FI75702B - Method for sending TV transmissions with stereo sound as well as a coder and decoder for implementing the method - Google Patents

Abstract

A method for stereophonic TV-transmission with a transmission coder and receiver decoder. On the transmission side, the right channel (R) and left channel (L) of stereophonic sound are jointly combined into the channel sum (R+L) and channel separation (R-L), and afterwards samples are taken from channel separation at a significantly lower rate, suitably at half the sampling rate of the sum channel. Signal sampling is performed in digital conversion phase, or from signals already in digital form. When sampling an analogue signal, for example, a 32 kHz sample rate is used respectively with a 16 kHz sample rate. A reduced or halved sample rate enables an aggregate signal to be sent through a multiplexer at a similarly reduced transfer rate (for example about 550kbit/sec), at which a good quality stereo sound signal can be made to fit, for example, a 7MHz channel division. When receiving, right and left stereo sound channels are formed after signal separation and matrixing. In this method, the only theoretical loss of quality is a lack of channel separation in frequencies higher than about 7 kHz, which does not have practical significance.<IMAGE>

Description

1 75702

Method for performing stereo audio TV transmission and encoder and decoder for implementing the method

The present invention relates to a method for performing stereo TV transmission, in which on the transmission side the information information of the two audio channels is digitized and suitably combined and transmitted via a modulator, and on the receiving side the information is again separated and converted to analog. The invention further relates to a method of this referred to the transmission-side encoder, a decoder and a single receiver, satellite receiver, the encoder, through which the signal is forwarded to a single TV vastaanottimel-le.

Some countries have already introduced a stereo and / or two-tone system for television broadcasting. The Federal Republic of Germany has introduced an analogue stereo and two-tone system, in which the necessary additional information is transmitted on a separate analogue subcarrier.

However, digital systems are taking over the industry. In the recording and transmission of sound, such are e.g. Compact Disc (CD) and television MAC satellite transmission standards.

In this context, various broadcasters have studied digital audio transmission methods for TV. The TV set itself is also digitizing at a rapid pace and it is possible to provide an inexpensive stereo and two-tone system if this is taken into account when choosing the transmission standard. One such digital TV circuit solution is ITT, for example. Similar circuits are also largely used in the so-called For decoding MAC satellite transmissions. In the C / D / D2-MAC standard recommended by the EBU (European Broadcast Union), it is possible to transmit several digital audio channels either with 2 75702 14-bit linear coding or using, for example, 10-bit so-called NICAM coding defined by the EBU. With NICAM coding, the required transmission rate is considerably lower and thus more sounds can be transmitted on the same transmission channel.

In the future, more and more program transmission to the consumer will take place via satellites. For example, about 60% of Finnish households are already connected to common antenna or cable systems, and in the future the number will probably increase as satellite reception equipment becomes cheaper. In order for the stereodigital sound of satellite broadcasts to be transmitted to the consumer in the future, the digital audio system of the TV must be such that the sound of the satellite broadcasts can be advantageously encoded at the reception point of the common antenna systems to suit the TV receivers.

Today, there are mainly the following proposals for broadcasting digital TV audio.

1. A system developed by the BBC.

The starting points for the BBC system have been as follows: - the system must be compatible with the 8 MHz channel split used in England - the coding must be as compatible as possible with the MAC satellite broadcasting standard so that the audio of satellite broadcasts can be conveniently encoded in a format suitable for TV.

Based on the above, the BBC has decided to use NICAM coding, which requires a transmission speed of 728 kbit / s for the transmission of two 15 kHz audio channels. This can be used for either two separate voices or one stereo broadcast.

A four-phase modulated subcarrier with a frequency of about 6.55 MHz is used to transmit this 728 kbit / s data. With the 8 MHz channel division in use in England, this auxiliary carrier 3 75702 wave propagation is possible without interference in the reception of neighboring channels.

2. The Dolby system studied by Swedish and Finnish broadcasters.

The main goal of the mentioned broadcasting companies, and especially of the Finnish Broadcasting Corporation, is to produce bilingual TV broadcasts and, secondarily, also stereo sound.

The Nordic countries use a 7 MHz channel division for VHF and an 8 MHz channel division for UHF. However, according to the BBC system described above, the bit rate of 728 kbit / s has been found to be too high to accommodate the 7 MHz channel split, as it causes too much interference to neighboring channels, at least when using older TVs. For this reason, Finnish and Swedish broadcasters have studied the ADM (Adaptive Delta Modulation) system developed by Dolby. This allows two high-quality digital audio channels to be transmitted at a transmission rate of 512 kbit / s on a subcarrier of 5.85 MHz. However, the disadvantage of the ADM system is its coding deviating from the C / D / D2-MAC satellite standard. Therefore, in common antenna and cable systems, the sound of satellite transmissions would first have to be detected on the audio frequency and further coded into ADM, in which case 1 ADM encoder is needed per mobile satellite channel. Due to the high cost of the ADM encoder, this means that only really large cable companies have access to such purchases.

The object of the present invention is to provide such a method in which it is possible on a transmission channel to transmit stereo and two-tone transmissions that meet the quality requirements at a transmission speed (e.g. 546 kbit / s) that this is also suitable for a 7 MHz channel division.

In order to achieve this object, the method according to the invention is characterized by what is stated in claim 1. When the invention is applied in the case of receiving and further modifying a signal transmitted by a satellite for transmission to a distribution network, the invention proceeds as claimed in claim 8.

It is therefore essential in the invention that at some point in the transmission interval the stereo channels are converted in a manner known per se into the sum of the right and left channels (R + L) to the difference (RL), respectively, and that the bandwidth of the separation channel is limited considerably less than the sum channel bandwidth. uses a significantly lower sampling rate when digitizing the separation channel. In this case, the combined multiplexed signal requires a correspondingly lower transmission rate, i.e. a lower bandwidth. Preferably, the separation channel bandwidth and sampling frequency may be half the sum channel bandwidth and sampling frequency.

In a preferred embodiment, the so-called C / D / D2-MAC satellite standard developed by the BBC is used. NICAM coding. It has been found that when a sampling rate of about 32 kHz is used for the sum channel and half the sampling rate for the separation channel, stereo sound can be transmitted at a transmission rate of about 550 kbit / s. This differs very little from the 512 kbit / s transmission speed of ADM studied by the Finnish Broadcasting Corporation, and it can be transmitted on the same subcarrier of about 5.85 MHz also on the 7 MHz channel division. As already stated, the use of the sum channel (R + L) and the separation channel (R-L) is already known mainly for radio transmissions. When this is now applied in digital form to stereo TV broadcasts in accordance with the present invention, while using half the sampling rate (16 kHz) for the separation channel, only about 3/4 of the original 728 kbit / s transmission rate provided by the BBC is needed to transmit a stereo broadcast.

Il 5 75702

When, in reception, the matrix again separates the channels into the left and right channels, the frequency repetition is again direct up to 15 kHz, provided that the transmission path is phase linear. The system according to the invention also does not cause deterioration in terms of distortion or signal-to-noise ratio. The only mild drawback (theoretically) is the lack of channel separation in the frequency range of about 7.5 ... 15 kHz. On the other hand, it is known that most of the stereo effect perceived by humans occurs in the frequency range of 300 to 7000 Hz. In the system according to the invention, the frequency range in which excellent channel separation is achieved is 0 ... 7 kHz. If necessary, the spatial effect could, of course, be artificially further improved at frequencies exceeding 7 kHz, for which known methods exist.

It is further noted that matrixing is not used in two-tone transmission, in which case one 15 kHz audio channel and one 7 kHz audio channel are available. These then correspond in coding to the HQ (high quality) and MQ (medium quality) channels of the C / D / D2-MAC standard proposal. The HQ channel is useful for good sound reproduction and the MQ channel for good speech reproduction.

The invention and its other features are described in more detail below in the form of a few examples and with reference to the accompanying drawing, in which Fig. 1 shows a block diagram of an encoder and transmitter for applying the method according to the invention, and Fig. 2 shows a receiver and decoder for applying the method.

In Fig. 1, the letters L and R on the left denote analog audio signals of the left and right stereo channels. From these, the sum R + L of the channels and the difference R-L of the channels are formed by the matrix 1. Reference numeral 2 denotes a phase linear low-pass filter having a cut-off frequency of fc, and correspondingly, reference numeral 3 denotes a phase-linear low-pass filter with a cut-off frequency of fc / 2. fc can be e.g. about 15 kHz.

Reference numerals 4 and 5 denote analog-to-digital converters (PCMs) in which the sum channel, respectively, is sampled from a separation channel at a certain sampling frequency fn, respectively fn / 2. As already stated above, the sampling frequency fn may be e.g. 32 kHz. Reference numeral 6 means the so-called a parallel-to-serial converter in which the digitized signal is converted to a suitable serial format for further transmission to a multiplexer 8. The multiplexer 8 is also provided with other data to be multiplexed 7 (frame data, spectrum spreading data, etc.). The multiplexer 8 is controlled at a clock frequency fk and the multiplexed data 9 is transmitted to the modulator and the transmitter 10 at a transmission rate of about 550 kbit / s.

It should be noted that the matrixing and digitization of the information R and L can also be performed in the reverse order, whereby the bandwidth limitation of the separation channel takes place in digital form.

Figure 2 shows a decoder and receiver suitable for receiving a transmission. The digital multiplexed audio signal is received and expressed in block 11. The clock frequency fk is regenerated from the signal and the sum and separation channels are separated by a demultiplexer 12. The data from the demultiplexer is converted to a parallel , in which case half of the sampling frequency fn / 2 is used in the difference channel.

Reference numerals 16 and 17 denote phase-linear low-pass filters with cut-off frequencies fc fc / 2, respectively. The analog sum and separation channels obtained from the filters are matrixed in the matrix 18 in a manner known per se to output the right and left stereo audio channels.

Il 7 75702

Filters 16, 17 remove the sampling frequency and higher frequencies from the audio signal. As on the transmitting side, the order of digitization and matrixing at the receiver can also be the opposite of that described, i.e. the matrixing takes place in digital form before the digital-to-analog conversion.

Claims (10)

A method for performing stereo audio TV transmission, wherein on the transmission side, the information information of two audio channels is digitized and digital samples are taken at a certain sampling frequency and transmitted after time multiplexing at a certain transmission rate to receivers in which the channel information is again separated, analogized combining the stereo audio channels in a manner known per se into the sum of the right and left channels (R + L) and the difference (RL), respectively, that the channel separation is sampled at a substantially lower sampling frequency (fn / 2) than the channel sum; L) and the difference (RL) at the sampling frequency again to the right (R) and left (L) stereo audio channels. Method according to Claim 1, characterized in that so-called NICAM coding is used in the digital coding. Method according to Claim 1, characterized in that the sampling frequency of the difference signal (R-L) is half the sampling frequency of the sum signal. Method according to Claim 1, characterized in that digitization is performed on the transmission side after the channel sum (R + L) and the separation (R-L) have been formed. Method according to Claim 4, characterized in that the frequency band of the sum signal is limited to about 15 kHz and the frequency band of the difference signal to about 7 kHz by means of a low-pass filter (2, 3). Il 9 75702 Method according to Claim 1, characterized in that the sum and difference signal (9) combined by multiplexing on the transmission side are modulated and transmitted at a transmission rate of approximately 550 kbit / s. A method according to claim 1, characterized in that on the receiving side, after demultiplexing, the sum and separation channels are first converted to analog form and then the signals are matrixed to form a right (R) and a left (L) stereo audio channel. An encoder for performing stereo sound TV transmission, comprising matrix means (1) for generating a difference (RL) of the sum of the right and left channels of stereo sound (R + L), respectively, and filter means (2, 3) for limiting the separation channel bandwidth substantially less than the sum channel bandwidth, characterized in that it comprises sampling means (4, 5) for taking digital samples from the sun channel at a certain sampling frequency and at a sampling frequency substantially lower than the separation channel, parallel series converters (6) for converting sum and difference channel digital signals to serial and for further delivery through the modulator to the transmitter. An encoder according to claim 8, characterized in that the sampling frequency of the sum channel is e.g. 32 kHz and the sampling frequency of the separation channel is e.g. 16 kHz. A receiver decoder for receiving a stereo audio TV transmission, comprising matrix means (18) for matrixing the sum and difference signal (R + L, RL) into stereo right and left channel signals (R, L), characterized by a combination comprising a detector (11) for detecting multiplexed serial digital signals and clock frequency (fk), for separating the detected signals from the demultiplexer (12), for converting the demultiplexed signals into parallel format, serial / parallel converters (14) for converting the demultiplexed signals into parallel format, and digital-to-analog converters wherein the digital-to-analog converters (15) are located before or after the matrix members (18).