. The mixers 46A-46D can e.g. be ordinary double-balanced mixers. The baseband audio spectrum of each of the signals from the pre-emphasis circuits 42-42D is substantially the same and corresponds to the generally shown portion of the spectrum denoted NAUD' in FIG. 3.

The output signals from mixers 46A-46D are respective pre-emphasized audio signals double sideband, suppressed carriers, amplitude modulated (DSB-SC-AM) at one of two center frequencies and either in-phase (I) or 90° out-of-phase or in cross-phase with a pilot tone . In this regard, a pilot tone of the desired frequency is applied to a common cross-phase carrier generator 48 which produces an in-phase signal I and a cross-phase signal Q at each of two separate frequencies related to the pilot tone frequency. These I and Q signals at each of the selected frequencies are used as carrier waves in the DSB-SC-AM multiple audio feature described in the following. The pilot tone was also fed to summing circuit 44 for transmission as part of the composite audio signal.

The selected frequencies in the embodiment shown are brought in relation to the horizontal synchronous frequency F„ of the television video signal since this frequency is advantageous and also has multiples which are uniquely suitable for the system of the present invention. The frequency 5F„ ti/2 (39.335 KHz) is specifically selected as a carrier frequency and the frequency 10F„ri/2 (78.67 kHz) is selected as the other. As will be seen later, this ensures about 10 kHz between each audio band in the composite baseband audio spectrum and causes the transmission of five audio signals (using the I and Q components) within the audio frequency bandwidth allocated in a television channel band.

By continuing to connect to fig. 2, the in-phase carrier signal 5FH/2 (0°) is applied to mixer 46A while the cross-phase carrier signal at the same frequency 5FH/2 (90°) is applied to mixer 46B. The in-phase carrier signal 10FH/2 (90°) is applied to the mixer 46D. The I and Q signals from mixers 46A and 46B are combined by means of a conventional summing circuit 50 and the I and Q signals from mixers 46C and 46D are applied to a similar summing circuit 52. The output signal from the summing circuit 50 is applied via a conventional low-pass filter 54 to an input terminal of the summing circuit 44. The output signal from the summing circuit 52 is supplied to the second input terminal of the summing circuit 44 and the output signal from the summing circuit is supplied via a conventional low-pass filter 56 as composite audio signal CAUD.

It should be noted that the signal from the combiner 50 is composed of the AUDIO A and AUDIO B signals on the I and Q components of the carrier SFr„i/ 2 (33.335 kHz). The highest desired frequency in this composite signal will be around 54.335 kHz since the composite audio signal will extend on both sides of the carrier for a maximum of 15 kHz (ie from 24.335 kHz to 54.335 kHz). The result is a frequency spectrum indicated by the curve AUD AB' in fig. 3 within the baseband audio frequency spectrum.

The combiner 52 likewise produces a composition of the audio signals AUDIO C and AUDIO D modulated on the I and Q components of the lOF.r,i/2 carrier. The center frequency of this composite audio signal AUD CD<1> is at 78.67 kHz and the same signal extends from 63.67 kHz to 93.67 kHz. When this composite signal is consequently combined with the composite signal AUD AB' and the pre-emphasized normal audio signal NAUD', the composite audio signal can be filtered for

to let through only signals with frequencies below the highest expected frequency of 93.67 kHz (filter 56).

The resulting baseband audio frequency spectrum of the composite audio signal CAUD is therefore the total baseband spectrum shown in FIG. 3. This audio spectrum will include pre-emphasized normal audio signals AUD N') in the baseband frequency range 50 to 15,000 Hz with a flat response between 50 and 2100 Hz and an emphasized response thereafter. The additional audio signals are included in the spectrum in I and Q phased, DSB-SC-AM signals centered around the pilot frequency and twice the pilot frequency. The pilot signal itself is provided as part of this component spectrum and, as shown, the pre-emphasis of each additional audio signal comes as a flat response at 2100 Hz on both sides of the center frequency and a rising curve thereafter.

The video horizontal frequency Fu n and its harmonics are shown in phantom relative to the baseband audio spectrum since they can present interference problems. AM-to-PM conversion can and does take place, e.g. in the high power stages of the video AM transmitter. These PM sidebands are found at fu, 2f„ and

ride n

higher harmonics. When the intercarrier wave detection is used at the receiver, the PM sidebands around the video carrier signal are intermodulated to the 4.5 MHz intercarrier sweep used for FM detection. In FM detection, the PM sidebands produce unwanted tones in the detected audio signal. For the standard broadcast system this does not pose a problem since unwanted tones are at F„ (15.734 Hz)

and higher harmonics that lie outside the 15 kHz audio band. These unwanted tones would occur as shown by phantom lines in fig. 3. To eliminate them as a problem, it should be noted that the decoder according to the present invention does not use an intermediate carrier detection, instead the audio intermediate frequency (IF) at 41.25 MHz is reshaped down to 10.75 MHz, filtered, amplified, limited and then detected. There will thus be no significant intermediate modulation of the video carrier PM on the audio signal.

As previously mentioned in connection with fig. 1, the composite audio signal CAUD is fed to a conventional FM exciter (modulator) and transmitted for broadcasting. In general

the exciter/transmitter will include pre-emphasis, which with a single audio signal at the transmission frequency would entail the pre-emphasis shown in fig. 3 for the NAUD<1->signal.

In accordance with the present invention, however, a

light emphasis approach undertaken. The initial pre-emphasis of the baseband audio signal is completed as previously described and instead of normal exciter pre-emphasis the entire baseband audio spectrum is pre-emphasized by a time constant

which would result in a flat portion on the baseband audio signal from 50 to a breakpoint of about 33,000 Hz (a break-

point well above the normal baseband audio range transmitted

in a television system) and an upward tilt, pre-emphasized

response above the 33 kHz breakpoint. For the audio FM modulator or exciter of a broadcast television system according to the embodiment shown, the biased time constant in the exciter (again completed via conventional inductive load or the like) will thus be modified to be of the order of 4.823 sec. to provide this 33 kHz breakpoint.

In addition, the FM exciter/transmitter in accordance with the present invention is preferably set for narrowband operation at modulation frequencies above the 3 3 kHz breakpoint with a transition to broadband operation at modulation frequencies considerably below 35 kHz. This can be completed in any suitable conventional way using e.g. setting transmitter modulation indices for a deviation around 18,734

Hz when a tone at 93.671 Hz is applied at the appropriate level.

With this broadband-low-frequency, narrowband-high-frequency transmission approximated, the bandwidth occupied by the modulated carrier wave would be about 189.102 Hz (2[880 + 93.671]),

a bandwidth well within the allocated 200-250 kHz bandwidth for the television audio signal.

According to the present invention, the additional audio signals can

be scrambled so that only authorized or "qualified" subscribers can use the information available in the additional audio signals. A way of scrambling the audio signals for secure transmission in the case of two additional audio signals is shown in fig. 4-6.

With reference to fig. 4 becomes a first information signal

A such as a Spanish language audio signal associated with the video signal which is transmitted is supplied to an input terminal of a conventional electronic selector switch 60. The information signal A is also supplied to an input terminal of an identical selector switch 62. However, if this signal A is supplied normally closed (e.g. normally conducting 1 side of the switch 60 then the signal A is supplied to the opposite or normally open

(e.g., normally non-conductive) side of switch 62. In this way, the same control signal (e.g., a binary signal) applied to switches 60 and 62 will cause signal A to appear at the output terminal of switch 60 while simultaneously blocking it from to appear at the output of the switch 6 2 and vice versa for each specific control signal level.

A second information signal B such as a French language version of the signal A is produced at a second input terminal of the switch 60 and at the opposite input terminal of the switch 62. In this way, the output signal at the output of the switch 60 will be the information signal B whenever the control signal is at a level and , at the same control signal-

level, the signal B will be blocked from the output signal of the switch 6 2 and the signal A will appear at this terminal. It will thus be clear for each control signal that while signal A appears at the output of switch 60, signal B will appear at the output of switch 62. When the control signal changes, opposite output signal conditions will be present.

A conventional pseudo-random binary sequence (PRBS) signal generator 64 produces control signals for switches 60 and 62 as shown.

In response to a clock signal, the PRBS generator produces output pulses having pulse widths that vary in accordance with a fixed repeating pattern. However, this pattern has no discernible logical frequency and it repeats after a sufficiently long period of time that the overall pattern appears to be random.

The PRBS generator signal pattern can be varied accordingly

with an initial code essentially loaded into the PRBS generator by means of a reset pulse RESET as shown. The reset pulse thus starts the PRBS generator at a point in a pseudo-random sequence determined by the start code and the pseudo-random sequence

continues from that point as a result of CLOCK. As will be seen later in the description, this allows the same pseudo-random pattern to be simultaneously generated at a remote location using an identical PRBS generator clocked in phase with and at the same hash height as the CLOCK signal and booted with the same start code at the same reset time ( relative to the timing of the received audio signal of course) .

The PRBS generator can also be opened and blocked using respective STV signals. The STV signal opens the PRBS generator to produce a pseudo-random sequence and thus scrambles signals A and B and the STV signal disables the PRBS generator and removes the scrambling.

It will be obvious that if only two further audio signals are produced as shown in fig. 4, it will be necessary to use only the NAUD' and AUD AB' parts of the baseband audio spectrum. Likewise, only the AUDIO A and AUDIO B channels of an encoder and combiner 12 in fig. 1 and 2. It should be noted that with only two additional audio signals, the AUDIO A and AUDIO C channels of the encoder and combiner 12 can of course be used and two additional audio signals transmitted with only I modulation which thereby further eliminates the combiner 50 and 52.

Fig. 5 shows the timing and the signal containing the AUDIO A and AUDIO B signals produced by the scrambler in fig. 4 in relation to control signals from the PRBS generator 64 in scramble or STV mode. When a binary "1" signal level is produced by the PRBS generator, the AUDIO B signal contains the information signal B while the AUDIO B signal contains the information signal A. When the PRBS generator's output signal switches to a binary "0" level, the AUDIO A signal contains the information signal A and the AUDIO B signal contains the information signal B.

It appears that neither the AUDIO A nor the AUDIO B signal contains a complete information signal, but together they both contain information signals in their entirety. Since both AUDIO A and AUDIO B signals are transmitted, a

correct reversal of the scrambling process each information signal in its entirety with relatively simple filtering without appreciable destruction. It should also be noted that proper reverse-

ing cannot be completed without the correct start code, the correct reset pulse and the correct clock signal (as well as an audio receiver that can receive and detect additional audio transmissions). Scrambling is therefore extremely secure even with only two audio signals added to normal audio signals.

When four additional audio signals have been provided and scrambled, an unauthorized descramble is of course practically impossible. Four information signals, SIGNAL A - SIGNAL D

can e.g. be supplied to a coder (scrambler) in accordance with the present invention as shown in fig. 6. The encoder shown includes four switch circuits 68-A-D (eg, a switch matrix) each connected in a conventional manner to select one of four input signals as an output signal. It is clear that the switch circuits are controlled so that at each eye-

glance, each of four information signals is selected as an output signal for only one switch circuit as shown by the curve in fig. 7.

In fig. 6, the switch circuits 68 receive four additional information signals, SIGNAL A-D and select one as output signal AUDIO A

in response to the control signals "o-3". The "0" control signal closes a switch "0" in circuit 68A, which connects the first input terminal to the output terminal and thus causes SIGNAL A to be applied to the output as the AUDIO A signal. The control signal "1" also closes an associated switch "1"

in the circuit 6 8A, which connects the second input terminal to the output and thereby supplies SIGNAL B to the output. The control signal "2" connects a third input terminal (SIGNAL C)

with the output terminal and the control signal "3" connects the fourth input terminal (SIGNAL D) to the output terminal.

The control signals "0-3" are produced in a suitable and usual way in response to the PRBS signal from the PRBS generator in fig. 4. The PRBS signal and the CLOCK signal previously described in connection with fig. 4 can e.g. be fed to a logic circuit 70 which includes conventional logic gates. The logic circuit produces the "0" control signal (eg, the "0" control signal assumes a high signal level when the "1-3" signals remain at a low level) as a result of simultaneous binary "NULL" logic levels in the PRBS and CLOCK signals. The "1" control signal is generated when the CLOCK signal is a binary "ONE" and the PRBS signal is a binary "NULL". The "2" control signal is generated when the CLOCK signal is a binary "NULL" and the PRBS signal is a binary "ONE" and the "3" control signal is generated when both the CLOCK and PRBS signals are binary "ONE". In this respect, the logic circuit 70 is operated as a normal binary to decimal decoder.

The information signals SIGNAL A-SIGNAL D are always supplied to the switch circuits 68B-68D, but with different input terminals on each switch circuit. While each switch circuit 68A-68D are identical and make identical connections for a particular control signal, the output signal of each will thus be different from a particular control signal As can be seen from Fig. 7 when the CLOCK signal is high (binary EN) and the PRBS signal is low (binary ZERO), for example the control signal "1" is produced and the respective B, A, D and C information signals supplied as AUDIO A, AUDIO B, AUDIO C and AUDIO D output signals.Each time the clock and/or PRBS generator output changes, the information signal contained in the audio output changes in an apparently irregular manner.

The code embodiment shown in fig. 6 is operated with equal time increments between audio signal containing changes, rather than pseudo-random increments produced in accordance with the two signal embodiment of fig. 4. The coding therefore lies in the pseudo-irregularity of the audio signal content more than in the timing. However, it should be noted that a further security element can be provided in the four signal implementation by introducing pseudo-irregular time intervals between the signal content changes as well as displayed pseudo-irregularities in the content. This can be provided e.g. by using a second pseudo-irregular bit sequence generator instead of the CLOCK signal as an input to the logic circuit 70 or in any other suitable manner.

Fig. 8 shows in more detail an embodiment of the audio decoder and the selector and controllers 30, 32 at a subscription site of the embodiment of fig. 1. Fig. 8 shows the composite audio signal CAUD (M.F.) at the intermediate frequency of the FM receiver (indicated by 24 in Fig. 1) and as pre-emphasized

by means of the FM exciter/transmitter at the vision station is fed to a conventional dual balanced mixer 80 together with a local oscillator signal L.O. The composite audio signal in the embodiment shown is a 41.25 MHz intermediate frequency and is mixed with a 30.5 MHz local oscillator signal. This produces a composite audio signal at an intermediate frequency (M.F.) of 10.75 MHz (as well as other mixer products) to the output of the mixer 80. The M.F. signal for the mixer 80 is fed to a conventional bandpass amplifier-limiter 82 and the output signal from the limiter 82 becomes supplied with a conventional discriminator 84. The band-pass amplifier-limiter 82 and the discriminator 84 are designed for operation at 10.75 MHz intermediate frequency. The composite audio signal at 10.75 MHz M.F. is generally filtered and amplified and limited by the circuit 82 and FM is usually detected by the discriminator 84 (e.g.

by cross-phase FM detection technique). The resulting output signal from the discriminator 84 thus becomes the pre-emphasized composite audio signal. A common de-emphasis circuit 68 with a 4.8228y time constant (33 kHz breakpoint)

to interact with that of the exciter/transmitter de-emphasizes the composite audio signal to produce the composite audio CAUD provided at the output of the audio encoder and combiner 12 of FIG. 1 and 2.

It showed the embodiment of the audio decoder and selector 3 0

on fig. 8 arranged to process four audio signals in addition to the normal audio signal. The normal audio signal is detected by the composite audio signal CAUD simply by filtering through a low pass filter 88 with an upper cutoff frequency of 15 kHz. The signal from the filter 8 8 is the NAUD' signal in fig. 3 and after the de-emphasis by a conventional de-emphasis circuit 90 having a 75ys time constant (2.1 kHz breakpoint) this signal becomes the normal audio signal NAUD with a flat frequency response. This NAUD signal is supplied to an input terminal of a conventional two-input terminal selector switch 92 designated STV switch in fig. 8.

The composite audio signal CAUD is also fed to a bandpass limiter 94 to detect the 39.335 kHz pilot tone and to four identical audio channels each including a conventional bandpass amplifier 96 and a conventional dual balanced mixer 98. The amplifier 96 and mixer 98 are labeled 9 6A-96D and 98A-98D respectively to denote the audio signal channels with which they are associated. Each of the bandpass amplifiers is designed with a lower cutoff frequency of about 24 kHz and an upper cutoff frequency of about the highest frequency of the composite audio signal in order to pass through the AUD AB' gg AUD CD' signals (fig. 3).

The pilot tone detected by the bandpass limiter 94 is applied to a conventional cross-phase carrier generator 100 which produces

in-phase and in-phase carrier signals at frequencies 5F„H/2 and 10Fu/2. The I-phase carrier signal at the carrier frequency 5FTT/2 from

ri H generator 100 is supplied to mixer 98A while the cross-phase signal at this frequency is supplied to mixer 98B. The in-phase and cross-phase carrier signals at the frequency 10F„ri/2 are supplied directly to the mixer 98C and 98D.

The output signals from the mixers 98A-98D are the respective baseband audio signals AUDIO A - AUDIO D as pre-emphasized by the respective circuits 42A-42D in FIG. 2. These signals may be suitably filtered (not shown) to remove demodulation components above 15 kHz and applied to the input terminals of an audio program selector switch 102. The audio signal selected from the switch 102 is suitably de-emphasized by a conventional de-emphasis circuit 104 having a time constant in 75 sec. The selected audio output signals, after de-emphasis, are then applied to the STV input terminal of switch 92 for selecting the audio output signal in STV mode.

In a system according to the invention where further audio signals are transmitted in an unscrambled form, the switches 102 can be a simple electronic or mechanical switch device which only selects the desired one of four AUDIO signals. Further selection between the normal audio signal and the selected one of the additional audio signals can be provided in a simple manner by means of the switch 92 or by adding switch contacts to the switch 102. In addition, one or more of the additional audio signals can be selected for use (or available all the time) at an output socket on the decoder. A sound track transmitted in stereo can thus e.g.

have one channel transmitted as a normal audio signal and the other as an AUDIO D signal, as the AUDIO D signal is available via auxiliary contact. Similar information for operating peripheral equipment can be provided by auxiliary contacts.

In the case of an STV system, the normal audio signal will be a descrambled signal as shown in the embodiment shown and will be selected as the audio output signal whenever the STV selection signal (e.g. the STV selector switch controlled by the subscriber) is in the "NON-STV"- mode (STV). Ordinarily in STV operation, the normal audio signal is an exclamation signal and the video signal is scrambled. However, until the video signal is accepted by the subscriber for descrambling, a scrambled video signal will appear on the television screen. When the subscriber selects STV or the descrambling mode, the video signal is descrambled and the selected of the additional audio signals is fed as audio output signal. It will of course be clear that this audio signal selection

with regard to the STV signal may be used whether or not the additional audio signals are scrambled, but it has been considered that they will be scrambled to provide an extremely high level of system security.

In the case of a STV system or other subscriber system where it is desirable to transmit additional audio signals in a scrambled form (e.g. as shown in fig, 4-7), the audio program selector switch can include a descrambling circuit controlled by the audio selection signals. The subscriber can e.g. enter the start code in their decoder and that code can be supplied to a regular PRBS generator identical to the one at the transmission station (see fig. 4). Alternatively, this code may be transmitted to the subscriber location on a periodic basis over telephone lines as in US Patent No. 4,163,254 assigned to the assignee of the present invention thereby addressing each decoder and producing the code as part of the transmitted signal as described later.

CLOCK and RESET signals necessary for the PRBS generator

on fig. 4 may be derived from the transmitted video signal or may be special codes introduced into this signal (e.g. at vertical rewind intervals).

A PRBS generator synchronized with the PRB generator at the transmission location is therefore provided in the decoder. The decoder and selector controller 32 in fig. 1 can e.g. include such a PRBS generator 110 as shown in fig. 9. The PRBS signal can be supplied to a logic circuit 112 such as the one previously described in connection with fig. 6 to produce in conjunction with CLOCK SIGNAL "0-3" control signals. In this connection, the audio program selector switch 102 in fig. 8 include a switch matrix such as that shown in fig. 6 to reverse the encoder process in response to the "0-3" control signals and thus produce the information signals, SIGNAL A- SIGNAL D.

A selector switch 114 can be turned by the subscriber to produce the signals A, B, C or D and thereby in the usual way select the desired of the audio information signals at the output of the switch 102 in fig. 8. A, B,,. The C and D signals may be sampled at switch 114 to produce the SV selection signal STV and STV or a separate subscription switch may be provided for this purpose.

It is clear that the placement of four audio signals in addition to a normal audio signal is provided by normal f iron vision audio signal bands and total television bands allocated to a single television channel and thereby provides additional advantages in both normal non-subscriber television systems and subscription systems. In a non-subscription system, a television program may be transmitted with the normal audio signal containing the English-language program soundtrack while additional audio signals may contain a Spanish-language soundtrack, a French-language soundtrack and/or other languages from which the viewer may choose.

In a mixed English language and Spanish language area, like this

as in Los Angeles, the normal audio signal can transmit one language, while an additional audio signal simultaneously produces another. If there is also a large Japanese population, this language can be present in another of the additional video signals. The other audio signals can simultaneously contain subtitle information to drive a character generator and produce subtitles for introduction on video and/or information from other services such as facsimile service, news or stock exchange listings (e.g. in connection with

a printer driven by information signals in the received additional audio signals, etc).

These latter types of services are considered to be subscription services paid for on a periodic basis or on a per-user basis. For subscription services, it may be desirable to scramble the additional audio signals so that only qualified subscribers can use the transmitted information even if the accompanying video signal cannot be scrambled.

In other words, a subscription service may include just the additional audio signals and the main television service over that channel (ie normal audio and video) may remain unchanged.

With such a subscription system, each user of an additional audio signal can be registered for billing. This billing information may be collected on a monthly basis. At the same time, the codes may be required for descrambling

of the audio signals be changed after it has been confirmed that the subscriber is still eligible to receive subscription audio programs and/or services.

In accordance with another feature of the present invention, a flat monthly rate may be charged for programs and services, but different types of programs and services •

is priced at different levels and the subscriber then pays in accordance with the program type and/service type to which he subscribes.

As shown in US patent no.

4 225 884, the program identification signal can be transmitted with program signals and only these program signals which have certain identification codes for which the subscriber has corresponding codes can be used at the subscriber station.

In the aforementioned US patent system, the program identification signal in the program information signals is detected and compared with signals that are produced locally by means of the subscriber's "category selector". A subscriber with a category selector that includes codes for all program categories can receive and unscramble all transmitted programs. However, a lockout feature gives the subscriber the ability to prevent unauthorized scrambling of certain program categories.

In accordance with a feature of the present invention, program identification codes can also be transmitted with program signals (i.e. with television programs including audio and video signals or precisely special audio services). These transmitted codes are detected at the subscriber location

, and compared to locally generated signals to determine whether or not a subscriber is eligible to unscramble the particular program. As in the aforementioned US patent, these program identification signals can be different for each program and thus identify each program individually (so that they can also be used for bill printing, for example) or be different only for different program categories. The category code comparator may alternatively be set to accept several different program codes that fall within a viewer's category qualification. The categorical feature for coding or comparison is preferred

for the sake of simplicity in a flat tax system of the type considered, but any other way can be used.

It should also be noted that program identification signals can also serve as codes used for scrambling and descrambling the program signals. E.g. by using different identification codes for different program categories and providing the subscriber with only these identification codes selected for a viewing period, the subscriber is limited to viewing and/or using only these program categories or audio services with preselected identification codes.

As mentioned above, the identification codes are supplied to the subscriber's decoders in accordance with an embodiment of the invention by separately addressing the decoders and loading these codes over the transmission of a decoder address for each subscriber followed by codes for the program categories that the subscriber is qualified to use.

As shown in fig. 10, assuming that the address and qualifier codes are transmitted as part of the video signal (e.g., in the vertical interval), a remote control all channel tuning circuit 120, preferably of the conventional electronic type, controls the channel tuning of the AM receiver 122 to AM and The FM receivers 16 in fig. 1. Descrambled or scrambled video signal VID (SVID) from the receiver 122 is fed to an address detector and comparator 12 4 and a coded detector and store 12 6. The address detector and comparator detects the address in the incoming video signal and compares the detected addresses with a stored address associated with the subscriber decoder. When a successful comparison is made, the address detector and comparator 124 opens the code detector and store 126 and subsequent codes in the video signal are detected and stored in the store for use by the video and/or audio decoders 26 and 30.

In the embodiment shown, the CLOCK and RESET signal is used for the audio decoder and the selector 30 is also transmitted as part of the video signal. The CLOCK signal can e.g. be a vertical or horizontal synchronous pulse or a coded periodically transmitted signal in the vertical return interval. The RESET signal can also be a specific synchronous pulse or a special code transmitted in the vertical return interval. The video signal VID (SVID) is accordingly also fed to a synchronous or code detector 128 to produce the CLICK signal and a reset code detector 130 to

to detect the RESET signal. The video signal is also supplied to the video decoder 2 6 as shown in fig. 1 which descrambles the SVID signal in a scrambled video system.

It should be noted that it will probably be necessary

to demodulate (detect) the video signal to produce

baseband video if the address and codes are transmitted therein.

Accordingly, a separate video detector (not shown) may be used in front of the address detector and comparator 124 and the other code detectors 126, 128 and 130. However, it may be necessary to provide baseband video for descrambling the video (eg where video inversion is used for scrambling or where it is necessary to reconstruct synchronous signals in order to provide a correct video signal to the television set). A more practical implementation may therefore be to use the baseband video signal from the video decoder 26 for input to the address and code detectors 124, 126, 128 and 130 instead of the VID (SVID) signal as shown.

With continued reference to fig. 10, the power supply switch 132 on the subscriber decoder is connected so that the energy is continuously supplied to the circuits 120-130 (and the video decoder 26 if it produces baseband video signals to the detector 124-130). The power supply switch 132 also applies an OFF signal to the polling circuit 120 to provide an indication of when the subscriber polling circuit is turned off. This OFF signal controls the tuning circuit in the usual way as follows

that the tuning circuit is tuned to a predetermined remote control channel setting (e.g. on channel 3) always when the power supply switch is off.

It has been considered that the program categories as a

subscriber wishes to subscribe to will be stored in the subscriber's decoder either at initial installation or by addressing the decoder after installation. However, the subscriber may wish to add or delete categories after a period of use. Accordingly, shortly after installation and/or at other times before the start of a new viewing period, when a subscriber has indicated a desire to change their category qualification, an address code for that subscriber's decoder is transmitted as part of the video signal. The address code in the video signal becomes

detected and compared to a stored address using the address detector and comparator 12 4. If the addresses match (preferably within certain tolerances)

the code detector and storage 126 is opened using the ENB signal and the category codes that follow the address codes are detected and stored in the storage by the code detector and storage 126. These stored category codes allow the subscriber's decoder to properly descramble and use scrambled signals in a category, but not in another category. The Kate gori code can e.g. be in the form of start codes for the audio signal decoder and the selector 30 and/or the descrambling codes for the video decoder 26. The start codes and the descrambling codes for the audio and video decoders can alternatively be arranged in a different way so that they can be changed more often than the category codes. If the category codes get the different categories to be changed on a periodic basis, this security is naturally eased. If the latter step is taken, e.g. the category codes stored by the circuit 12 6

be compared to the category codes in the received signal to either enable or disable the descrambling of that signal.

It is clear that the scrambling and I/Q transmission technique described herein can be used to ensure the transmission of signals other than video signals. Video signal transmission via satellite can e.g. using in-phase and cross-phase components of a carrier signal and alternating modulation of these components with two different video signals to ensure the satellite transmission. It will also be clear that the scrambling codes for the video and/or the audio information can be generated and transmitted in different ways.

The PRBS generator at the transfer point can e.g. be free-running and its state at the moment can be transmitted periodically so that the PRBS generator at the receiver can be synchronized with it at the time of transmission. The PRBS codes can be provided in a number of ways including transmission

of the video signal as described or the audio signal.

Based on the above-mentioned description, it is made clear that the present invention can be used in other special forms without deviating from the purpose or the essential characteristics thereof. Present described embodiments are therefore to be considered in all respects as illustrative and not as limiting, as the scope of the invention is indicated in the subsequent claims more than in the preceding description and all changes that fall within the meaning and equivalence range of the claims and are therefore intended to be covered by these.

Claims (14)

1. Method for transmitting program information signals in a television system, with the generation of a television video program information signal and several television audio program information signals each associated with the video program information signal, characterized by selective scrambling of at least two of the audio program information signals by mixing parts of the information signals with each other on a time basis in a predetermined manner, transmitting all the information signals including any scrambled audio program information signal, each being transmitted on a separate carrier signal at a frequency within a television channel bandwidth, reception of the sent information signals, descrambling any scrambled audio program information signal by the received separate carrier signals, selecting any of the plurality of audio program information signals, and supplying the selected audio program information signal and the video program information signal to a television video display and sound generating device. 2- Method according to claim 1, characterized in that the transmission steps are characterized by: transmission of a television audio program signal and a television video program signal within respective normal television audio signal and video signal frequency bands in a total television frequency band allocated for the transmission of a single television channel, the audio and video program signals being transmitted at respective normal audio and video center frequencies associated for normal transmission of respective audio and video signals within normal television audio and video frequency bands, simultaneous transmission of at least two additional audio signals within said total television frequency band and at least one center frequency that lies within the television audio frequency band, at least one center frequency being offset -vet from the normal audio center frequency with a value such that during their transmission the two additional audio signals at said at least one center frequency do not interfere with television audio or videop rogram signals and does not exceed the limit of the total television frequency band. 3. Method according to claim 2, characterized in that at least two further audio signals are transmitted by modulation with the further audio signals of an in-phase and a quadrature component to a carrier signal which has at least one center frequency. 4. Method according to claim 2, characterized in that at least two additional audio signals are the scrambled audio program information signals and are sent by modulation with the additional audio signals of an in-phase and a quadrature component to a carrier wave signal that has at least one center frequency. 5. Method according to claims 2-4, characterized by the additional steps: reception of the normal television video signal frequency bands and detection of the television video program signal, reception of the normal television audio signal frequency band and detection of the television audio program signal and two additional audio signals, selection of the television audio program signal or one of the two additional audio signals for generating sound, and producing a television presentation as a result of the detected television video program signal while simultaneously producing sound as a result of the selection of one of the three audio signals. 6. Method according to claim 1 or 2, characterized in that the transmission steps include the steps: transmission of the video program signal in a first frequency band, transmission in a first form within a second frequency band of one and then the other of two audio signals in time sequence intervals such that parts, but not all of the two audio signals are transmitted in the second frequency band in the first form, transmission within the second frequency band in a second form separable from the first form, of one and then the other of the audio signals in the time sequence intervals so that one of the audio signals is transmitted in a time interval in the second form while the other has been transmitted in a time interval of corresponding length in the first form and vice versa, with parts of the one and the other audio signal that have not been transmitted in the first form thereby being transmitted in the second form, so that the two audio signals are transmitted in their entirety in the second frequency band, but are transmitted with a part in d a first form and a part in the second form. 7. Method according to claim 1, where the scrambling of the two information signals is characterized by modulating an in-phase component of at least one carrier wave signal in response to one and then the other of the audio program information signals in a time interval sequence so that parts , but not all of the one and the other of the audio program information signals , are carried by the in-phase component , modulating a quadrature component into at least one carrier signal in response to one and then the other of the audio program information signals in the time interval sequence such that portions, but not all, of the one and the other audio program information signals are carried by the quadrature component and such that i- the phase component and the quadrature component together carry the two audio program information signals in their entirety, and combining the modulated in-phase and quadrature components of the transmission. 8. Television system for carrying out the method according to claims 1-7 for the transmission of program information signals including device for generating a television video program information signal and several television audio program information signals each associated with the video program information signal, characterized by a device for selectively scrambling at least two of the audio program information signals by mixing parts of the information signals with each other on a time basis in a predetermined manner, device for transmitting all the information signals including any scrambled audio program information signal, each of which is sent on a separate carrier wave signal with a frequency within a television channel bandwidth, device for receiving the transmitted information signals, device for descrambling any scrambled audio program information signal with the received separate the carrier signals, means for selecting any of a number of audio program information signals, and means for supplying the selected audio program information signal and the video program information signal to a television video projector and sound producing means. 9. System according to claim 8, characterized by means for transmitting a television audio and video program signal within the respective normal television audio and video signal frequency band to a total television frequency band allocated for the transmission of a single television channel, where the audio and video program signal was transmitted at a normal audio and video center frequency associated with normal transmission of respective audio and video signals within the normal television audio and video frequency bands and device for simultaneous transmission of at least two additional audio signals within the total television frequency band and at least one center frequency that lies within the television audio frequency band, said at least one center frequency being shifted from the normal audio center frequency by a value such that during its transmission the two additional audio signals at at least one center frequency do not interfere with the television audio or video program signals and do not exceed the limits of the total remote the visual frequency band. 10. System according to claim 9, characterized in that the device for simultaneous transmission of the two further audio signals is a device for modulating further audio signals, of an in-phase and a quadrature component to a carrier signal which has at least one center frequency. 11. System according to claim 9, characterized in that devices for simultaneous transmission of the two audio signals are a device for modulating the in-phase component of the carrier signal with one of the additional audio signals while simultaneously modulating the quadrature component with the other of the additional audio signals, and device for periodically reversing the modulating signal so that the second of the additional audio signals is modulated by the in-phase component. to the carrier signal and vice versa. 12. System according to claims 9-11, characterized by: device for receiving the normal television video signal frequency band and detecting the television video program signal, device for receiving the normal television audio signal frequency band and detecting the television audio program signal and the two additional audio signals, means for selecting television audio program signals or one of the two additional audio signals for producing sound, and device for producing a television presentation as a result of a selected television video program signal while simultaneously producing sound as a result of a selected one of three audio signals. 13. System according to claims 8-12, characterized by: first and second sources for at least two different audio signals and a third source for at least one television video program signal, first transmission device for transmission of video program signal in a first frequency band, second transmission device for transmitting in a first form within a second frequency band one and then the other of the audio signals in a sequence of time intervals such that parts, but not all, of each of the two audio signals are transmitted in the second frequency band in the first form, the second the transmission means includes means for transmission within the second frequency band in a second form separate from the first form and then the second of the audio signals in the sequence with time intervals, so that one of the audio signals is transmitted in a time interval in the second form, while the other is transmitted in a time interval of similar length in the first form and vice versa, being parts of the one and the other dre the audio signal is not transmitted in the first form, thus is transmitted in the second form so that the two audio signals are transmitted in their entirety in the second frequency band, but are transmitted in part in the first form and part in the second form. 14 . System according to claims 8 - 13, where the second transmission device is characterized by: device for producing in-phase and quadrature components of a first carrier signal, device for modulating the in-phase component with one and the other of the audio signals for to produce the first form, and means for modulating the quadrature component with one and the other of the audio signals to produce the second form separable from the first form.