DE4012850A1 - Digital transmission path quality parameter measurement - using measuring signal with sequence of test cells each containing binary test pattern - Google Patents

Abstract

The quality parameter measurement is effected using a measuring signal inserted in a data stream having a cell structure, with each cell having a field which can be freely occupied by data, the measuring signal evaluated at the end of the transmission path. The measuring signal has a sequence of test cells, each with a field which is occupied by a number of binary bits providing a test pattern, with an autocorrelation function corresponding to a Dirac pulse. - The received test patterns are each compared with an identical pattern, or an inverted pattern at the reception end.

Description

The invention relates to a method for determining Quality parameters of a transmission link for digital Data streams with cell structure, with each cell having a free Has data usable useful field in which a measurement signal given the transmission route and after going through the Transmission path is evaluated.

Transmission paths for digital data streams also show physical transmission lines other components, such as switches, standardized connection places, adapters for the connection points and terminals (e.g. telephones or fax machines). About the transmission ranges in digital communication systems are in high Speed in an organized form digital data streams headed. For assessing a digital communication system it is essential to have quality parameters of such an over to know the transmission line. As a quality criteria for example, whether parts of the data stream on the transmission stretch, the duration of the data is lost of the transmission link, whether the order of the Data has been preserved and whether individual binary data (bits) were transferred incorrectly.

From communication technology it is known to have transmission links to test by entering a measurement signal on the input side and that the measurement signal after passing through the transmission distance is evaluated. The test signal can preferably in empty areas of a digitally to be transmitted operationally Data stream are introduced. However, one is certain casual detection of the inserted into the digital data stream  Measurement signal required.

According to the invention in a method is the one mentioned at the beginning Kind of provided that

• a sequence of test cells is used as the measurement signal, their fields of use, each with one of a number of binary data (bits) existing test patterns are occupied, where the auto-correlation function of the test pattern is approximate corresponds to a Dirac impulse and where the position of the test pattern within the sequence of the test cells varied repetitively in the useful fields of the test cells is
• - That every cell of the data stream after passing through the transmission stretch with or identical to the test pattern Test pattern is inversely correlated reference pattern
• - And that the maximum of the cross-correlation function obtained regarding its location with respect to the beginning of each Cell and its height is evaluated.

Under Dirac impulse is a theoretical impulse with one infinitely high amplitude and infinitely short duration to understand with the area l (see e.g. Otto Mildenberger "Basics of system theory for communications engineers", Hansa Verlag, 1981, pages 48 to 50). An advantage of the invention According to the method is that a test cell as such in one operational digital data stream can be detected without that there is additional data - to be carried along in the data stream animals would be - needed. In this way - depending on the structure of the Cells - in the best case, the entire cell can be checked, because for no such data characterizing the test pattern cell positions are required. By evaluating the Maximum of the cross-correlation function obtained with respect its amount can also fix bit errors in the transmission be put. Be in a relatively wide range there is an inversely proportional linear combination hang between the height of the maximum and the number of bit errors.  

Depending on the threshold for the height of the maximum, there are also Test patterns recognizable that contain one or more bit errors exhibit. However, this increases the risk - if the digital data stream also cells transmitted during normal operation has - that a correctly transmitted operational cell the test pattern or bit error in its useful field contains corrupted test patterns corresponding data pattern. This danger can, however, be caused by the increasing length of the test pattern be significantly minimized. By evaluating the location of the Leaves maximum with respect to the beginning of the evaluated cell also determine the number of the test cell without doing so additionally to be transferred - and therefore not verifiable and that Reduce useful field - data for the number of the test cell required are. Through that in relation to the test Test patterns shifted at the beginning of the cell are systematic errors additionally recognizable in an advantageous manner; if everyone Variations of the test pattern in the useful fields of the test cells taken together cover the entire useful field of the test cell, every space in the useful field is replaced with at least once Binary date of the test pattern occupied.

Further advantageous developments of the invention Procedures are contained in the subclaims.

If a particularly high number of test cells are to be used, then according to the invention, the measurement signal is obtained by further test cells complements, the fields of use of which partly with one to the Inverse test patterns are occupied. So that is without Additional data doubled the number of test cells recognizable. Yet another increase in the number of test cells is he wishes, this can be done by using an additional sequence Number that is written in binary in the useful field. This sequence number should preferably be an exponent function to have. This means that after reaching a be agreed number (for example 100) of test cells whose Number is recognizable through the cross correlation, the exponent  is set and the variation of the test cells again from the beginning begins. This would increase the number of test cells in the example 200 increased.

The invention is based on its application in the following a broadband communication system (D-ISDN) based on a Drawing explained. It shows

Fig. 1 to illustrate the differences between the asynchronous transfer mode and the synchronous transfer mode, the time to enter certain cells of specific data flows;

Fig. 2 shows schematically an arrangement for performing the method according to the invention;

Fig. 3 shows a possible structure of a cell of a digital data stream;

Fig. 4 is a test cell structure;

Figure 5 shows the displacement of a test pattern into the available payload field in response to the test cell number.

Fig. 6 shows schematically the evaluation of the cross-correlation function;

Fig. 7 shows the structure of the ATM analyzer shown in Fig. 2 and

Fig. 8 shows the conditions when using further test cells, the useful fields are each partially with a test pattern inverse to the test pattern be.

The method according to the invention is particularly so-called th Asynchronous Transfer Mode (Asynchronous Transfer Mode, ATM) applicable for the broadband ISDN method. The ATM technology seems a universal solution for the transmission of different To be services with a wide range of bit rates.

Due to the ever increasing demands on the communication networks through the different services that are connected development must be carried out at the same time new transmission technology has become necessary. The advantage nes new network should be flexible in the transmission rate and are also open to low transmission rates. With a Broadband capacity of 140 Mbit / s is suitable ISDN for image transmission; e.g. B. for video conferencing or Transmission of large amounts of data such as those used to manufacture Magazines are necessary. The ATM transmission method is in the B-ISDN technology can be advantageously used for all known today Services and offers a great one due to its flexible bit rate System openness for future services.

As an extension of the previous ISDN (Integrated Service Digital Network) the B-ISDN will be introduced in the future. Essential Un The difference to ISDN and SONET (synchronous optical network) is one asynchronous transmission mode.

Fig. 1 shows an irregular occurrence of cells of the data streams 1 , 2 , 3 , 4 in the asynchronous transfer mode ATM of the B-ISDN in contrast to the regular occurrence in time in the synchronous transfer mode. It is particularly important that - in contrast to the synchronous transfer mode (STM) - a particular time slot is not assigned to a transmission channel. In the asynchronous transfer mode, the order of the cells assigned to the respective channel 1 , 2 , 3 , 4 , etc. is not specified.

The arrangement of the cells (frequency distribution) allows practical to achieve full utilization of the channel capacity implement any bit rates in ATM mode for the user. This procedure is therefore open to future developments in the Transfer rates. All data can be transferred  transmissions (telephone, moving image transmission, computer networking, etc.) integrate into a transmission system.

A corresponding arrangement for carrying out the method according to the invention is shown in FIG. 2. A transmission link SUT to be examined with regard to quality parameters is specifically stimulated by means of an ATM generator ATG with a data stream DSS serving as a measurement signal and a received data stream DSR is by means of an ATM -Analysers ATA examined for possible interference from the SUT transmission link.

From a recommendation of the CCITT (Comite Consultatif International Telegraphique et Telephonique), an ATM cell structure is known, which is subdivided into payload and header with a length of 48 or 5 octets (8 bit word) ( Fig. 3).

Essential information that you want to get from the test cell, are the test cell number and the bit error rate in the useful field. Furthermore, it is important to recognize whether it is at all is a test cell. This information is in the previous Suggestions contained in separate data fields in the useful field. (Test cell number, bit error field and test cell identifier).

As shown in FIG. 4, each test cell 10 can be subdivided into a header 11 , a useful field 12 used for internal data (e.g. test cell number, test cell identifier, time stamp) and a useful field 13 available for a bit error measurement.

The space remaining in the available useful field 13 for the bit error measurement is relatively small. One purpose of the invention is to combine as much information as possible from the test cell in one signal, which at the same time enables bit error measurement. To this end, the data for the bit error measurement, the test cell identifier and the test cell number are combined according to the invention.

The ATM generator ATG ( Fig. 2) produces a sequence of test cells T 1 , T 2 , T 3 , T 4 , T 5 , the useful field available for the bit error measurement is described with a test pattern which is the Eigen shaft has that its autocorrelation function AKF is ideally a Dirac pulse. M-sequences, pseudorandom number sequences and the Barker code fulfill e.g. B. this condition is sufficient.

FIG. 5 further shows the header 11 combined into a field 21 and the useful field 12 used for internal data (cf. FIG. 4). The length l of a test pattern is shorter than the length L of the available useful field 20 . This enables the position of the test pattern to be shifted from test cell to test cell preferably in each case by an equal distance AB, in particular by one bit to the right (or to the left), until the pattern has covered the entire available useful field 20 ; this is then repeated cyclically if necessary. The area 22 of the available useful field 20 which is not described with the test pattern 19 is either assigned only the binary value "1" or only the binary value "0".

The ATM analyzer ATA ( FIG. 2) now forms the cross-correlation function KKF over each received cell with a generated or supplied reference pattern which has a known phase reference to the beginning of the cells. A maximum M is obtained for the available useful field 20 , the position of which varies depending on the current test cell number.

The cross-correlation function KKF obtained for the area of the available useful field is evaluated in accordance with FIG. 6. The height m of the maximum of the cross-correlation function KKF obtained determines the similarity of the received test pattern with the reference pattern and thus enables the test cell to be recognized and a statement about the bit error rate. The relative position n of the maximum of the cross-correlation function KKF to the start A of the respective cell determines the test cell number.

The method according to the invention has the following major advantages: Largest possible test area for bit error rate measurement in the available usable field; the bit error rate is determined by height m of the maximum. It is not a synchronization for pattern recognition necessary for the sent test pattern. Detection of the test cell takes place via the cross correlation function KKF of the Testmu sters (no additional data required). The test cell number is by the position n of the maximum of the cross-correlation function  KKF determined; no separate data is required for this. Due to the variable position of the test pattern, the detection sy static bit error possible.

The signal flow for the ATM analyzer ATA is shown in FIG. 7. A received cell 30 is fed from a receiver 31 to a correlator 32 , which contains the available useful field 33 , in which the test pattern is located, with a cell and bit-syn chronous reference pattern 34 (or with the binary inverted reference pattern) from a reference Generator 35 correlates.

The generally unipolar, digital signals of the cell 30 and of the digital reference generator 35 are converted into bipolar signals by adding an offset size, since the known test patterns, the autocorrelation function AKF of which is a single pulse, are generally bipolar signals. If this is not taken into account, the uniqueness of the amplitude of the cross-correlation function KKF is lost for the similarity analysis.

A detector 36 determines the height m and position n of the maximum M from the cross-correlation function KKF thus obtained. From this, the test cell number, the test cell identification and the bit error rate can be determined as stated above.

A doubling of the scope of the available test cell numbers can be achieved in that the ATM generator ATG once the test pattern 19 and once the inverse test pattern 19 '(ie the binary values "1" or "0" of the test pattern 19 are in the Binary values "0" or "1" of the inverse test pattern 19 'converts). The cross-correlation function KKF then, depending on whether the transmitted test pattern is inverted or not, ideally has a positive 40 or a negative 40 ′ maximum, as can be seen in FIG. 8.

Claims (4)

1. A method for determining quality parameters of a transmission link for digital data streams with a cell structure, wherein each cell has a freely usable data field, in which a measurement signal on the transmission link give and is evaluated after passing through the transmission link, characterized in that

• - A sequence of test cells (T 1 , T 2 , T 3 , T 4 , T 5 ) is used as the measurement signal (DSS), the useful fields ( 20 ) of which are in each case partly with a test pattern ( 19. ) consisting of a number of binary data (bits) ) are occupied, whereby the auto-correlation function of the test pattern ( 19 ) corresponds approximately to a Dirac pulse and wherein within the sequence of the test cells (T 1 , T 2 , T 3 , T 4 , T 5 ) the position of the test pattern ( 19 ) in the useful fields ( 20 ) of the test cells (T 1 , T 2 , T 3 , T 4 , T 5 ) is varied without repetition,
• - That each cell of the data stream ( 1 ) after passing through the transmission link (SUT) is correlated with a test pattern ( 19 ) identical ( 34 ) or to the test pattern ( 19 ) inverse reference pattern
• - And that the maximum (M) of the cross-correlation function obtained is evaluated in terms of its position (s) with respect to the beginning (A) of the respective cell and in terms of its height (n).

2. The method according to claim 1, characterized in that the respective remaining area ( 22 ) of the useful fields ( 20 ) of the test cells (T 1 , T 2 , T 3 , T 4 , T 5 ) is assigned identical binary values ( Fig. 6). 3. The method according to claim 1 or 2, characterized in that the position of the test pattern ( 19 ) in successive test cells (T 1 , T 2 , T 3 , T 4 , T 5 ) varies by an equal distance (AB) without repetition. 4. The method according to any one of the preceding claims, characterized in that the measurement signal (DSS) is supplemented by further test cells, the payload fields (20) are each partly covered with an inverse to the test pattern (19) test pattern (19 ').