DK159353B - PROCEDURE FOR THE TRANSMISSION OF DIGITAL INPUT SIGNALS IN A FILM TRANSFER - Google Patents

Abstract

1. A method of converting digital input signals of a light waveguide transmission system from a first binary code comprising 32 different code words each comprising five binary digits corresponding to 32 possible amplitude values into output signals which occur in a second binary code comprising code values comprising six binary digits having a positive and a negative mode, where the mode corresponding to the continuous digital sum occurs at the end of the preceding code word comprising binary digits, wherein the 32 possible amplitude values (0...31) of the input signal are sub-divided into two equal sized sections each comprising 16 code words where the first section comprises the amplitudes values between a lowest amplitude value (0) and a first middle amplitude value (15) and the second section comprises the amplitude values between a second middle amplitude value (16) and a maximum amplitude value (31), wherein the code words for the positive mode of the output signal of the first section are selected to be such that the greatest possible conformity occurs between the input and output signals, and wherein for the lowest amplitude value of the second section the code word (49) for the positive mode corresponds to the inverted code word (14) of the negative mode of the highest amplitude value of the first section, and the code word (49) for the negative mode corresponds to the inverted code word (14) of the positive mode of the highest amplitude value of the first section, characterised in that for the second lowest amplitude value of the second section, the code word (57) for the positive mode corresponds to the inverted code word (6) of the negative mode of the second highest amplitude value of the first section, and the code word (17) for the negative mode corresponds to the inverted code word (46) of the positive mode of the second highest amplitude value of the first section, that the code words for the other amplitude values of the second section are formed correspondingly, and for the highest amplitude value of the second section, the code word for the positive mode corresponds to the inverted code word of the negative mode of the lowest amplitude value of the first section and the code word for the negative mode corresponds to the inverted code word of the positive mode of the lowest amplitude value of the first section, so that a diagonally inverted symmetry occurs between the two sections of the amplitude values of the input signal and the output signal.

Description

DK 159353 B

The invention relates to a method for converting digital input signals into a fiber optic transmission system, which consists of a first binary code of 32 different passwords each with five binary digits 5 corresponding to 32 possible amplitude values, to output signals appearing in a second binary code of passwords. consisting of six binary digits with a positive and a negative mode, corresponding to the current digital sum at the end of the preceding code-10 word with six binary digits, where the 32 possible amplitude values of the input signal are divided into two equal ranges with every 16 code words, the first range comprising the amplitude values between a minimum amplitude value and a first intermediate amplitude value, and the second region the amplitude values between a second intermediate amplitude value and a maximum amplitude value, the positive mode code words of the output signal in the first range being selected thus; that the greatest possible is achieved correspondence between the input signal and the output signal, and where the smallest amplitude value codeword in the second positive mode region corresponds to the negative mode inverse code for the largest amplitude value in the first range and the negative mode codeword to it inverted 25 positive mode passwords for the largest amplitude value in the first range.

The development of attenuated glass types for fiber-optic cables, on the one hand, and the extension of the life of the light-ends, on the other, enables the introduction of fiber-optic transmission systems as cost and energy-friendly alternatives for systems with copper cables. Due to the non-linear properties of the optical transmission elements, digital signals are preferably transmitted - i.e. in binary code - whereby "light" corresponds to near 35, and "no light" corresponds to binary 0, or vice versa. This causes two problems.

2

DK 159353 B

Since intermediate generators usually perform both amplitude regeneration and temporal regeneration of the signals, the transmitted signal must contain a sufficiently large clock information. This is not ensured by any data signal. In addition, the mean value (constant proportion) of the signal must be constant, so that only the alternating proportion must be amplified in the receiver, and the known and constant proportion is first included in the amplitude assessment. Both problems can only be solved by redundancy. Since, in the case of fiber-optic systems, a slight increase in the transmission bandwidth can generally be allowed, a block coding is used in fiber-optic systems at 34 or 140 Mbit / s, where five binary digits in the input signal are converted blockwise into six binary digits in the output signal. In this case, the input signals may also consist of n-block passwords of every five binary digits, correspondingly converted to n blocks of every six binary digits.

20 From NTG-FACHBERICHTE ISSLS 80, Volume 73, Septem ber 1980, pages 31-35, a digital subscriber system for copper lines is known in which six bit words are transmitted, which are generated by encoding five bit words in accordance with the 25 table published on page 32 1. The 32 possible amplitude values in Table 1 can also be divided into two equal ranges of 16 code words, the first range comprising the amplitude values between a minimum amplitude value and a first intermediate amplitude value, and the second range comprising the amplitude values between a second intermediate amplitude values and a maximum amplitude value. Furthermore, the code words for the positive mode of the output signal in the first region are selected so as to obtain the greatest possible match between the input signal and the output signal. Also, for the smallest amplitude value in the second region, the positive mode password corresponds to that

DK 159353B

3 inverted negative mode code words for the highest amplitude value in the first range and the negative mode code word corresponds to the positive mode inverse code for the highest amplitude value in the first range, the β-bit words can each - in particular, consist of three "l" bits and three "0" bits corresponding to a rated value of the directional component of 0 or of four "l" bits and two "0" bits. Since, in the latter case, more "1" bits than "0" bits occur, a normalized value of the DC component of +2 appears. The 6-bit words for the negative mode are formed so that at a rated value 0 of the direct component of the 6-bit for the positive mode, this is taken over to the negative mode, while at a standardized value of the direct component of +2 for the 6-bit word for the positive mode, the inverted 6-bit word applies to the negative mode, so that a normed value of -2 is obtained for this. In a single case, there is an association between 20 5-bit words and 6-bit words which differs from the code schema of the present application.

A method of re-encoding information contained in a binary code is known from DE Publication No. 23 39 868. By this method 25, the n elements of each codeword in the first code are transferred to the first n element spaces of a codeword in the output of the the original order, after which the elements are transferred to the last n element locations in the reverse order.

30 From DE Publication No. 27 47 018, a method for translating binary passwords is known.

In this method, the binary passwords are converted to ternary passwords.

From the journal "Frequenz" 34 (1980) 2, pages 35 45-52, a method of the kind mentioned in the introduction is known. By this known method 4 is transformed

DK 159353 B

likewise of five binary digits consisting of six binary digits output signals which appear in positive or negative mode.

The positive mode then contains only passwords where the number of binary digits 1 is equal to the number of binary digits 0 or one greater, while the passwords appearing in negative mode are so constructed that the number of binary digits 0 is equal to or one greater than the number of binary digits 1. The occurrence of passwords in positive or negative mode is obtained by the continuous digital sum of the six-digit passwords in the output signal formed by setting binary digit 1 to the value +1 and the binary digit 0 to the value -1.

Table 2 of the journal "Frequenz" shows the association between input passwords and output passwords. For conversion, in addition to a serial parallel converter and a parallel serial converter, a suitable clock conversion is required as well as on the receiving side in addition to the actual recoding device a synchronization device. This creates a relatively large need, which can grow further as a result of monitoring devices for said components.

The invention is intended to provide a method of the kind mentioned above, which is suitable for fast light-conducting systems, which can be realized with a relatively low supply and which, in the turnover, entails a small error multiplication factor.

According to the invention, this is achieved by the code word 30 for the second lowest amplitude value in the second region of the positive mode corresponds to the inverse code word for the negative mode for the second largest amplitude value in the first region and the code word for the negative mode corresponds to the inverse code word for the 35 positive mode for the second largest amplitude value in the first region,

DK 159353B

5 digits in the second region are similarly formed and the password of the largest amplitude value in the second region of the positive mode corresponds to the inverse code word of the negative mode of the lowest amplitude value of the first region and the password of the negative mode corresponds to the inverse password of the positive mode for the lowest amplitude value of the first range, so that a diagonally inverted symmetry is obtained between the two ranges of the amplitude values of the input signal and the output signal.

By means of the described division into two areas and the symmetry between the two areas there is an advantageous opportunity to carry out the connection by means of gate circuits and controlled inverter steps.

15 To further reduce the bidding, it is appropriate that in both ranges for the smallest and the largest amplitude value, passwords that are the same for the positive and negative modes are selected.

A preferred embodiment of the method according to the invention is set out in claim 3. In this embodiment, an advantageous reduction of the error multiplication of an average of 1.28 is obtained compared with the method according to the prior art, where an error multiplication factor of 1.41 is obtained.

Further embodiments of the method of the invention are set forth in claims 4-6.

The invention will be explained in more detail below with reference to the drawing, in which fig. 1 is a table of the conversion from 30 5-bit words to 6-bit words; and FIG. 2 is a supplementary table to the table of FIG. l.

Table 1 lists the 32 possible different five-binary code passwords that constitute the input signal, the consecutive numbers 0-31 indicating the amplitude steps concerned. In conjunction with 5-bit or-6

DK 159353 B

these are shown in the positive and negative modes of the 6-bit word of the output signal, which results from a turnover. In addition, the amplitude steps corresponding to the 6-bit words and the two sequences are shown. As already mentioned 5, the 6-bit words - controlled by the ongoing digital sum - can appear in positive or negative mode. The order of the 5-bit words and the 6-bit words is indicated by the letters A B C ...., A always being entered or given first.

It turns out that the 32 possible amplitude values for the 10 input signal can be divided into two equal ranges, one comprising the amplitude values from 0 to 15, and the other the amplitude values from 16 to 31.

The association between the 6-bit words and the 5-bit words in the first region is chosen so that the greatest possible match between the input signal and the output signal is obtained when the first digit of the 6-bit word is not taken into account. . This ensures not only a simple but also a fast connection in the amplitude ranges that occur.

In order to achieve a simple realization without the use of extensive fixed value stores, the interconnection for the second region is chosen so that the code word for the smallest amplitude value in the second region of the positive mode corresponds to the inverted code word for the negative mode for the greatest amplitude value in the first range. The code word corresponding to the amplitude value 49 binary is the inverse code word for the amplitude value 14. Since the code words for the largest amplitude value in the first region are the same for the positive and negative modes, this also applies to the smallest amplitude value in the second region.

For the second-lowest amplitude value in the second region, the positive mode codeword 57 corresponds to the inverted codeword 6 for the negative mode for the second largest amplitude value in the first region. Furthermore, the negative mode password 17 corresponds to that

DK 159353B

7 inverted password 46 for the positive mode of the second largest amplitude value in the first region.

Thus, assuming a line of symmetry between the amplitude values 15 and 16 in the 5-bit words 5 and the corresponding values in the 6-bit words, Table 1 is diagonally symmetrically inverted. Similarly, the code word 19 for the fourth-lowest amplitude value in the second region is also equal to the inverted code word 44 for the negative mode for the fourth-largest amplitude value in the first region. In this case, too, since the codewords 44 in the first region are the same for the positive and negative modes, this will also be the case for the fourth least amplitude value in the second region. Finally, the positive mode code word for the 15 largest amplitude value in the second region is equal to the inverse code word for the negative mode for the smallest amplitude value in the first region. For the smallest amplitude value in the first region, the code words for the positive and negative modes are the same, so that this also applies to the largest amplitude value in the second region.

On the receiving side, the reverse conversion of the six-digit binary words into five-digit binary words takes place using the same connection in the reverse direction.

When receiving reverse transfers, there may be a problem that the transmission caused by a disturbance has falsified the originally sent 6-bit word. If it is assumed that at the time of the disturbance only a single digit is falsified, several results are possible. It may happen that after decoding, there is not just a single bit, but several bits that are counterfeit, or no bits are counterfeit. Also, an initially unused password 35 may have emerged. The error multiplication factor amounts to the association according to Tables 1 and 2

Claims (6)

The realization of the connection according to Tables 1 and 2 by means of an optimized combinatorial network according to known methods using gate circuits and controllable inverters can conveniently be done in highly integrated coupling technique, eg 15 using a so-called mask programmable logic component of the SH100 series. A method of converting digital input signals into a fiber-optic transmission system, which consists of a first binary code of 32 different passwords each with five binary digits corresponding to 32 possible amplitude values, to output signals occurring in a second binary code of password consisting of six binary digits with a positive and a negative mode appearing corresponding to the current digital sum at the end of the preceding six binary digit password where the 32 possible amplitude values (0___31) of the 30 input signal are divided into two equal ranges with each 16 passwords, the first range comprising the amplitude values between a minimum amplitude value (0) and a first intermediate amplitude value (15), and the second range the amplitude values between a second intermediate amplitude value (16) and a maximum amplitude value (31) , where the code words for the positive mode of the output of DK 159353 B 9 the first region are selected thus, a t, the greatest possible correspondence is obtained between the input signal and the output signal, and the code word (49) for the smallest amplitude value in the second region of the positive mode corresponds to the inverse code word (14) for the negative mode for the largest amplitude value in the first range and the code mode (49) of the negative mode to the inverted code word (14) of the positive mode of the largest amplitude value in the first region, characterized in that the code word (57) of the second lowest amplitude value in the second region of the positive mode corresponds to the inverted password (6) of the negative mode for the second largest amplitude value in the first region and the code word (17) of the negative mode corresponds to the inverted password (46) of the positive mode of the second largest amplitude value in the first region, that the codewords for the second amplitude values in the second region are formed in a similar manner and that the codeword for the largest amplitude sputter value in the second positive mode region corresponds to the negative mode inverse password for the lowest amplitude value in the first region and the negative mode password corresponds to the positive mode inverse password for the lowest amplitude value 25 for the first region, such that a diagonally inverted symmetry is obtained between the two ranges of the amplitude values of the input signal and the output signal. Method according to claim 1, characterized in that for both the regions of both the smallest and the largest amplitude value, the code words are mutually identical for the positive and negative modes. Method according to claim 1 or 2, characterized in that the conversion of the input signal to the output signal is carried out in accordance with the code diagram in Table 1. DK 159353 B 10 Method according to one or more of claims 1-3, characterized in that the reverse conversion from six-digit to five-digit passwords is done using the same connection in the reverse direction. The method according to claim 4, characterized in that the reverse conversion of passwords that have been falsified by the transfer to an initially unused six-digit password is carried out in accordance with the code form in Table 2. Method according to one or more of the preceding claims, characterized in that the order of the bits in Table 1 and Table 2 is reversed both within the 5-bit words and within the 6-bit words.