WO1998009399A2

WO1998009399A2 - Method for the characterization of transfer errors occurring on a digital message transmission route

Info

Publication number: WO1998009399A2
Application number: PCT/DE1997/001947
Authority: WO
Inventors: Dieter Filbert; Thilo Tiburtius
Original assignee: Tektronix, Inc.
Priority date: 1996-08-30
Filing date: 1997-08-29
Publication date: 1998-03-05
Also published as: DE19636341C1; EP0860063A2; WO1998009399A3

Abstract

The invention relates to a method of production of an error signal indicating transfer errors on a digital message transmission route wherein a bit string (ES) is fed into the end of a transmission route and a bit string (AS) is received at the other end of the transmission route and a bit error vector (FV) and a spacing vector (AV) are formed by bit-by-bit comparison. The spacing vector(AV) is compared with the spacing vectors (AVB (1)) stored in the error library. An error signal is produced upon concurrence between the produced spacing vector (AV) and one of the known spacing vectors (AVB(1)). In order to be able to detect any overlapping of transfer errors, sub-vectors(UV(k)) are formed from the spacing vector (AV)produced, each of which describing a part of the bit error that has occurred. The sub-vectors (UV(k)) are subsequently compared with the spacing vectors(AVB (1)). In case of sameness or similarity between a sub-vector (UV(k)) and one of the known spacing vectors(AVB(1)) at least one other error signal is produced.

Description

description

Method for characterizing transmission errors that have occurred on a digital communications link

The invention relates to a method for generating transmission errors on a digital signal transmission link which characterizes error signals, in which a bit error vector is formed by bitwise comparison from a bit sequence fed in at one end of the transmission link and a bit sequence then received at the other end of the transmission link , a distance vector is generated from the bit error vector, which characterizes the time intervals of the bit errors that have occurred, the generated distance vector is compared with known distance vectors characterizing certain transmission errors, and if the generated distance vector and one of the known distance vectors match, an error signal is generated.

Before carrying out such a known method (Erwin Lehmann, "Fault structure measurements on digital transmission systems", technical report 44 TB 113, Deutsche Bundespost Research Institute at FTZ, June 1990, Darmstadt), a large number of existing digital message transmission links were examined with regard to transmission errors An error library was created which was saved and made available for further measurements for evaluation.If a transmission error is found when checking a transmission link to be examined, a distance vector characterizing the error is formed and this is combined with the known distance vectors the error library. If there is a match between the measured distance vector and one of the known distance vectors, an error characterizing the known transmission error is signal generated. The known method can only be used for transmission links that have a very low error rate, since only one error that has occurred can be successfully assigned to a known transmission error. An overlay of different errors can only be recognized if precisely this overlay is stored in the error library with a specific, characteristic distance vector.

The invention is based on the object of specifying a method for characterizing transmission errors which have occurred on a transmission link and which, even with any superposition of any number of transmission errors, supplies the error signals which uniquely characterize the transmission errors which have occurred.

To achieve this object, the method according to the invention proceeds in such a way that subvectors are formed from the generated distance vector, each of which describes a part of the bit errors that have occurred, the subvectors are compared with the known distance vectors and, in the case of equality or similarity, between a subvector and one of the known ones Distance vectors at least one further error signal characterizing further transmission errors is generated.

The main advantage of the method according to the invention is that in the event of a superimposition of different transmission errors due to the use of sub-vectors, each partial error can be detected individually.

Since a source of error on a transmission link can only be reliably identified if it generates certain bit error patterns, ie at least two or more bit errors, it is advantageous in the context of the method according to the invention considered when the subvectors are formed by selecting at least two bit errors in the bit error vector and assigning the components of the subvector time intervals between successive, selected bit errors.

In this way, it is possible to generate a large number of sub-vectors which characterize transmission errors which may have occurred; however, a subset of this large number of sub-vectors is sufficient for the detection of superimposed transmission errors. Such a subgroup can be formed particularly easily and thus advantageously if at least the first and the last bit error in the bit error vector are selected.

In order to achieve the fastest possible evaluation of the measurement results, it is considered advantageous if the respective error signals are generated for all known or similar distance vectors that have subvectors.

In order to be able to determine the superimposition of various transmission errors as error-free as possible, it is considered advantageous if, after the equality or similarity has been established, a new bit error vector is formed between one of the sub-vectors and one of the known distance vectors, which contains all the bit errors of the old bit error vector with the exception of that Has sub-vector corresponding bit errors, a new distance vector and new sub-vectors are formed from the new bit error vector, which with the known

Distance vectors are compared, and if there is similarity or equality between one of the new sub-vectors and one of the known distance vectors, the further error signal is generated. The advantage of this method is that the wrong bits of the first contained in the bit error vector found transmission errors are eliminated, so that a simple and reliable identification of the remaining bit errors is possible. Of course, this process loop can be repeated not only twice, but as often as desired, until either all of them have occurred

Transmission errors detected or the remaining bit errors can no longer be assigned.

A similarity signal indicating the similarity between a sub-vector and a known distance vector can be formed in different ways, but this is possible particularly quickly and thus advantageously by generating a similarity signal indicating the similarity between a sub-vector and a known distance vector when the sub-vector is compared component by component with the known distance vector, the deviation per component is potentiated, all potentiated deviations are added up and the resulting sum falls below a threshold value.

To explain the method according to the invention, an embodiment of an arrangement for carrying out the method according to the invention is shown in the figure.

An input bit sequence ES is fed into a message transmission path 5 and at an input E61 into an XOR element 6. On the output side, the message transmission link 5 is connected to a further input E62 of the XOR element 6. Downstream of the XOR element 6 is a distance vector generator 7, which is connected on the output side to an input E91 of an evaluation device 9. On the output side, the evaluation device 9 is connected to a mask generator 13, the output of which is connected to an input E141 of an error vector generator 14. Another input E142 of the error vector generator 14 is connected to the output of the XOR Link 6 connected. Downstream of the error vector generator 14 is a further distance vector generator 15, which is connected on the output side to a further input E92 of the evaluation device 9.

The method according to the invention is carried out as follows. The input bit sequence ES is fed into the message transmission link 5 and into the XOR element 6. An output bit sequence AS which is then present at the end of the message transmission link 5 also reaches the XOR gate 6. An error vector FV is generated in the XOR gate 6. This indicates a logical .1 * error and a logical, 0 "correctly transmitted bits. For the input bit sequence

ES = (1,0,0,1,0,1,0,1,1,0,1,0,0,1,0,1)

and the output bit sequence

AS = (1,1,0,1,1,0,0,1,0,0,1,0,1,1,1,1)

the error vector results

FV = (0.1.0, 0.1, 1.0, 0.1, 0.0, 0.1, 0.1.0).

A total of six bit errors have occurred on the transmission path 5. The error vector FV arrives at the distance vector generator 7, in which a distance vector AV is formed. This indicates n = 5 time intervals between the six bit errors that have occurred. Since the second bit error in the error vector FV occurred three clock periods after the first bit error, the distance vector AV has a 3 in the first position. The third bit error occurred a period after the second bit error, so that there is a 1 at the second position of the distance vector AV. The fourth bit error has a time interval of three period lengths to the third bit error, so that the third component of the distance vector AV again has a three. The fifth bit error occurred four period after the fourth bit error and the sixth bit error occurred two period after the fifth, so that the last two components of the distance vector AV have a 4 and a 2. The distance vector therefore has the form

AV = (3,1,3,4,2).

The distance vector AV reaches the evaluation device 9, in which sub-vectors UV (k) are formed. Each sub-vector UV (k) identifies a part of the bit errors that have occurred that contain at least the first and last bit errors in the form of distance vectors. The sub-vector UV (1) has all errors that have occurred, so that it has the same shape as the distance vector AV: UV (1) = (3,1,3,4,2). The second sub-vector UV (2) identifies the first, second, third, fourth and sixth bit errors, so that it has the following components: UV (2) = (3,1,3,6). The third sub-vector UV (3) identifies the first, second, third, fifth and sixth bit errors and is: UV (3) = (3,1,7,2). In this way the sub-vectors UV (4) = (3,4,4,2), UV (5) = (4,3,4,2), UV (6) = (3,1,9), UV <7) = (3,4,6), UV (8) = (3,8,2), UV (9) = (4,3,6), UV (10) = (4,7,2) , UV (ll) = (7,4,2), UV (12) = (3,10), UV (13) = (4,9), UV (14) = (7,6), UV (15 ) = (11.2) and UV (16) = (13). The sub-vectors UV (k) thus each identify a sub-group of bit errors, each sub-group containing at least the first and last bit error of the bit error vector FV. So there are (n + 1) bit errors

2 ⁿ = 16 sub-vectors are generated, so that for k:

1 ≤ k ≤ 2 ⁿ ~ ^l . The sub-vectors UV (k) are compared with the known j distance vectors AVB (l) (l≤l≤j) stored in the evaluation device 9, which identify certain transmission errors on message transmission paths known from test measurements carried out in advance. If a sub-vector UV (k) is identical or similar to a known distance vector AVB (l), then this is transmitted from the selection device 20 contained in the evaluation device 9 to the mask generator 13. Various similarity measures (quadratic error, cubic error, etc.) can be used as a basis for the assessment of the similarity. In the evaluation device 9, the similarity between a sub-vector UV (k) and a known distance vector AVB (l) is determined by comparing the sub-vector UV (k) with the known distance vector AVB (l) component by component, the deviation per component being squared, all squared deviations are added up and the measure of the similarity is formed by the resulting sum. If the resulting sum falls below a predetermined threshold value for a sub-vector UV (i) (1 ^~ i 2 2 ^{n ~} ), this sub-vector UV (i) is transmitted to the mask generator 13. Known distance vectors AVB (1) necessary for carrying out the method according to the invention can be found in the above-mentioned technical report.

A sub-error vector UFV (i) is formed from the sub-vector UV (i) in the mask generator 13 and has the form of an error vector: If, for example, the sub-vector UV (6) = (3,1,9) was selected, the sub-error vector UFV (6) the form UFV (6) = (0.1.0.0.1.1.0.0.0.0.0.0.0,1.0). The sub-error vector UFV (i) thus, like the error vector FV, identifies bits which have occurred with a logical .1 "and correctly transmitted bits with a logical .0". The sub-error vector UFV (i) arrives at the one input E141 of the error vector generator 14, at its further input E142 the original error vector FV

FV = (0.1.0.0.1.1.0.0.1.0.0.0,1,0,1,0)

is present. In the error vector generator 14, a new error vector FVN is formed from the sub-error vector UFV (i) and the original error vector FV, which has all bit errors with the exception of the bit errors identified by the sub-error vector UFV (i). The error vector FVN generated in this way thus has the form:

FVN = (0.0.0.0.0.0.0.1.0.0.0,1,0,0,0). The new error vector FVN is stored in the error vector generator 14 and arrives at the further distance vector generator 15, in which a new distance vector AVN is generated. The mode of operation of the distance vector generator 7 and of the further distance vector generator 15 are identical, so that the new distance vector has the following components: AVN = (4). The new distance vector AVN arrives at the further input E92 of the evaluation device 9, in which new sub-vectors are formed from the new distance vector AVN, which in turn are compared with the known distance vectors AVB (l).

The process loop formed by the evaluation device 9, the mask generator 13, the error vector generator 14 and the further distance vector generator 15 is run through until the equality or similarity between a sub-vector UV (k) and a known distance vector AVB (l) can no longer be determined. In this case, in the error generator 14 after the first loop pass to generate the new error vector FVN, the original error vector FV present at the further input E142 is used, but the new error vector previously stored during the run. The known distance vectors AVB (1) for which in the evaluation device 9 equality or similarity Subvectors UV (k) have been identified, characterize the transmission errors that have occurred on the message transmission link 5.

In conclusion, it should be pointed out that the method according to the invention is carried out in practice by means of an electronic data processing system; The block diagrams shown therefore serve primarily to illustrate the method according to the invention.

Claims

claims

1. A method for generating transmission errors on an error signal which characterizes a digital message transmission link, in which

a bit error vector (FV) is formed from a bit sequence (ES) fed in at one end of the transmission link (5) and a bit sequence (AS) then received at the other end of the transmission link, - a bit vector (FV) from the bit error vector (FV) - a distance vector ( AV) is generated, which characterizes the time intervals of the bit errors that have occurred,

- The distance vector (AV) generated is compared with known distance vectors (AVB (D) characterizing certain transmission errors and

- if the generated distance vector (AV) and one of the known distance vectors (AVB (l)) match, an error signal is generated, characterized in that - if there is no match, sub-vectors (UV (k)) are formed from the generated distance vector (AV), each of which describes part of the bit errors that occurred,

- The sub-vectors (UV (k)) are compared with the known distance vectors (AVB (l)) and - if there is equality or similarity between a sub-vector (UV (k)) and one of the known distance vectors (AVB (l)) at least one further Transmission error characterizing further error signal is generated.

2. The method according to claim 1, characterized in that the sub-vectors (UV (k)) are formed by - at least two bit errors in the bit error vector (FV) are selected and - The components of the sub-vector (UV (k)) are assigned time intervals between successive, selected bit errors.

3. The method as claimed in claim 2, so that at least the first and the last bit error in the bit error vector (FV) are selected.

4. The method according to claim 1 to 3 d a d u r c h g e k e n n z e i c h n e t that for all equality or similarity to sub-vectors (UV (k)) having known distance vectors (AV), the respective error signals are generated.

5. The method according to claim 1 to 3, d a d u r c h g e k e n n z e i c h n e t that

- After established equality or similarity between one of the sub-vectors (UV (i)) and one of the known distance vectors (AVB (D), a new bit error vector (FVN) is formed, which contains all bit errors of the old bit error vector with the exception of the sub-vector (UV (i)) has corresponding bit errors,

- A new distance vector (AVN) and new sub-vectors are formed from the new bit error vector (FVN), which are compared with the known distance vectors (AVB (l)) and

- If there is similarity or equality between one of the new sub-vectors and one of the known distance vectors (AVB (D), the further error signal is generated.

6. The method according to claim 1 to 5, characterized in that a similarity signal indicating the similarity between a sub-vector (UV (k)) and a known distance vector (AVB (D) is generated if

the sub-vector (UV (k)) is compared component by component with the known distance vector (AVB (D),

- the deviation per component is potentiated,

- all potentiated deviations are added up and

- the resulting sum falls below a threshold