DE2718631A1 - Compander system for analog signals - uses compressor at transmitter and corresponding expander at receiver with digitised transmission path - Google Patents

Abstract

At the transmitter the analog signals pass through a Dolby analog compressor (DKOM) and then through a low pass filter (TP) which limits the band-width. The signals are then treated in an analog/digital converter (A/D), quantized at 12 bits and encoded. The digital signals are then compressed from 12 bits to 10 bits in a compressor (MK) and the compressed digital signals are then transmitted. At the receiver the 10 bits signals are expanded to 10 bits in an expander (ME), then reconverted to analog signals in a digital/analog converter (D/A) and filtered by a low pass (TP). The analog signals pass through a Dolby expander (DEXP) and can then be reproduced.

Description

Process for the digital transmission of high quality audio signals The invention relates to a method for the digital transmission of high quality Sound signals based on the principle of dynamic limitation on the transmission side and dynamic expansion on the reception side. Such a method can save transmission time, Transmission bit rate or storage capacity can be used. It creates a synchronous Output bit rate with constant code word length and can therefore without buffer storage can be used in conjunction with digital transmission channels.

For the economical transmission of high quality audio signals with bandwidths from 40 Hz to 15 kH; are different types of PCM processes today with digital companding known. A basic block diagram is indicated in FIG. 1. The analog audio signal first runs through a pre-emphasis, is band-limited by a low-pass filter, in an analog / digital converter with a Frequency of 32 kiiz sampled and with 16384 steps, corresponding to 14 bits per sample, linearly quantized and encoded. The digitized audio signal is in a compressor compressed from 14 to 10 bits per sample and then via the transmission channel transfer.

In the receiver, the 10-bit signal is converted into a digital expander 14 bits per sample expanded. After the reconversion in the digital / analog converter and subsequent low-pass filtering, the analog audio signal is de-emphasized and can then be played back. To compand the digital audio signal in the transmission method according to (1) Hessenmüller, H. "The Transmission of Broadcasting Program in a Digital Integrated Network "Proceedings of the" 1972 Zurich Seminar " an instantaneous value compander is used, which is based on a logarithmic 13 segment A characteristic is working. Two more methods according to (2) C.C.I.R. 1974 - 78 Doc. CMTT / 42 - E, and (3) Osborne, D.W. and Croll, M.G. Digital sound signals; bitrate reduction using an experimental digital compander BBC RD 1973/41, work on the principle of linear block companding.

The Compenseur Preeque Instantane according to (2) sums up the sending process in each case 32 consecutive samples linearly quantized with 14 bit to form a block together. According to one of 5 linear 10-bit characteristics, their quantization levels differ in size by a factor of 2, all sample values are of a block uniformly quantized depending on the largest amplitude occurring in it and coded. A scale factor of 3 bits divides the receiver at the beginning of each block in addition, which characteristic has been switched on on the transmit side. The scale factor and other information is stored in the coder at the point of the least significant bit (LSB) every second 10-bit code word to be transmitted is added. After the above The principle described also works the Near-Instantaneous Compander (3) except for the deviations listed below. At the transmitter end, the samples are sent in Analog / digital converter only quantized linearly with 13 bits, so that between 4 linear 10 ut characteristics must be switched.

In a multiplexer, a 2 bit scale factor is used for each one to be transmitted Preceding code word block. For marking the higher quantization background noise an analog "edither" signal is added in the analog / digital converter.

The transmission system with instantaneous value companding according to (1) is shown is characterized by a high signal-to-noise ratio at low levels and by a constant code word length. The transmission of the digitized audio signal requires only a word synchronization.

Compared to the method according to (1), the transmission systems achieve with block companding according to (2) and (3) a larger signal-to-noise ratio high levels. The need to transfer a scale factor leads either to a loss of the signal-to-noise ratio (2) that can in principle be achieved, or to a not constant code word length (3). As a result of the block formation, the transmission of the digitized audio signal, in addition to word synchronization, block synchronization is also performed necessary.

The task leading to the invention was to specify a method at least one for the digital transmission or storage of audio signals Dynamics of 79 dB achieved using analog / digital and digital / analog converters whose resolution with linear quantization does not exceed 12 bits per sample is.

The method should have a synchronous output bit rate with a constant code word length of not more than 10 bits per sample and for transmission or storage of the digitized audio signal only a word synchronization but no block synchronization require. The method is intended to make it possible to record compressed analog data To transmit signals directly digitally without expansion or to save them digitally.

The object is achieved as described in claim 1.

Appropriate arrangements are described in the subclaims.

This hybrid compander according to the invention uses the advantages of instantaneous value companding and brings out the favorable properties of an analog compander. In order to it succeeds compared to the solutions described in (i), (2) and (3), a favorable To achieve the signal-to-noise ratio over the entire modulation range. Further advantages are that no additional switchover information is transmitted mun and a block synchronization for the transmission is not necessary, then inexpensive analog / digital and digital / analog converters are used in the implementation with a linear resolution of only 12 bits per sample value.

The difference to the suggestion made in (1) is that dan in (l) the total companding gain from a digital instantaneous value companding is achieved with a 13 segment characteristic according to the logarithmic A law, while In the invention described here, the companding gain is made up of two parts composed. The first part is generated in an analog Dolby A compander. The second part is determined by a digital instantaneous value compander with a 9 segment characteristic after the logarithmic A law achieved.

In contrast to the suggestions made in (2) and (3) the invention is not block companding.

The basic interaction of the components of the flybrid compander results from Fig. 2. At the transmitter end, the analog audio signal first passes through a Dolby A compressor. The analog compressed signal is band-limited by a low-pass filter, sampled in the analog / digital converter, linearly quantized with 12 bits per sample and coded. The samples of the digitized audio signal are stored in an instantaneous value compressor MK compressed from 12 to 10 bits and then transmitted. On the receiving side, the 10 bit code words expanded to 12 bit in an instantaneous value expander ME. After Reconversion in the digital / analog converter and subsequent low-pass filtering goes through the analog audio signal has a Dolby A expander and can then be played back. The digital instantaneous value companding works according to a 9 segment characteristic. In Fig. 3, the characteristic part is shown for positive code words. The negative one Part of the characteristic is in the 3rd

Quadrants of the coordinate system. It is based on the part of the characteristic curve shown by a 180 ° rotation around the origin of the coordinates.

As is to be shown using an example in FIG. 4, this results in Interaction of the Dolby A compander with the 9 segment instantaneous value compander favorable signal-to-noise ratio curve over the entire control range. In the diagram of FIG. 4, the signal-to-noise ratio of the method is shown for comparison plotted against the relative signal level according to (1). It denotes 0 dDr the maximum permissible level of a sinusoidal signal at the input of the analog / digital converter before the Use of limitation.

Compared to the 13 segment characteristic used in method (1) the 9 segment characteristic curve has an initial steepness that is 4 times lower. For the The digital instantaneous value companding used in the invention results therefrom a 12 dB lower gain at low levels. The associated loss The signal-to-noise ratio is increased by the gain of the analog Dolby-A-Sompander up to balanced to a residual of 2 dB.

At a high level, the 9 segment characteristic is steeper than the 13 segment characteristic curve, so that compared to method (1) that in the invention used instantaneous value companding at high levels a signal-to-noise ratio that is 2.5 dB greater achieved, which is further improved by the gain of the Dolby A compander. The Dolby win depends on the signal statistics in the area of higher levels, what themselves explained from the functionality of the Dolby A system.

In the Dolby A processor, the entire frequency band is divided into four sub-bands split up. A separate level control works in each sub-band, so that the Dolby A system can be caught as a controlled filter, its characteristics constantly adapts to the spectrum of the signal and a white spectrum for the dolby Tried to adjust the signal. The resulting filter will through the invention exploited, on the one hand, an advantageously large signals-to-noise ratio to achieve and on the other hand to influence the quantization noise subjectively favorably, that is created by the digital instantaneous value companding.

The filter gain of the Dolby A compander at levels above about -45 dßr is significantly higher for audio program signals than the illustration in Fig. 4, in which the Dolby level was selected as -6 dBr. The one shown there Course of the Dolby gain, taken from (4) K. Bertram, dynamic improvement with the Dolby Stretcher, TV and Cinema Technology 24/1970, Issue 4, only takes into account the gain that occurs when the same level occurs in each sub-band.

The signal-to-noise ratio of the entire hybrid compander, computer-aided measured using one proposed by the European Broadcasting Union (EBU) Test signal 'Violinc' is plotted against the signal level in FIG.

Because of its high peak factor, the violin signal is at levels above - 15 dBr no longer free of limiting noise. At these levels Determined measured values are shown in the diagram of FIG. 4 with dash-dotted lines.

Fig. 6 shows a compander characteristic of the Dolby A compander like him in "Fernseh- und Kino-Technik", special print from volume 24 (1970), issue 4, page 123 to 126 is described.

Fig. 7 shows a compander characteristic of the Telcom C4 compander, such as it is described in the "Funkschau" 1975, issue 18, p. 571, by Jürgen Wcrbuth, with which even more favorable results can be achieved than with the Dolby A compander.

The basic principle for generating such a compression characteristic with a 33% gradient over such a large area can be explained as follows: There will be three identical, controllable amplifiers connected in series, the control inputs are operated in parallel.

A regulation ensures that at the exit of the last Amplifier the level remains constant, no matter what the input voltage of the first amplifier.

This method has the advantage that it is in the work area of interest - due to a high negative feedback of the Regel systems - works exactly. In addition, the control range of each of the three amplifiers is one third of that Reduced input dynamics, which results in favorable requirements for realizing this Let Schaltullg derive. The recovery of the original dynamics in the expander works in a complementary way. Since the Telefunken compander Telcom C4 with vicr from each other independent control systems works - each system is a specific frequency band assigned - the breathing tape noise and the modulation noise of the tape self suppressed.

The explained instantaneous value companding with 9 segment characteristic for the liybrid compander according to FIG. 2, a logic switching network can be used on the transmitter side for coding (coder) and on the receiver side by a logic switching network for decoding (Decoder). As shown in Fig. 5, the encoder assigns each code word x12x11x10x9x8x7 x6x5x4x3x2xl one code word each at the output of the analog / digital converter y10y9y8y7y6y5y4y3y2y1 to which is transmitted. The decoder assigns each received Code word y10y9y8y7y6y5y4y3 one code word each w12w11w10w9w8w7w6w5w4w3w2w1w0 to, which is fed to the digital / analog converter. The bit w carries 0 no information; it is only used to display representative values in the Sign. & Magnitude code.

Tables 2 and 3 give a possible coding or decoding rule to the Realization of the 9 segment instantaneous value companding if analog / digital and Digital / analog converters with Sign & Magnitude code can be used.

The coding takes place differently in the individual segments of the characteristic curve.

To distinguish them, the segments are numbered as in FIG. 3. That second half segment of (5) and segments (6) to (9) - not shown in FIG. 3 - form the negative part of the characteristic. The segmentation given in Table 1 uses the zero point symmetry of the 9 segment characteristic. The segments (1) and (9), (2) and (8), (3) and (7), (4) and (6) are combined in pairs. For every couple only one coding and decoding rule is required from the table 2 or 3 can be read in the line given by the segment. Segments 1 which form a pair can be distinguished from one another using the sign bit xi2.

Positions of the code word x12x11x10x9x8x7x6x5x4x3x2x1 at the encoder input, the information of which is lost during coding, are highlighted in table 1 with hatching. rooo with 12> - Xg X1 X7 X5 xS J x3 x2 x1 y10 Yg V1 o 5s3 "21 12 11 10 oo 3 2 1 N oo 00000 000 5 Æ OOOOL xxx xxx> ~ xxxy 0 0 0 0 L yyyyyy L 000L0 SS><>^>~> O0OLL 0 LLO 000LL 6>><>> 3 "OJO zo CO O0LOL L00L OOOJ >> x12 oooo J oo J os> -1 o -1 0 0 0 0 0 0 0 - 0 0 5 OLL 0 OOOOOOJJ JLOL xLLLCLCx2x7x5xx ° LL.> y3 21 L o 0 0 12 ic LOL LLLO LOL 1 9 x12 L xN xn xw xx > n xn x- xi x xb «N {D b > xxxxx > XXXXK > - XZ OJOJOJOJOJ > ZOOJJOOJJOOJJ >OOJJOOOOJX> J Xn OOOOOJJJJJJXJ o PYY> N > xxxxx XN XN Xn me7 XC X ~ X ~ X- Xw Xw S vn «n 0 0 0 ~ XXXKXX UZ. ZD 0 40 XXXXXX Xo OJOJOJ Xb Xb X XOOJJOOJOJXX « Xo OOOOJJJOOJOJ Xç XOOOOOOOJJJOOJ X- OOOOOOOOOOJJ XY wY tY X- XW X- 'approx Table 1 Segmentation Table 2 Coding Table 3 Decoding rule regulation x12 x11 ... x1 input code word of the 9 segment compressor shown in y10 y9 ... y1 code word to be transmitted # Sign + magnitude code w12 w11 ... w0 output code word of the 9 segment expander in representative value representation L eerseite

Claims (4)

P a t e n t a n S} rü c h e 0 Process for digital transmission of high-quality audio signals based on the principle of dynamic compression on the transmitting side and the dynamic expansion on the receiving side, characterized in that that on the transmission side the analog signals are compressed with an analog compressor that the required for a given signal dynamics to be transmitted Resolution of the analog / digital converter is reduced that the digitized signals additionally to a minimum transmission bit rate with a digital instantaneous value compressor be compressed so that on the receiving side with a corresponding digital instantaneous value expander, Digital / analog converter and analog expander return the dynamic range to its own original value is stretched and that the analog companding gain for enlargement the distance between signal power and quantization noise power of the digital Instantaneous value companding is used. 2. The method according to claim 1, characterized in that an analog Compander is used, its linear gain characteristic in the range smaller Level is so deformed that in the compressor small levels are raised and in the Expander are lowered in a complementary manner and that towards the middle level range Deformation decreases and large levels remain unchanged (Fig. 6). 3. The method according to claim 1, characterized in that eiii more analog Compander is used, whose linear gain characteristic is independent of the Aussteurung runs in such a way that the signals in the compressor are raised by 33% and in the Expander complementary to this can be lowered by 33% (Fig. 7). 4. The method according to claim 2 or 3, characterized in that the digital instantaneous value compander based on a 9 segment characteristic curve with the following function curve works: Ax y = (0.75) for 0 # x # 1 / A 1 + hnAx y = (0.75) for 1 / A # x # 1 with A = 21.76 1 + hnA