Method and device for processing and representing signals
Patent Information
- Authority / Receiving Office
- EP · EP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2024-05-03
- Publication Date
- 2026-04-08
AI Technical Summary
Existing digital oscilloscope technologies face challenges in displaying multiple signal tracks with varying starting times without losing information, especially in handling high data rates and short-term signal fluctuations, often resulting in data loss and increased computing time.
A method and device that utilize a minimum-maximum tree structure for parallel data processing to compress and display signal data, allowing for simultaneous display of multiple signal tracks with individual starting times, and control signals indicating changes in column entries, reducing data volume and computing time by several orders of magnitude.
This approach ensures minimal information loss and efficient data processing, enabling the display of high-data-rate signals without losing short-term fluctuations, and reduces the data volume and computing time required for display by more than a power of ten compared to conventional methods.
Smart Images
Figure EP2024062290_28112024_PF_FP_ABST
Abstract
Description
[0001] Method and device for signal processing and display
[0002] The invention relates to a method for processing and displaying signals, for example, measurement data from an oscilloscope. Furthermore, the invention relates to a device designed to carry out such a method.
[0003] Possible displays on an oscilloscope are discussed, for example, in US Pat. No. 6,104,374 A. The goal here is to use a digital oscilloscope to create a visual impression that is as close as possible to an analog oscilloscope. In particular, US Pat. No. 6,104,374 A addresses options for adjusting the brightness of individual pixels.
[0004] EP 1 139 103 B1 discloses a device referred to as a general-purpose oscilloscope with digital television signal reproduction capability. This device processes both digital and analog signals.
[0005] EP 0 350 557 B1 discloses a method for compressing digital time-serial data for display on a digital oscilloscope. This method involves grouping data points and determining maximum and minimum values.
[0006] The invention is based on the object of further developing the display of signals, for example measurement data, which are to be displayed on a digital oscilloscope, compared to the prior art, whereby information losses are to be avoided as far as possible, regardless of the data rate.
[0007] This object is achieved according to the invention by a method for the simultaneous display of multiple signal traces according to claim 1. A device according to claim 11 is suitable for carrying out the method. The method according to the application serves to display multiple signal traces of at least one stream of sampled measurement data, which are to be presented simultaneously on a screen, wherein each of the signal traces has an individual start time. In this case, the start time can be given in particular by a trigger, wherein the start time is, for example, identical to a trigger time or has a fixed time interval from a trigger time. A trigger condition can be fulfilled either by the stream of sampled measurement data itself or by any other signal.Measurement data can be acquired on a single channel or on multiple channels, allowing any number of streams of sampled measurement data to exist. For example, there may be a single signal trace per channel or multiple signal traces to be displayed. Signal traces to be displayed simultaneously can be spaced apart in time or overlap. This applies to signal traces on different channels as well as to signal traces on the same channel.
[0008] In any case, the assignment of consecutive time segments of the signal traces to be displayed to columns of the screen display is adjustable. In at least one of the selectable settings, i.e., operating modes, more than two measurement data items exist per column and signal trace. In each operating mode, the minimum and maximum of the measurement data of the signal traces assigned to each column of the screen display must be displayed.
[0009] The minima and maxima to be displayed in the individual columns are determined by first building at least one minimum-maximum tree on the data stream containing the signal traces through at least partially parallel data processing. This minimum-maximum tree forms a common database for the screen displays in the various settings and has a width, expressed as the number of measurement data points, that corresponds to a power of two. A section is cut from the data stream which, in at least one of the selectable settings, does not correspond to a power of two, has a maximum length equal to the length of the signal traces to be displayed, and is to be displayed on the entire screen, i.e., across all columns. These sections are divided into the sections of the data stream to be displayed column by column according to the number of columns and the selected screen display setting.In these sections, the minimum and maximum of each signal trace are determined by tree traversals. The traversals through the tree (i.e., the minimum-maximum tree) are performed for multiple columns of the screen display using parallel data processing. This allows for column-by-column selection of minimums and maximums from the measurement data, depending on the selected screen display, i.e., the time base of the display. The selection of the setting-specific minimums and maximums is performed using the same previously created minimum-maximum tree and is used for an envelope display.
[0010] In short, the display device used in the display method according to the application is designed to display digitized measured values column by column, wherein at least in one of several operating modes there are more than two measured values per column to be displayed.
[0011] In particular, the presentation process includes the following steps:
[0012] Determination of the minimum and maximum of the measured values assigned to a column by at least partially parallel data processing, generation of display bars, each to be displayed in a column, from the minimum-maximum pairs for each start time, in particular trigger time, so that the totality of the display bars forms an envelope curve representation of the measured values,
[0013] Control of the display device with control signals which indicate exclusively changes in the column entries in each column, such changes being present in particular at the lower and upper edge of each display bar.
[0014] Overall, this results in a two-stage reduction of data: First, the
[0015] Measurement data is hierarchically compressed by recursively extracting the minima and maxima from sections of the data stream. This is done by building structures on the measurement data and then traversing them in various ways, depending on the selected settings. These structures can be thought of as trees or pyramids. The extraction of the minima and maxima is an essential prerequisite for ensuring that even the briefest fluctuations that occur in the incoming data stream, i.e., in the stream of measured values, are not lost. This is conceivable, for example, through averaging as provided for in conventional methods. By forming the minimum-maximum pairs, the data is - figuratively speaking - compressed on the time axis without having to accept any significant loss of information, especially with regard to short-term signal excursions, so-called glitches or spikes.
[0016] The second step of data compression, which concerns the bar display, occurs entirely without any loss of information. By transmitting only the changes in values, which are to be visualized as pixels in the individual columns, in the form of control signals instead of absolute values, a reduction in the data volume to be displayed and the computing time required for the display can be achieved by more than a power of ten.
[0017] Overall, the display method according to the application reduces the data volume required to control the display device, as well as the computing time required for display, by several orders of magnitude compared to conventional methods. This makes it possible to simultaneously use the device intended for displaying data, which can be attributed in particular to an oscilloscope, for digitally processing analog data acquired by the data acquisition device, which is received as an uninterrupted data stream.
[0018] The signal pattern recognition method disclosed in DE patent application 10 2021 130 772.4 of November 24, 2021, is particularly suitable for determining the minimum and maximum values assigned to a column of the display device. This means that the minimum and maximum values to be displayed in each column are determined by constructing at least one min-max tree on the digitized measured values. This is achieved by defining the measured values as the first layer of the tree and constructing further layers of the tree on this layer, which, unlike the first layer, are formed from min-max pairs. Each min-max pair is formed from the minimum and maximum of the two pairs or sample values located in the most recently created or already existing layer. The tree construction is carried out by parallel data processing, at least in some of the layers, including the first layer.Subsequently, for each start time, especially the trigger time, and for each column of the screen display, the minimum and maximum values are calculated using parallel tree traversals for the set of measured values included in the respective column of the display. The number of measured values used for each column of the display can vary across the columns. This can be the case, for example, if the time increment of consecutive display columns is a non-integer multiple of the measured value time grid.
[0019] In particular, by using the signal pattern recognition method according to DE application 10 2021 130 772.4, it is possible to divide the measured values into blocks, whereby measured values from more than one block, for example an integer or non-integer multiple of a block, measured values of exactly one block or measured values of a section of a block are displayed on the display device, and whereby the number of measured values per block can be more than the square of the number of columns of the display device.
[0020] If there are multiple blocks from which measured values are to be displayed simultaneously, it may be expedient, in order to minimize the data volume to be handled and the computational effort, to first compress the data from at least one of these blocks hierarchically before merging it with data from another of the aforementioned blocks. The aforementioned minimum-maximum trees make a significant contribution to reducing the data processing effort. This makes it possible, among other things, to keep only a more recent portion of the measured values available in a memory at the time of display, while the digitized measured values of at least one older data block have already been overwritten in a ring buffer. Instead, only the data required to display the display bars, which includes the minimum and maximum values for each column, is kept available.Incremental block processing is particularly useful for long signal trace durations.
[0021] According to various possible method variants, the same min-max pairs generated using the teaching of DE application 102021 130772.4 are accessed to determine the start times, particularly in connection with triggering, and these pairs also form the database for displaying the envelope curves. For example, to display an eye diagram, several display bars can be displayed simultaneously, spaced apart, in at least some of the columns.
[0022] The measured values to be digitized are, for example, acquired at intervals of no more than 10 ns. In an operating mode with an extremely long main time base of the oscilloscope, the values to be displayed in the individual columns in bar form can be provided as min-max pairs of at least 10 6To extract measured values. Conversely, an operating mode with an extremely short main time base can also exist, in which only a single measured value is displayed per column.
[0023] In cases where there is a very large number of measured values per column to be displayed, there is a particularly large effect of data saving without any relevant loss of information, whereby even in such cases the envelope display comprises vertical display bars to be displayed in individual columns of the screen, whereby control signals are used to control the screen, i.e. the display device, which in each column indicate only changes in the column entries, comprising the lower and upper edge of each display bar.
[0024] The columns of the screen can also be used for graduated brightness displays, whereby the brightness of a pixel in the column is set higher the more frequently a signal was detected at that time, relative to the start time of the signal trace. For example, the brightness is displayed in at least 16 levels, with the screen being controlled in such a way that for each column, not absolute brightness values are transmitted, but only changes in brightness that occur when moving through the column.
[0025] The device according to the application for displaying a plurality of signal traces of at least one stream of sampled measurement data to be presented simultaneously on a screen generally comprises a display device controlled by an arithmetic and logic unit, wherein the arithmetic and logic unit, in cooperation with a data acquisition device provided for recording the signal traces, is configured to carry out the method according to claim 1. Said arithmetic and logic unit can also be configured for digitally processing analog data acquired by the data acquisition device in the form of an uninterrupted data stream. In particular, the arithmetic and logic unit can check the data stream without interruption for compliance with a trigger condition, whereby triggering does not result in any dead times.
[0026] An embodiment of the invention is explained in more detail below with reference to a drawing. In the drawings:
[0027] Fig. 1 shows a block diagram of an apparatus for carrying out a method for signal processing and display,
[0028] Fig. 2 shows in a diagram features of the method that can be carried out with the device according to Fig. 1,
[0029] Fig. 3 Possibilities of grouping and processing data in different settings of the device according to Fig. 1,
[0030] Fig. 4 shows, in a more detailed diagram compared to Fig. 2, further features of the method which can be carried out with the device according to Fig. 1.
[0031] A signal processing and display method, visualized in different ways in Figures 1 to 4, assumes that measurement data is received as a continuous data stream DS. In this case, the measurement data are electrical signals acquired at a clock rate of 1.2 ns.
[0032] A signal processing and display device 1, which is designed to carry out the method and is shown in a rough schematic in Fig. 1 and which is an oscilloscope in the exemplary embodiment, comprises a data acquisition device 2, which is designed to acquire the data stream DS, and a data processing device 3 connected downstream of the data acquisition device 2.
[0033] Components of the data processing device 3 are a ring buffer 4 and a plurality of arithmetic units 5, in the present case more than a thousand. Embodiments with a smaller number of arithmetic units 5, for example, 128, are also possible. The signal processing and display device 1 is configured, among other things, to perform all the functions described in German patent application No. 10 2021 130 772.4, in particular trigger functions. 6 denotes the display device, i.e., the screen, of the signal processing and display device 1.
[0034] The method for signal processing and representation that can be carried out with the device 1 according to Fig. 1 is illustrated in broad outline in Fig. 2, according to which a division into three steps S1, S2, S3 is given.
[0035] The measured values acquired with the aid of the data acquisition device 2 are digitized in the first step S1 and are available in blocks, i.e., as data blocks DB. As can be seen in the middle section of Fig. 2, which relates to step S2, min-max trees BA are constructed on the digitized data, each consisting of a plurality of levels E0, E1, ... The actual number of levels E0, E1, ... far exceeds the number of levels E0, E1, ... shown in Fig. 2. The min-max trees BA are constructed according to the method disclosed in DE application 102021 130772.4, whereby all minima and maxima of the time periods considered are retained. The min-max trees BA are constructed by first forming pairs of sample values in the lowest level E0. The number of pairs of sample values assigned to a single tree BA corresponds to a power of two, for example, thirty-two or more.One of the two values of each pair represents a minimum value, the other a maximum value, although both values may also coincide. In each case, the corresponding min-max pair is written to level E1. This is done based on a large number of parallel data processing steps performed by the arithmetic and logic units 5.
[0036] From level E1 onwards, pairs of values are compared rather than individual values. Thus, for each comparison, there are a total of four values from which the minimum and maximum are selected. The next min-max pair selected in this way is written to the next higher level. This process also uses parallel data processing. Synchronization occurs where necessary. The construction of the min-max tree BA is complete when the min-max pair of the entire tree BA, i.e., the tip of the tree BA, has been found. The min-max trees BA, each built on a data block DB, represent static structures for the further process that can be used multiple times. Later tree traversals, also called depth-level traversals and required for the screen display explained in more detail below, are not bound to tree boundaries.
[0037] Illustratively, the structure of the trees BA, which can be used multiple times, especially with different time base settings of device 1 including display device 6, and in which minimum and maximum values are entered, means a compression of the data set on the time axis, as indicated in Fig. 2, step S2, by horizontal arrows pointing toward each other. It is crucial that no minimum or maximum values to be displayed are lost.
[0038] In the third step S3, at the very bottom of Fig. 2, the previously compressed data is displayed column by column, in the form of bars, with only two of these columns, Spl, Sp2, being sketched in Fig. 2. The minimum and maximum values of each column Spl, Sp2 represent the lower and upper limits, respectively, of a display bar, which is to be visible in the corresponding column Spl, Sp2 on the display device 6. In the simplest case, each display bar is composed exclusively of "1" pixels, whereas the pixels outside the display bars are set to "0."
[0039] With conventional control of the display device 6, each pixel within the display bar would now be controlled separately with "1". All pixels that are located in the same column Spl, Sp2 outside the display bar would be controlled individually with "0" in an analogous manner.
[0040] The method used in the present case for controlling the display device 6 differs fundamentally from this conventional control method in that it is not absolute values assigned to pixels that are transmitted to the display device 6, but rather only changes in these values. Thus, in the simple case shown in Fig. 2, in which no fine gradation of brightness is provided, when scanning through a column Spl, Sp2, a change occurs once from "0" to "1" and once from "1" to "0". Overall, this represents a drastic reduction in the data volume that must be transferred to control the display device 6. The vertical arrows pointing towards each other next to the two exemplary columns Spl, Sp2 are not intended to symbolize a pushing together of the display bars, but rather a reduction in the amount of data.The display device 6 is operated by the same data processing device 3, which simultaneously receives and processes the data stream DS (Fig. 2, step S1). Thus, data is recorded and displayed without interruption, regardless of the triggering.
[0041] Fig. 3 shows relationships between the data stream DS and various possible settings of the display device 6. First, min-max trees BA of equal size are constructed, which connect to one another without interruption. The width of an individual tree BA, expressed as the number of digitized measured values displayed as individual points in the lowest level E0 of the trees BA, can be smaller or larger than the time period to be displayed on the screen 6. In Fig. 3, it is assumed that a time base of the display device 6 is initially selected which enables data from a time period that is greater than the width of an individual tree BA to be displayed. The simultaneous display of several signal tracks, which are also generally referred to as traces, will not be discussed here. Display blocks ABI of uniform width are, as can be seen from the middle section of Fig.3, are significantly wider than the individual blocks BA, but do not represent an integer multiple of a block BA. An individual display block ABI is to be displayed on the display device 6 as a section of the data stream DS, at least in part, namely in the form of minimum and maximum values, which together form an envelope curve. Boundaries between consecutive sections ABI only coincide with boundaries between trees BA in isolated cases. Otherwise, boundaries between the sections ABI lie within a tree BA.
[0042] The statement regarding the position of the boundaries between sections of the data stream DS, which are defined for the purpose of screen display, also applies to the time base setting of device 1, i.e., the oscilloscope, shown at the bottom of Fig. 3. In this case, sections AB2 are formed that are shorter than the temporal extent of a tree BA. An individual section AB2, i.e., a single portion of the data stream DS intended for display on screen 6, can lie entirely within a tree BA or extend from a first tree BA into a neighboring tree BA.
[0043] Fig. 4 shows the simultaneous display of several signal traces SKI, SK2, SK3. Without loss of generality, it is assumed that the first two signal traces SKI and SK2 are acquired on a first channel of the data acquisition device 2 and the third signal trace SK3 is acquired on a further channel of the data acquisition device 2. The signal traces SKI, SK2, SK3, which are also generally referred to as signal curves, have a uniform length. In the present, simplified case, the total length of each signal trace SKI, SK2, SK3 is divided into two sections of equal length SKa, SKb. The columns Spl, Sp2 are assigned to these sections SKa, SKb on the screen display. In reality, there is a much larger number of sections SKa, SKb and columns Spl, Sp2. Each signal curve SKI, SK2, SK3 begins at a time tl, t2, t3, which is triggered by a trigger as described in the DE patent application 10 2021 130 772.4, is precisely determined. Either the same data stream DS, which is partially displayed in columns Spl, Sp2, and other columns of the display device 6, or any other signal can be used for triggering. In the diagrams sketched in Fig. 4, which show arbitrary shapes of the signal curves SKI, SK2, SK3, where the amplitude is designated by A, the signals SKI and SK2 have a positive sign, while the sign of the signal SK3 is negative.
[0044] Based on the digitized measured values, which each form a signal curve SKI, SK2, SK3, a table TAI is first filled for each signal curve SKI, SK2, SK3. In Fig. 4, this step is illustrated for the two signal curves SKI, SK2, but not shown for the third signal curve SK3. The table TAI is made up of several table columns TS1, TS2, the number of which corresponds to the number of sections SKa, SKb of the signal traces SKI, SK2, SK3, and is therefore much larger than two in reality. The length of each column TS1, TS2 represents the ordinate of the corresponding diagram in which the signal trace SKI, SK2, SK3 is recorded. For example, if the TS1 column is traversed from bottom to top, the value zero is entered in the table column TS1 if this amplitude value does not appear for the corresponding amplitude A within the SKa section. Otherwise, the value one is entered in column TS1.Analogously, the next column TS2 and other columns not visible in Fig. 4 are filled, whereby this process is carried out for a plurality of table columns TS1, TS2, ... by parallel data processing.
[0045] Subsequently, based on each table TAI, another table TA2 is populated, which has significantly fewer entries than the first table TAI. In terms of the number of columns TS1 and TS2, the second table TA2 does not differ from the first table TAI. Likewise, the assignment of the table columns TS1, TS2, ... to the subsequent columns Spl, Sp2, ... of the screen display is retained. In contrast to the first table TAI, the second table TA2 only contains entries to the extent that table entries in the first table TAI change when passing through the individual columns TS1 and TS2, as already explained in the third step S3 in Fig. 2. In the case of the first section SKa of the first signal trace SKI, for example, this means that the first column TS1 of the second table TA2 only contains two entries, namely the entry "+1" and the entry "-1".Reading the table column TS1 from bottom to top, these entries are to be understood as meaning that "+1" increases the displayed value from zero to one, and "-1" decreases it back to zero. In the same way, each subsequent table column TS2, ... is generated until the second table TA2, which refers to a single signal track SKI, SK2, SK3, is completely filled.
[0046] After the tables TA2 for the individual signal tracks SKI, SK2, SK3 have been completed, a third table TA3 is created from these tables TA2 by summation, which is shown in Fig. 4 in a simplified manner exclusively for the two signal tracks SKI, SK2. In fact, the third signal track SK3 is also included in the summation. Instead of filling individual tables TA1, TA2, TA3, a final summary table, i.e., table TA3, can be updated directly, whereby several signal tracks SKI, SK2, SK3, which are also generally referred to as events, are processed in a synchronized manner.
[0047] As a result, in table TA3, individual columns TS1, TS2, ... also contain entries that are greater than one. In the case outlined in Fig. 4, this is the case in the positive range of columns TS1, TS2. This means, as shown in the diagram at the bottom right of Fig. 2, that display bars visible in screen columns Spl, Sp2 sometimes have a width of one and sometimes a width of two. Deviating from the procedure illustrated in Fig. 4, it is also possible to increment table entries by non-integer values. In this case, an increment can depend, for example, on the event, i.e., on the signal trace SKI, SK2, SK3.
[0048] In actual operation of the signal processing and display device 1, instead of the
[0049] Width of the display bars, whose brightness varies, with each display bar being placed close to the next, so that—unlike in Fig. 4—there are no gaps between the columns Spl, Sp2, etc. Likewise, different strengths of a display bar can be visualized by color coding. There can also be a relationship between the display type of a display bar and its height.
[0050] The screen display shown on the display device 6, as shown in Fig. 4 (bottom right), represents a small section of an eye diagram. The brighter a section of the screen display appears, the more often one of the signal traces SKI, SK2, SK3 was detected in the corresponding area. Instead of varying the brightness, or in addition to varying the brightness, the persistence duration of sections of the recorded signal traces SKI, SK2, SK3 can also be varied.
[0051] In any case, the minima and maxima of the signal traces SKI, SK2, SK3 (Fig. 4, diagrams top and middle left) are included in the screen display without any loss of information. The screen display parameters are selected such that even minimal values other than zero in the summed table TA3 can still be clearly seen in the corresponding column Spl, Sp2, ... of the display device, regardless of the time base setting of the signal processing and display device 1.
[0052] List of reference symbols
[0053] 1 signal processing and display device
[0054] 2 Data acquisition device
[0055] 3 Data processing device
[0056] 4 ring buffers
[0057] 5 Calculator
[0058] 6 Display device, screen
[0059] A Amplitude
[0060] ABI, AB2 display blocks, sections
[0061] BA min-max tree
[0062] DB data block
[0063] DS data stream
[0064] E0, El levels, layers
[0065] SI, S2, S3 steps
[0066] SKI, SK2, SK3 signal curves, signal traces
[0067] SKa, SKb sections of a signal track
[0068] Spl, Sp2 Columns of the screen display t Time tl, t2, t3 Start times
[0069] TAI, TA2, TA3 tables
[0070] TS1, TS2 table columns
Claims
Patent claims 1. Method for displaying several signal traces (SKI, SK2, SK3) of at least one stream of sampled measurement data to be displayed simultaneously on a screen (6), wherein - each signal trace (SKI, SK2, SK3) has an individual start time (tl, t2, t3), the assignment of successive time segments (SKa, SKb) of the signal traces (SKI, SK2, SK3) to columns (SP1, Sp2) of the screen display is adjustable, in several of these settings there are more than two measurement data items per column (Spl, Sp2) and signal trace (SKI, SK2, SK3), in each of the said settings the minimum and the maximum of the measurement data of the signal traces (SKI, SK2, SK3) assigned to each column (Spl, Sp2) of the screen display is to be displayed, the minima and maxima to be displayed in the individual columns (Spl, Sp2) are determined by o building at least one minimum-maximum tree (BA) on the data stream (DS) comprising the signal traces (SKI, SK2, SK3) by at least partially parallel data processing, which forms a common database for the screen displays in the various settings and a width,expressed as the number of measurement data, which corresponds to a power of two, o a section (ABI, AB2) is cut out of the data stream (DS), which in at least one of the settings does not correspond to a power of two, has a maximum length of the signal traces (SKI, SK2, SK3) to be displayed and is to be displayed on the entire screen (6), i.e. the entirety of the columns (Spl, Sp2), o the sections (ABI, AB2) are divided into the sections (SKa, SKb) of the data stream (DS) to be displayed column by column according to the number of columns (Spl, Sp2) and the selected screen display setting, o in the said sections (SKa, SKb) the minimum and maximum of each signal trace (SKI, SK2, SK3) are determined by tree traversals, whereby the traversals through the tree (BA), i.e. minimum-maximum tree, are carried out for several columns (Spl, Sp2) in the form of parallel data processing, the measurement data representing minima and maxima selected column by column for the various screen displays which differ from one another in terms of time base, are used for an envelope display with access to the same minimum-maximum tree (BA).
2. Method according to claim 1, characterized in that the stream of measurement data is divided into data blocks (DB), each of which extends over a power of two of measurement data, wherein at least one minimum-maximum tree (BA) is constructed for each data block (DB) and at least one of the starting times (tl, t2, t3) lies within a data block (DB) in such a way that the part of the data block (DB) beginning with the starting time (tl, t2, t3) does not correspond to a power of two.
3. Method according to claim 2, characterized in that start times (tl, t2, t3) are used which correspond to trigger times.
4. Method according to claim 3, characterized in that the envelope representation of the signal traces (SKI, SK2, SK3) dependent on the position of the trigger times (tl, t2, t3) is determined in the following steps: Definition of the samples of a data block (DB) lying in defined time intervals (SKa, SKb) with the help of the trigger times (tl, t2, t3) as the first layer (EO) of the minimum-maximum tree (BA), construction of further layers (El) on the first layer (EO), whereby the further layers (El), in contrast to the first layer (EO), are formed from min-max pairs, whereby each min-max pair consists of the minimum and the Maximum of the two pairs or samples lying in the last created or already existing layer (EO, El, ...) is formed, and wherein the tree construction is carried out by parallel data processing at least in a part of the layers (EO, El, ...), which includes the first layer (EO), Carrying out tree traversals, some of which cross block boundaries, for each column (Spl, Sp2) of the screen display, wherein firstly each individual sample value is defined as a min-max candidate and two consecutive min-max candidates are replaced by the min-max pair lying as the associated node in the layer above (E1, ...), provided that all samples assigned to this min-max pair also lie within the section to be displayed in a single column (Spl, Sp2), and identifying the minimum value and maximum value assigned to the column (Spl, Sp2) on the basis of the min-max candidates found by the tree traversal, wherein one and the same min-max tree is used for a plurality of tree traversals to be carried out in parallel, Generation of an envelope curve belonging to the signal curve (SKI, SK2, SK3) from the minimum and maximum values specific to the column (Spl, Sp2).
5. Method according to one of claims 2 to 4, characterized in that there is at least one setting of the screen display in which the number of measured values per data block (DB) is more than the square of the number of columns (Spl, Sp2) displayed on the screen (6).
6. Method according to one of claims 1 to 5, characterized in that the envelope curve display comprises vertical display bars visible in individual columns (Spl, Sp2) of the screen (6), wherein control signals are used to control the screen (6) which indicate in each column (Spl, Sp2) exclusively changes in the column entries, comprising the lower and upper edge of each display bar.
7. Method according to claim 6, characterized in that the columns (Spl, Sp2) of the screen (6) are used for graduated brightness displays, wherein the brightness of a pixel in the column (Spl, Sp2) is set higher, the more often a signal was detected at the corresponding time, relative to the start time (tl, t2, t3) of the signal trace (SKI, SK2, SK3).
8. Method according to claim 7, characterized in that the brightness is displayed in at least 16 levels, the screen (6) being controlled in such a way that for each column (Spl, Sp2) no absolute brightness values are transmitted, but only changes in the brightness which occur when passing through the column (Spl, Sp2).
9. Method according to claim 7 or 8, characterized in that several display bars are displayed simultaneously in at least some of the columns (Spl, Sp2) to form an eye diagram.
10. Method according to one of claims 6 to 9, characterized in that the measured values are recorded at time intervals of not more than 10 ns, wherein in a setting with an extremely long main time base the values to be displayed in the individual columns (Spl, Sp2) in bar form are each displayed as min-max pairs of at least 10 6 measured values are extracted.
11. Device for displaying a plurality of signal traces (SKI, SK2, SK3) of at least one stream of sampled measurement data to be displayed simultaneously, comprising a screen (6) controlled by means of a data processing device (3), wherein the data processing device (3) is set up in cooperation with a data acquisition device (2) provided for recording the signal traces (SKI, SK2, SK3) to carry out the method according to claim 1.
12. Device according to claim 11, characterized in that the data processing device (3) is also designed for digital processing of analog data acquired by means of the data acquisition device (2) in the form of an uninterrupted data stream.