JPH07504539A - Simultaneous transmission method of signals from N signal sources - Google Patents

Abstract

A process is disclosed for simultaneously transmitting signals from N-signal sources over a corresponding number of transmission channels. The individual signals are subdivided into blocks and the blocks are converted into spectral coefficients by transformation or filtering, and the latter are then subjected to a data reduction process. The invention is characterized in that the blocks belonging to the individual signals are subdivided into section. The momentary sections of all signals are processed together, the admissible disturbance is determined for each section by using a perception-specific model and the momentarily required total transmission capacity is calculated. The allocation of the maximum available transmission capacity for each individual signal is calculated from the total available transmission capacity and the total momentarily required transmission capacity. Each individual signal is coded and transmitted with the thus determined capacity.

Description

[Detailed description of the invention] name of invention Method for simultaneous transmission of signals from N signal sources Technical field The present invention synchronizes signals from N signal sources through a corresponding number of transmission channels. The present invention relates to a method of time transmission.

Conventional technology An individual (time) signal is divided into blocks, and the blocks are transformed or is converted into a spectrum coefficient by filtering, and the spectral coefficient is Department.

undergo a data reduction process per minute or each coded by data reduction. There are known methods to do this. For example, Jürkufbert (Joe) on studio techniques Perceptual speech coding (perceptual speech coding) Audio-Coding) J or Elrad Steph y-- -Article “Data Diet” by Stefanie Renner In other words, data organization in digitized audio signals (Datan-Di aet.

Datenreduction bei digitalisierten A (1991). An overview of these The article and PCT publication number WO 88101811 are expressly set forth in this application. Not all terms and method steps are specifically cited for explanation.

In many cases, a corresponding number of transmission channels are used to simultaneously carry signals from a large number of signal sources. It is necessary to transmit the data via

The simplest example for this is stereo signals via two transmission channels. transmission.

When transmitting signals from N signal sources through the corresponding number of transmission channels , there is a problem of determining the size of the transmission channel.

Each transmission channel is By sizing each channel, on average a relatively large amount of transmission capacity is left unutilized. It will remain in use.

When transmitting signals from a large number of signal sources through a corresponding number of transmission channels In order to design the transmission channel only for "average demand" and to balance short-term increased demand on one channel by allocation from other channels. It is known from digital telephone technology. In that case, the assignment is exclusively for the signal This is done using statistical tables.

Regarding the prior art, the following documents were referred to. Namely, Dr. H. Gerhauser ( Dr. H. Gerhaeuser) “Digital prediction control word classification” Linguistic interpolation method ionsverfahrenmit praediktionsgesteue Rter (Wortaufteilung) J (1980), R. Woitowitz's [linguistic interpolation method by instantaneous priority assignment (ein)] digitalessprachinterpolationsverfahr en mit momentanerprioritaetszuteilun g) J (1977) or G, G, Kranerbucher (K1) ahnenbucher)'s ``Digital Language Interpolation with Block Preferential Assignment'' Law (ein digitales 5 prachinterpolations verfahren mitblockweiser Prioritaets Reference was made to Zuteilung J (1978).

Fluctuations in demand when transmitting a large number of signals over a corresponding number of transmission channels The normal digital telephone technology method for balancing the If the source signal has undergone data reduction in advance, for example by the so-called OCF method, is known to produce no negative results. The present invention aims to solve such problems. It is something.

Description of the invention The object of the invention is to transmit signals from N signal sources via a corresponding number of transmission channels. A method for simultaneously transmitting signals, the transmission channels having a signal capability of Simply "average" without causing any perceptible loss, e.g. The size of the data is determined only by the demand, and the data-organized signal is used for the transmission. The purpose is to provide a method for transmitting data over a channel.

Means of achieving this object of the invention are set out in claim 1. Improvements of the invention are dependent This is the subject of the section.

In the present invention, signals are transmitted simultaneously from N signal sources through a corresponding number of transmission channels. Individual signals from a statistical point of view when balancing fluctuating requests while being transmitted. Rather than assigning Based on the idea of balancing fluctuating demand through appropriate means in the manufacturing process. Based on.

The basic idea of this invention will be explained in detail with reference to embodiments shown in the drawings.

FIG. 1 is a block diagram for explaining the method according to the invention.

2a and 2b are diagrams showing the structure of signals in the present invention.

In the method according to the invention, individual signals are divided into blocks and The locks are converted into spectral coefficients by transformation or filtering. fluctuate In order to balance the demand for and each current section of all signals is processed simultaneously.

This is illustrated by the corresponding "functional blocks" in FIG.

Using a perceptual specific model, for example, a psychoacoustic model when transmitting an audio signal, The tolerable interference for each section is determined and based on the currently required The total transmission capacity required is calculated. Calculation of such total transmission capacity, i.e. The required number of bits is calculated simultaneously for all blocks. currently available From the total transmission capacity and the total transmission capacity currently required, the current An allocation of the maximum currently available transmission capacity is calculated. Bits assigned to each signal The individual signal is encoded according to the number of signals, and the transmission of this individual signal is performed. be exposed. In the simplest case, it is possible to balance or equalize the individually required transmission capacities. Equalization is performed only between channels.

The improvement as claimed in claim 2 provides a storage location for the transmission capacity, i.e. a so-called bit storage. If the total transmission capacity required exceeds the average transmission capacity available, then If the transmission capacity is met, the transmission capacity is allocated.

This bit repository is used when the requested transmission capacity is less than the available transmission capacity. The conditions are always satisfied (Claim 3).

In either case, the transmission capacity is If the transmission capacity is much smaller than the available transmission capacity, It is necessary to forcefully allocate a set (Claim 4). This forced allocation is performed for each signal source only on channels that signaled a demand greater than the average demand. It is desirable to A demand that is significantly larger than the average demand is This means that encoding the signal is quite difficult.

In any case, according to claim 9, all blocks are coded separately from the signal source. It is desirable that the signal be formed from all the signals that have been This entire block is A fixed area that contains information that can be determined, and a flexible area that accepts encoded signals. It consists of a large number of regions of various lengths. This is illustrated in Figure 2a.

Furthermore, identical input signals are recognized and transmitted only once using the appropriate transmission format. By doing so, transmission capacity can be saved (claim 6). This means that This is illustrated in Figure 2.

In either case, accurately determine or simply estimate the currently required transmission capacity. (Claims 7 and 8).

Furthermore, the method according to the invention can be carried out in parallel to a considerable extent.

For this purpose, according to claim 11, the coding of the individual signals is already performed on each signal. Advantageously, this is done during the calculation of the transmission channel allocation for the transmission channel.

A further advantageous embodiment of the basic concept according to the invention is specified in claim 11.

If the required transmission capacity exceeds the available transmission capacity and is allocated from bit storage If this does not occur, ensure that the requested transmission capacity does not exceed the available transmission capacity. It is possible to increase the allowable interference value for all signals (claim 11).

A numerical example of the method of operation for audio signals is given below. Furthermore, the basics of the present invention The idea is not limited to audio signals, but rather video signals or intellectual It goes without saying that other signals that undergo sensory evaluation can also be processed in the same way. Ru.

Examples of possible ways of operating the audio signal are as follows.

Assume that y(t) is the sampling value of the audio signal.

1) Audio signal y is decomposed into sampling values (y (t)) using a known method. digitized. Digitized sampling values are blocks of length 2n In the selected embodiment, an overlap with n overlaps is decomposed into It is a block that works.

x(k, b)=y(b*n+k) Here, k=0. ．． 2n (b block number) 2) Each block of length n is , the spectral coefficients by a transform, e.g. fast Fourier transform or cosine transform. is converted to

x (j, b) = SUM (1 = 0., 2n; x (1, b) ・f (1) ・cos(pi ・(21+1+n)(2j+1)/(4n))) where j=0 ．． , n and f(1)=S(lrt(2)・5in(pi・(1+0.5) / (2n)) 3) Each block is divided into sections, and each section The energy density for the tion is calculated.

E(i,b)=(SUM(k=a(f)+1.,a(i+I　X(k,b)A2 ))/(a(i+1)-a(i)) where i=1. ,,c.

In this case, the coefficient a(i) was taken from Table 1 below.

4) For each section, by an appropriate psychoacoustic model (for this (see above-mentioned literature) allowed interference was calculated. Allowable interference to between bands Masking is obtained.

T(i, b)=MAX(k=1.,,i-1;E(k,b)・z(i- k)) Band masking is expressed by the following equation.

s(i, b) = max(E(i, b)・e(i), T(i, b)) In addition, masking between blocks is expressed by the following formula.

5s(i, b) = max(s(i, b-1)/16.s(i, b)) next Then, for each block, the required number of bits is calculated.

5) Calculation of the required number of bits for a block a) 0CF (Fuchman code Regarding encoding such as in the case of p=po+sUM(i=1.,c; (a(i+1)−a(i)(s(i,b) /5s(i,b)))b) About PCM encoding (SNR=6dB/bit) For each section, the scale factor and number of bits per sampled value are added. Transmitted as information.

p=po+sUM(i=1.,c; (a(i+1))40/6・log(E( i,b)/5s(i.

b))) Below, in table form, the valid values for the individual numbers and constants are shown.

n=512 C=23 For pO=12000CF (average number of bits per block) pO=345 P Regarding CM (scale coefficient: 10 bits/section: number of quantization stages) Encoding: 5 bits/sexy ) Table 1 i 1 234567891011121314a(i) 0 4812162 02428323640465260i 15 16 17 181920 2 1 222324a(i) 70 82961141361641982402 96372 table 2 i1234567 e(i) 1e-11e2 1e3 1e4 1e5 1e6 1e-7e(i ) Ie-81e-9a (1) 0 where i>9 Table 3 i123456 e(i) 0.00040.00040.00040.00040.00040 ．． 0004i 7 8 9 10 11 12 e(i) 0.0004 0.0020.0020.0020.004 0.0 1i 13 14 15 16 17 18e(i) 0.015 0.025 0.04 0.06 0.06 0.06i 19 20 21 22 23 e(i) 0.08 0.08 0.11 0.14 0.18 Following this, An assignment of the number of bits to the individual signals is made. For this, the sign of the input signal is k (k)-bits are required, while the number of available bits is psoll Assume that

psum=SUN (p(k)) A case-by-case distinction is necessary here.

1) When psum=psoll Each signal receives the requested number of bits.

That is, z(k)=P(k) 2) When psum<psoll Each signal receives more than the requested number of bits.

z(k)=(psoll/psum)'p(k) For example, K=2. psol 1=1600. p(1)=540. p(2) = 660 psum = 1200 z(1) = 1600/1200・540=720 (180 bits or more) z(2 )4600/1200・660=880 (220 bits or more)3) pso When ll>psum Each signal receives no more than the requested number of bits.

a) About OCF: z(k)=(psoll/psum), p(k)b) Regarding PCM: For each signal, the minimum number of bits must then not be less than:

z(k)=pO+((psoll-K・pO))・(p(k)-po) For example , K=2. psoll=1600. po=500.1)(1)=600゜p (2)=1200 At this time, psum4800 z(1)=500+(1600-2・500)/(1800-2・500)・( 600-500) = 575 (25 bits or less) z(2)500+(1600-2・500)/(1800-2・500)・(1 200-500) = 1025 (175 bits or less) To correct for tolerable interference, p bits are required for each signal but are not allocated. In the case of 2 bits, the following case classification is required.

1) If the number of allocated bits is equal to the number of requested bits: No modification necessary.

2) If more bits are allocated than requested: About OCF: No modification necessary.

About PCM: The number available for quantization of each section is increased by (z-p)1512.

3) If fewer bits than requested are allocated: About OCF: 5s(i, b) = s(i, b) + (z-pO) / (p-oO・(s s(i, b) - s(i, b)) where p>p.

5s(i, b) = s(i, b) Here, p<=pO About PCM: In each section, the number of bits available for quantization is (z -, p) 151 Increased by 2.

For PCM, it is necessary to make the bits per ATW an integer approximation. this To do this, all bits/ATW are first rounded down to the next lowest integer, The resulting bit sum is then determined.

If there are still bits available, start with the lowest band and continue with the available bits. For each band until the number of bits is reached, one more bit is added in the first estimation run. The port/ATW is placed in a usable state.

example: 104 bits are available.

Assignable: 10 bits The present invention has been described above using preferred embodiments. in a comprehensive inventive concept Of course, various modifications are possible.

That is, employing a fixed total block length while filling the bits that are in use. is also possible, and transfers to encoders that have not yet been completed may also be employed.

Furthermore, we have developed a flexible block that performs time averaging after specifying the maximum block length. It is also possible to adopt the lock length.

international search report PCT/DE 92100905 Continuation of front page (72) Inventor: Dieter Seitzel, Federal Republic of Germany, Day-91054 Alangen Humboldaustraße 14 (72) Inventor Sporat trout Federal Republic of Germany Day-90766 Fattou Wilhelmshalf Ener Strasse 29

Claims (11)

[Claims] 1. Simultaneous transmission of signals from N signal sources through a corresponding number of communication channels. A method in which the individual signals are divided into blocks, the blocks being transformed or filtered. The spectral coefficients are converted into spectral coefficients by data sorting. In the method of processing, blocks belonging to the individual signal are divided into sections. , Each current section of the entire signal is processed simultaneously and the allowance for each section is The tolerable interference is determined using a perceptual specific model and all currently required The transmission capacity is calculated, The allocation of the maximum transmission capacity that can be used for each individual signal is the total available transmission capacity. and the total transmission capacity currently required, characterized in that each of said individual signals is encoded and transmitted with said determined capability; A method for simultaneous transmission from N signal sources. 2. If the total requested transmission capacity exceeds the available average transmission capacity. , characterized in that the transmission capacity is stored by a bit repository in which the allocation is made. , the method according to claim 1. 3. If the requested transmission capacity is less than the available transmission capacity, the 3. A method according to claim 2, characterized in that the storage is filled. 4. If the requested transmission capacity is much less than the available transmission capacity, the A forced allocation occurs to avoid the storage space becoming too large. 4. The method according to claim 3, wherein the method comprises: 5. Specify that the forced allocation occurs only if there is a demand greater than the average demand. 5. The method according to claim 4, wherein the method comprises: 6. The same input signal is recognized and transmitted only once using the appropriate transmission format. 6. A method according to any one of claims 1 to 5, characterized in that: 7. Claims 1 to 3, wherein the currently required transmission capacity is accurately determined. The method described in any of Section 6. 8. the determination of the currently required transmission capacity is merely estimated; A method according to any one of claims 1 to 7. 9. The entire block is formed from the signals from the signal sources, encoded separately, and the entire block is The lock consists of a fixed part containing information that can determine the division of the individual signals, and a flexible part. Claim 1, characterized in that it consists of a large number of regions having a length of The method described in any of paragraphs 8 to 8. 10. The encoding of the individual signals does not require calculation of transmission capacity allocation for each signal. Claims 1 to 9, characterized in that the invention has already been made while Any method described. 11. If the requested number of bits exceeds the total number of available bits, , the tolerable interference for all signal sources increases and the bit requirement decreases. 11. A method according to any one of claims 1 to 10, characterized in that: