DE2149717A1 - Procedure and computing system for the automatic processing of data - Google Patents

Description

Method and computing system for automatic processing of data.

The invention generally features a method and a method Computer system for the electronic processing of data preferably according to selectable programs, each accessible under the control of a modular control! or execution program In particular, the invention relates to a computational one Daxenao education or data mapping system.

So far, no data processing system is known, that with modular programs the processing of three-dimensional Data carried out Above all, there is still no system that can work with a centralized control program, which is not in the central core or main memory of the computer system. For processing various types of three-dimensional Up to now, data have always been separately the programming, data transmission or processing, and input and output processes carried out. Regarding the number of possible The known systems are limited exclusively to the core memory. There was still no satisfactory process for treating complicated regularities or imperfections. Obwohidie respective type of data, requirements of preprocessing, special features of the respective technical area and various other requirements must be taken into account, which are conditioned by the special nature of the respective data In any case, the processes are treated as uniform and separate undertakings. No attempt has yet been made the computing system by one that is not in the computer itself central program to control various additional computing processes of selected programmable modules, of which each to perform a specific data processing function

209848/0989

is programmed to be ordered and executed as required.

The invention provides a data processing system that under the control of a central execution program module and additional selected process or processing program modules handles three-dimensional data and provides an output in the form of a map or a grouping field »Der Execution program module coordinates the entire data processing, and since it is not in the computer's memory, is a maximum accessible core storage for each processing program-ilodule possible Furthermore, the ■ programming so eriolgen that from a selected modular manufacturing process a connection is made directly to another modular processing process and a return is possible The data processing system can receive irregularly spaced three-dimensional data Evaluate data and identify them in a regular, single coordinate or trrexi network, whereupon the Unified data values are available for further processing with selected additional code processes according to a separate programming and control by the Execution program Hodui »

Precautions are taken, coordinates for known ones indicate sudden discontinuities. These data are required for the neta formation in order to ensure a correct evaluation of all To ensure nec ^ e in the vicinity of the discontinuity «. The procedure was with excellent results for the evaluation and profiling (contour formation) of geoscientific data applied from areas exposed to a disturbance »

To achieve this, the T'riindung creates an open data processing system that continuously new in the hunt 13t processors, d _o h. Operation modules, increasing in a cumulative manner. Furthermore, the directive specifies a "program for an AD training or control system for three-dimensional data, which is not restricted by the specialist,

209848/0 98 9

BATH

2U9717

"O class the iarten ^ Rospe independent of the memory size and an appropriate level of power or a suitable Interpolationnrate for the respectively considered special Kart are selected enter ei cli can"^"7 eiterhin the invention provides a central Veraroeitungsprograimii-Fodul, the various results subordinate and specialized """elaboration program t'oduie and thereby enables the programs to be carried out with the least amount of effort ₉ without the need for separate inputs into the bucket receives irregular three-dimensional data, which it unifies, for example, to a certain plan arrangement, or three-dimensional data in regular form and then carries out certain data treatments by selecting and executing special Irogi ^ amm modules of the calculation system.

Annaiid of the drawing, the invention is explained in more detail will. Show it:

Pig. 1 a system with various program modules according to the.

Pig. 2 is a block diagram of a for performing the Method according to the invention suitable data processing system:

Pige 3 a sector diagram to explain an invention Type of heta evaluation used j

Figures 4A and 4B are sequential views of the data sampling a npaü-like arranged card genrlns of the invention;

FIGS. 5-A and 5 B show successive views of a map segmentation according to the invention ";

209848/0989

BATH ORIGINAL

-4 ~ 2U9717

6A shows a modular program operation as a block diagram;

Pig β 6b one of the operations according to Jj'ig. 6A the following modular Program operation;

Pig 6C a modular program operation following the operation according to FIG. B;

7 shows a schematic representation of a core memory, how it is used in the preferred data processing system;

8 shows a flow chart to explain the connection process by a computer monitoring system according to the invention;

9 shows a flow diagram of a typical program module, how it is used by the entire bystem according to the invention;

10 is a flow chart of the first half of the operation the execution program module;

11 is a flow chart of the second half of the operation of the execution program module;

Figure 12 is a map table showing the classification and identification of the various used throughout the system Cards shows;

13 shows a flow chart of the first part of a TTetz-1 phase of the network program module;

14 is a flow diagram of the second part of the network 1 phases the programming operation according to FIG. 13;

209848/0989

-5- 2U9717

15 is a diagram of the first part of a Hetz-2 phase of the network program module;

16 is a flow chart of the network 2 phase of the program according to FIG. 15;

17 shows a flow chart of a further network 2-phase program; and

18 shows a flow chart for explaining an LTetz-2B phase of the Netzprogra, nm module

It should first be noted that the computing system according to the invention is able to do much more than just depict or map three-dimensional input data. So the computing system is able to perform two-dimensional Fourier transformations to process seismic section data (as a so-called "Record Section Processor"), to process two-dimensional filter data ^ or in general all kinds of data that can be represented in a three-coordinate system. Mostly there are irregularly beaten Data (in the form of x ·? ·, Y and z values) into the computing system entered. This data can be in almost any form, and almost describe any information that can be printed out in three coordinates o For example, the x, y, and z data values represent the top of a well or borehole, sand thicknesses, Bouguer anomaly values, seismic travel times, velocity values, etc., and the values can also be in different Units can be entered.

_{The given x, s} y and z values, if any, consisting of different units, can in turn be measured in different scales when correcting calculation inconsistencies in the computer program. Regular or uniform coordinate data, as is usually available for seismic section data, can be entered directly. With the keching system there is the possibility of identifying

209848/0989

-6- 2U9717

purposes. BelieDen with jeaem Daxen coordinate value or checkpoint mitgeiünrt a plurality of alphanumeric characters weraeno Such Kontrolipunkt-Indentifizierungswerte and the x, y and z values uefinden in bystem in an external memory such as a magnetic tape-transport unit, so that virtually any any number of such control points is possible in the cysteine.

information is entered into the system in Suwei Grundionnaten angegeoeno The first format ordered from such data in which There does not necessarily have to be a regular, even spacing between any of the x, y or z data values, as above was mentioned. The second format consists of information in which the x-values and the y-values are replaced by some original x value (Zero point) and a y original value and the remaining x and y data values by uniform step spacings In the x and y directions of a plane coordinate system, »Every x and y point on A z-value can then be assigned to this defined coordinate network which also has a special, so-called z-kil representation includes, which indicates that at a certain point of x and y data values at a given point in time no real z-value is available

By inserting the z-values in an ordered rectangular arrangement, which is referred to herein as a card that can (x, y) position of each z-lVertes determined by its position in the array werüen ₀ In this Yieise the information can be used for x, y and ζ are simply stored in the computing system by simply storing the z-values, the x-origin, the x-step distance, the y-origin, the y-step distance, the number of values in the x-direction and the number of Values are stored in the y-direction «In order for data of this kind to be processed satisfactorily, this right-angled arrangement must not be restricted to any pail in the core memory of the computer. To this end, precautions are taken to include one or more of the columns

209848/0989

2H9717

ordering card assemblies in some hot store De, for example in the form of a disk or a magnetic tape zii save ο Each arrangement or card is given its own name to which the parts of the system refer.

The computing system 10 shown in Figo 1 is through a A plurality of program sections formed, namely one section 12 for Singangsverarbeitungj a section 14 for general Processing, a special processing section 16 and an output section 18, all from a central Control program section 20 ", each of the section processes is formed by a plurality of separately programmed processing modules that correspond to the respective Select a desired data operation to perform. The individual program modules are explained in more detail below. The various processes and their corresponding program modules are controlled by the central control section 20, which in turn is under the command of an execution program module (EXEC) with an additional control through an associated library program source 24 and a so-called LIST-MEH ^ E program 26 stands for a list of card header instructions and a flow chart for later reminder, as will be explained below will be"

The computing system 10 and the various programs which the Dilden or enable the system are suitable for various purposes Commercial data processing systems, especially for the so-called Sigma-7 computer from Xerox-Data Systems Inc. This is a fast data processing system, which is particularly suitable for processing seismic data and for the various peripheral i'-units that are connected to used with the overall program of the system 10 described here will. For example, according to the program, facilities such as a random or direct access memory, the so-called Cf ^ convolver part of the system and various belt and slat drive mechanisms used and controlled «It goes without saying, however, that other commercially available data processing systems,

209848/0989

namely the IBM 360 system with the program described here bzwo modular program system can work.

In Fig. 2, a data processing system 30 is shown as a block diagram, which can be controlled by the program according to the invention. It contains a computing unit 32 with the necessary link arrangements and internal storage devices. In the case of the Sigma-7 computer, the Jiechn unit 32 has sixteen registers including 7 index registers (not shown) and a 64K magnetic core memory ₀ K corresponds to 1024 words in this unit, and the coding format requires four characters per word »

The program input in the computer unit 32 via a line 36, a standard design can of a card reader 34 are used _e The output, or the print information from the computer unit 32 can be performed by a line printer 38 for example in the form of a known list protocol apparatus, or graphical output can be displayed using a graphical printer such as that supplied by the California Computing Company. Although not specifically shown, the data can be digitized for input, for example in a Calma Digitizer Model VIP 485 from the Calma Company of Sunnyvale, California "

The convolver part 40 of the system can, for example, be a Type DFE-3 convolution filters from Xerox Data Systems, Inc ». The operations carried out by the CFE part can also be carried out from the instruction set of the Sigma-7 computer, but the speed of operation will be significantly reduced

An outer Hpeicherung takes place in a plurality of disk storage 42, with which the processing unit 32 is connected _e The disk storage 42, standard units, which are supplied with the system of the Bigma-7-computer ₀ Each of them has a

209848/0989

2U9717

Storage capacity of 6 million characters. Another memory access option is via a tape access connection 44 to each of several 9-track tape transport units 46 and 7-track tape transport units 48.

As has already been explained above, a square in the computing system 10 is a right-angled arrangement of numbers which have an assumed origin and distances between them in the x and y directions. The Re.chensystem 10 can handle a predetermined number of cards in a given pass "The cards can be identified by columns stored 48 either in a disk storage 42 or in one of the magnetic tape units 46 or _0. Each card can by a single four-character names which is referred to throughout all parts of the system. Once a map has been created, it can be used as input information for other processes as often as desired. The disk space on disk drives 42 can be re-used if the same card is used for subsequent output cards, but the previous contents of the particular card are lost. Any number of cards can be used in subsequent job runs on magnetic tape units 46 or 48 be preserved. The effective program module relates to a card according to its four-character format. Input / output routines search for the required card located in the modules 46 or 48 or in the disk drives and ensure that it is read directly, such as it is requested by the current processing program module of the computer system 10 "In a given job process or finite arithmetic process, a card can also be transferred several times from the tape units 46 or 48 to the disk storage 42 if it is required as input information."

For identification, each card also has a sequence file containing the event sequence with it, which specifies precisely how to create this special card

209848/0989

2Η97Ί7

To rather identify run file records in disk storage, a special key is provided which has a specific value for each column or sequence file record Disk storage 42 "

Kin disk key, which consists of four bytes, namely byte 0, byte I _8, byte 2, byte 3,

has the following statement:

Α «Byte 0 indicates that the record is an expiration file record for a special card (Byte 1 will be 0) »

Bo Byte 1 indicates that the record is a card column for a special card (byte 0 becomes Be 0).

0 · Bytes 2-3 designate the map column or sequence file recording to which the key is assigned » Bytes 0 and 1 are hereinafter referred to as "key specification" «

D. This means of identifying columns of cards is one way of explaining or doing everything Columns up to 55 cards with a 32-bit word.

Up to 55 Cards are located. Any number of cards within a limit the tape storage capacity may be reduced to the reserve tape storage spaces «The program modules requesting them can refer to these cards as often as they like is full, there is a possibility of overflow from the disk storage to tape storage,

209848/0989

2H9717

A column of a particular card is obtained by calling an input / output routine where it is specified is in terms of key, disk or tape storage, and field position in core storage where the column should be placed. A subroutine checks the key to see if the card is in disk storage 42 or tape storage units 46 and 48, and then locates the appropriate column for the specified card. In Cards are issued to the system in a similar way. Byte 0 of the key is processed by the EXEC module (execution program module) “Then the information connection to the various modules is determined by the global variable area of the inner one Core spear determined.

The global variables are set before the job starts (of the various work required) entered into the system »They consist of numerous coded values according to the following compilation:

Global variable puncture

Program modules

D. ITU 2) IOU 3) IGP 4) IR2BIS 5) IRBIAS 6) IWBIAS 7) IPBIAS 8th) NMAP

- Unit number of control cards for each module

- Unit number for normal print output

- Unit number for the control point memory (.Band)

- Reading specification for secondary card

- Reading specification for primary card

- Write specification for issued card

- Pilter specification

- Number of cards in the buffer

- Filter scale multiplier

EXEG EXEG EXEG

EXEG EXEG

Output modules

209848/0989

2U9717

Global variable

10) GSIZX

11) GSIZY

12) ZSCL

13) ZBIAS

14) GSIZE

15) XOR

16) XLM

17) YOR

18) YLM

Note:

19) KROW

20) ITCOL

21) ZMAX

-12-

function

- X-scale for the map in units / network

- Y-scale for the map in units / network

- Z-scale for the control point values

Program modules

EXEC EXEC

INPT or INPS

- An additional default that can be used when the data values have low contract and polarity in common INPT

- Network size in inches per network EXEC · * X origin in units for the processing area EXEC

- X-G-renze in units for

the processing area EXEC

- Y-Urepresentation in units for

the processing area EXEC

- Y limit in units for

the processing area EXEC The input module works with a different scale and a different origin in order to get from read scales and origins to the required units _o The scales defined under (15-18) apply to the possible output cards.

- Number of lines in the EXEC card

- Number of columns in the EXEC card

- Maximum Z-values of the control points INPT or

INPS

2098A8 / 0989

2U9717

Global variable

22) ZMIK

23) NGC

24) KEYOVR

25) KEYTP

26) IHBIAS

27) IWKSPB 28) * MAPI1

29) * MAPI2

30) IERR

31) MPO

32) IZMNG1

33) IZMXG1

34) ZS0LG1

35) IZMNG2

36) IZWXG2

37) ZSCL & 2

38) IZNL

!Function

Minimum z-value of the
Checkpoints

Number of control cards for the current module

Key at which the change from RAD to tape storage occurs (-1000 indicates that RAD is not yet saturated)

Key type - 0 for new key - 1 for old key for output card

Specification for output card attachment

Program modules

INPT or INPS

EXEO

EXEO

Workspace specification
Name of the card corresponding to IRBIAS

Name of the card corresponding to IR2BIAS

Set to 1 when an error is to be eliminated

Number of words per column Minimum packed Z value for card according to IRBIAS EXEG Maximum packed Z value for card according to IRBIAS EXEC scale (multiplier) for card according to IRBIAS EXEG Minimum packed Z value for card according to IR2BIS EXEC Maximum packed Z- " ^v ert for

Card according to IR2BIS EXEC scale (multiplier) for card according to IR2BIS EXEO

Fixed point value for networks that are not evaluated (in the current case X »FFFF8000») EXEC

EXEC

EXEG

EXEC

Running module

EXEC

2098Λ8 / 0989

Global variable

39)

ISORG

40-43) EBCDIC

44-47) EBCDIC

48-49) EBCDIC

50) IZMt

51) IZM2

52) HXWD

53) IXOJND

55) ZNIL

-14 function

Program modules

If it is equal to I ₉ , the trlobal values 10, 11, 15, 16, 17 »18 of the (geographical) latitude and longitude must be converted into special units»

Representation of a system conversion EXEC

Display of time and date of the current run EXEC

CONMAP billing or account number EXEC

0 if there are no NZIL ^ values in the card is in accordance with IRBIAS;

not equal to O, if the card contains ZNIL values EXEC

0 if there are no ZNIL values in the card are according to IR2BIB;

0 if card contains ZNIL values EXEC Number of extra words at the end of each column EXEC

0 will expand the map by copying boundary lines and boundary columns to the outside be to filter delay

Output module

to compensate

0 will expand the map with values zero.
Running presentation index key

Sliding Nothing (Nile Control Points)
(in the current case - 1.0 * 10 ** 30)

209848/0989

2H9717

The (ilobal values 201-260 contain the current header block.

The global values 261-450 are available in the event that additional variables are exchanged between modules have to",

* These values are only set if the card is in Permanent tape storage is located. The user module should not make any reference to them

* Running module suppresses the operating module if an error occurs occurs-o

209848/0989

2H9717

The EXEC program module 22 determines from the input information the size of the card to be processed and other critical parameters. This information and the other critical ones Data is stored in a communication or (iio balspeicherbereicli of the internal memory for access by other modules set or programmed. The various primary functions of the EJiEC program module 22 include the following processes:

1o Finding all cards and putting them together or programming the information required for input / output routines in the various modules with module sections 12 to 18 of the computing system 10.

2 β maintenance of the process files for all output cards, each of which describes exactly how the special map was set up.

3. Checking the existence of all necessary Modules. Since the riodule used by the surface system 10 perfectly are flexible, the disk stores 42 are examined to see if all the required modules are present. Is this if this is not the case, the work process is aborted without taking up the time of the computer.

4. Entry of all control cards for all processes to be checked by the EXEC program module 22 “All necessary information is obtained through the EXEC module 22 «, then the necessary Card tables, job or work schedule tables etc. built up,

5. The EXEC program module 22 brings the selected first module of a work process (job) into information connection with the computer. In the connection process, the execution program gets a fixed point in the disk space and the location is kept where the execution was interrupted. The module _T for which a connection was established is entered in the core memory and superimposed on that area where the EXEC module 22 was previously located. Then the

209848/0903

Operations of the new module program carried out, the EXEC module is saved again, and the execution in the EXEO module 22 begins at the point where it was interrupted »Other subsequent program modules can work in the same way be associated until the operation (job) is completed. These connection processes are discussed below are explained in more detail.

The modular structure of the system allows arbitrary additional processes for one of the four main functions of the system perform, namely input, data processing for general purposes, data processing for special purposes, and output « The following series of modules is just one of the many possibilities o

An nrPS module 49 forms an input program which irregularly spaced control points of cards, tape or disk storage reads o There are several with this module Methods for entering control point information into the system., The required coordinate information of the x, y and z-data can reside on almost any storage medium, and from input it can be almost any form. The input medium can be a tape that is known when using a Device, such as the Oalma digitizer or cards, FOliTRAN output tapes, etc., purely receive every control point Any identification of up to twelve characters is possible for a basemap assignment or for other purposes Geographical marks such as the corners of an urban area can be entered. Different methods for creating scales and for TJrjump implementation are also provided in the program steps

The control points are read from the corresponding input medium, for example the nine-track tape unit 46 according to FIG. After üKaienbildung and implementation of t unkte the x- and y-coordinates for the map borders are

2 0 9848/0989

-18- 2U9717

examined and if all parameters are correct, the x, y coordinates are converted into meshes, after all control points has been processed and written to a suitable medium (e.g., tape or disk), the data is under control of the program of a SORT module for general purposes (such as will be explained in more detail, whereupon the control point data sorted by columns. If necessary, you can scale factors and original conversion values are specified. For some input forms, the scaling and transformation of the supplied independently of the converters (or subsets) of the data.

The true control point values are determined according to the following Calculated formula:

^z t = ⁺ «" default

Here ζ mean the value on the tape and z. den wanren zv / erto The system also works with z-values of the size Eu 11 (z-Nile or nothing values), which can be used for white areas or earth boundary conditions Sorted order, whereby all control points in the columns are sorted in increasing order by rows. The sorting depends on the Uetz size in the x-direction and y-direction. If the units per batch are to be changed in one of the two directions, a different sorting is required

Has the currently used control point information following format:

Word 1-x-value in floating ^ notation; Word 2-y-value in floating, notation; Word 3-z-value in floating, notation,

209848/0989

2U9717

The identification is arbitrary, and the initial character is in a specific location in the tape header or header specified. Also indicates the number of characters in a "particular word of the header record. A zero display there is no identification for the number of dykes. Each checkpoint is marked by a record on the Komsrafl. ·! - point tape shown, and each record must be a multiple of four bytes.

A head record of a checkpoint tape can be compose as follows:

A. The ports 1-40 contain general header information, which are determined by the current HDR1 and HDR2 control cards.

B «. Words 41 to 60 are used for parameters of the following reserved;

41. Number of lines in the card

42. Number of columns in the map

43. link size in inches per race

44. X-scale for the card in units / netζ

45. Y-scale for the map in units / network

46. X origin in units for the map

47. Y origin in units for the map

48th ji limit in units for the card 49 · Y limit in units for the map

50. Z-scale (sliding) divider for control points

51. Z-specification (sliding,) - addend for control points i) 2. Minimum control point Z- vert 53 c maximum control puiiKtz —

54. initial of control point identification 55 number of characters of identification (12)

56-5y _e reserved for future needs 60. Ά alii of the control points on aem tape

209848/0989

BATH ORIGINAL

_2o- 2H9717

A module operation can generate data in the form of irregularly spaced x, y and z coordinates · You can click on Magnetic tape is stored on the disk device and after its input through all phases of the system "or the one described here Process are processed through. However, before the irregularly spaced data is recorded, it can pass through Modular information (modular options) with regard to the formation of scales and data translation brought into a new coordinate setting will"

To process the irregularly spaced data becomes one through a two-dimensional interpolation and extrapolation from the irregularly positioned control points right-angled arrangement of data or map values is calculated, as already explainedβ the interpolation-extrapolation process is referred to as meshing and is performed, for example, by the GRID module 50, which is located in the general processing program sector 14 of the computing system 10 is located. The network formation provides values on the networks that are close to the control points fit within the appropriate areas of influence, and a sufficiently monotonous surface in areas exposed by the Controls are removed so that optimum results can be achieved o In these cases it is essential that maximum one control point per network is allowed, and most types of data require area data to be networked that do not necessarily fit into the core memory of the computer

As an option, the system also offers the option of using coordinates in the field when creating a network for sudden coordinates Insert discontinuities · This can help in the formation of agitation natural discontinuities or boundaries are taken into account in the interpolation process and anomalous details such as geographic Discontinuities (imperfections) are represented by the recording system.

209848/0989

2H9717

In carrying out the two-dimensional interpolations under the control of the program of the GRID module 50 in the general processing program section 14, the GRID module 50 supplies a uniform grid network based on irregularly spaced data. Missing values are also filled in at the same time due to the surrounding mass or intersection points of the network. The GRID module 50 works mainly in two phases, namely a first so-called GRID-I program and a second GRID-2 program. known data values in the form of x ~, y and z values _o The GRID-1 program then carries out the following steps *

1) Calculate the centroid of all control points in a given network;

2) Divide the area around the centroid into eight equal sectors with the centroid or control point as Center, as shown in Fig. 3;

3) Examine each sector to find the closest control;

4-) Use the closest control in a method the evaluated least squares to compute the best plane through the data; and finally

5) Use the equation of the plane to determine the value at each of the four dimension points »

Only one control should be used that is in the same fault or store part block as the focus. If an impurity crosses the network, every part of the impurity block must to be charged separately.

209848/0989

_22_ 2H9717

Describes the GRID-2 phase of the operation program then the real control and then calculates the remaining mesh points based on the nets that go through the G-RID-1 program phase had been determined. The G-RID-2 program performs the following steps:

1. Set the minimum sector requirement "ITS" to 6; 2β Find uncalculated agitation in their neighborhood there are at least three calculated networks;

3. Create eight sectors with the non-computed network as the center;

4 · Examine each sector for the closest ones calculated networks, ignore the neighboring control;

5o Palis at least "NS" sectors have a control, fajjTe away, and if not, leave the net unaccounted for;

6, Combine neighbor control and sector control, to maintain the value of the network;

7, This operation will continue until all of the above Conditions corresponding networks have been calculated; and

8, If a concave polygon has been specified, the process stops after the need for six meshes is exhausted is.

If a convex polygon specification (option) has been specified, the six-mesh requirement is reduced to five and the The entire process is repeated when the entire card is hand-made is calculated to the edge, the network requirement is reduced to four, and the process continues until all networks are calculated. As in the G-RID-I-Pall, there is only one control in the same impurity block used in which the calculated network is located,

209848/0989

If at the beginning a map partially designed as a scrap is available, it is possible to directly access the GRID-2 program. The GRID-I and GRID-2 phase programs are explained in more detail below using flow charts will.

Programs include three specifications (options) that provide control over the degree of extrapolation _o The first option calculates only the incitement that lie in a polygon, following which tightly aufden circumference (perimeter) of control »The second option only calculates the Networks that lie within a convex polygon that defines the control. The third option calculates all networks. Yet another selection parameter allows the network formation to be terminated due to the convergence of the process, and a further parameter controls whether an error process (troubleshooting) should be used in the calculation and control.

An essential advantage of the invention described here is that, in contrast to most two-dimensional network processes, the area to be calculated does not have to be fully and / or continuously contained in the core memory. For example, consider the case shown in Fig. 4A where the card consists of L columns and a minimum number K of columns will fit in the core memory. For optimal results, the overlap area shown in FIG. 4A axs section m should contain a control point within sector 1 for each network along the boundary between η and m _Q. At some point the width of the area m could be limited due to the distance assessment used in the meshing. In practice, a width of 20 columns for the area m has proven to be sufficient for various types of data. The area K is then subjected to a network formation as a separate map, so that the areas j and 3 complete columns are output.

moved so that the areas j and η result, which are

209848/0989

«24-

2U9717

Area N is shifted to the left, and the columns from area M are reinserted in their original form, since nets extra-polished from the first pass or operation are only based on data from the left. Then J new columns are added, resulting in the new configuration shown in Figure 4B. The area n _Q is used for checking to the left of area m, so that continuity across the segment boundaries is guaranteed. The areas m, J ₁ and m .. are then output as completed columns. The area m _{2 is} entered into the core memory in its original form. The process is repeated until all columns have been completed. In the last step, all columns that have not been previously written are output.

Of the numerous program modules indicated in FIG. 1, the execution or EXKC module 22, the INP6 module 39 and the GRID module 50 were mentioned in particular. The contour or ONTft module 52 in the output program sector 18 should now also be particularly emphasized. The ONTR module draws contours for a specified map through the network of individual networks. Kontrolipunictorte, the identification of z-values and geographical brands can optionally be distributed. Contours can be drawn for the entire map or for any area within the map. It is also possible to enter control point information without drawing the contours. The module also opens up the possibility of adding comments about their values to the contours.

So that maps outlined with the desired flexibility that have too many networks for the available core memory of the computer, or so that cards are possible, that are either too long or too wide for the paper of the recorder is one method of automatic segmentation intended. FIGS. 5A and 5B serve to explain this map segmentation. As shown there, enable

209848/0989

Horizontal subdivisions of the 'output card a consecutive Output of the vertical sectors to the output printer. The map segments fit together exactly at the borders So theoretically it is possible to use any "arbitrary" card To create dimension

According to the program of the GNTR module 52, the network examined for the intersection points of all contours · The (x, y) point where each contour intersects the mesh is examined recorded. Then, in a sorting process, all contours with the same value are connected or combined, whereupon optionally the contour can be smoothed by means of a curve fitting process. The amount of overlap on the boundaries of a segment is a parameter that can easily be changed in the program. It is currently at three or more times set from this so that there is a suitable adaptation to the contour segment limits

The ONIER module 52 includes options such as the following Control processes:

1 »Options for entering controls, markings an identification;

2, Annotation options that specify (a) which contours, (b) how often along the contour;

3 · The contour interval, which can be specified as an absolute number or as an approximate number of contours over the area of the map;

4 · Contour boundaries that are within a range of z-values can be restricted;

5. Segment-shaped (χ, ν) boundaries for the format formation when drawing outlines;

209848/0989

.26-2H9717

6β Interpolated smooth contour that can be optionally drawn can be; and

7β The option to display the map on an IDI cathode ray tube device in connection with a Sigma-2-0 oscilloscope, an XDS peripheral device, instead of the OAIOOMP printer "

The various processing modules or programs in the various program sectors 12 to 18 each perform performs a special puncture of the overall signal processing operations. The input program sector 12 thus contains an INPT program module 54 »which serves as a general entry process, but also for special use in a geophysical one Gravitational Process »A storage routine is carried out by a SAVE module program 56 through which the card in random or direct access disk storage to a selected one of the Bazjd memory units 46 or 48 written and there stored for selective later retrieval. The INPS program module 49 is used, as has already been explained, for a more general input for reading irregularly spaced Checkpoint data from maps, baghd records, or a disk file · An OOPY module 60 causes a previously Card stored on magnetic tape in the random access disk storage is read.

In the program section 14 for general data processing, in addition to the Hetz module 50, there are further modules for one General signal processing provided. A sort operation 62 can be performed on a commercially available peripheral device programmed in a known manner. An NSFL module 64, a Non-standard filter program, creates special filters with selected ones Switching frequencies and / or averaging the area. A TREN module 66, the trend surface program, ensures rather surface (currently 1 / 13th order) for adaptation the operation control points, and it performs a computation

209848/0989

-27- 2U9717

through each network point. The FLiR module 68 is a comprehensive program, a two-dimensional based on a room filter result binding (flat winding) performed

Another processing program, the APLN module 70, can be called up to form flat surfaces that have selected locations and values. A $ SAM module 72 which is an algebraic The manipulation program is used for various operations such as scaling, subtraction, addition, multiplication, division including logical operations for transferring Limit restrictions «The. $ FFT module 74 is used for frequency processing, since his program transfers rectangular data arrays into the frequency domain, and he also performs some operations for retransmission into the spatial domain.

In addition to the Eontur or GHTR modules 52, the output program section 18 also contains several output program modules. An OSED program module 76 can be accessed when there is a desire to show the position of the surface along any profile in the map area. Retention Routines are enabled by the SAVE-J.Iodul 78, whereby a card is written to tape in the direct access memory, so that it can be used at a later time. A printer display b-Irogranmi or PPiBP module 80 initiated the printing of values on the map at each network point. A printer contour program or .PGNT module 82 divides the range from Card values of ten units and forms a printer representation using one-digit values

The program section 16 is reserved for those program modules that work according to special processing methods. A remainder calculation program, the RESI-LTodu! 84, subtracts or adds two cards, this addition being more specialized and faster than in the aforementioned IbAK- ^T module 72. A DERV module 86 consists of twenty-seven

209848/0989

-28- 2U9717

Standard line filters for use in processing of gravitational data. A program for a three-dimensional gravity model is also included in M0D3 ~ module 88, that uses polygonal plates or strips to construct a three-dimensional model and that of that model An MMPO module 90 calculated the gravitational effect caused Constructs a three-dimensional magnet model using right-angled quills and then calculates that of this Model caused magnetic anomaly o

A seismic migration (migration) program is included in the $ MIG module 92 * This operation provides below Using a time map and a speed map, a shifted horizon representation. A VEIKAP module 94 enables the speed mapping of seismic input data A data manipulation program, the DMAIT module 96 uses selective data from the database, whereupon questions are asked by data groups and resulting answers into the computing system 10 are introduced. The VSEM module 98 is a data retrieval program, which retrieves groups of data selected using the database. These selection processes contain area distribution and / or other identification parameters o For copying groups of information from a card onto a special tape that designates the database a data storage program is available, namely a νσΤΤ module 100.

Fig. 6A ₉ 6b and 6C illustrates in functional block representation a part of a processing operation is. The inner core memory 110 of the computer according to Figo 6a are input, the global variables in a global area 112 of the memory core 110 while the monitor data of the computer constantly in Area 114 of the memory 110 can remain o The remaining and larger core memory area 116 is for the periodic relocation of foreign, i. _E. reserved for data not in memory

209848/0989

_ ₂₉ - 2H9717

Storage area 116 the program for the execution operations module (EXEO module 22) · Several other processing programs there are all module operations TTo. 1 in one program look 118, as a module operation No * 2 in a program block 120, and so on up to a module operation MTo. K in program block 122. The module operations of the program blocks 118 to 124- only designate in a general form the repetitive programs within of program sections 12, 14f4.6 and 18 of Fig. 1. Each this program is completely stored with the correct address in disk memory 4-2 (Pig 2)

6B shows the part of a process in which the execution program has been pushed out of the core storage area 116 for the purpose of a checkpoint storage at the disk storage location 126 and at the same time the program of the module operation No _β K has been entered from the program block 122 into the core storage area 116.The process of the program block 122 or the module operation K is carried out regardless of the supplied data, while the EXEG program is kept at the checkpoint memory position 126. After the completion of the program of the module operation No · K in program block 122, the execution program is read back into the kernel storage area 116, The program for operation K in the core memory is destroyed. The program for the module operation K is still available in the disk memory. The execution program is now re-entered into the core memory area 116. Thereafter, a next selected data processing operation or a corresponding program can be carried out in a similar manner, the execution program being pushed out to the checkpoint memory position 126 and the selected program being pushed into the core memory area 116 to store the data treat in the way you want.

A combination of consecutive processing operations or program modules is also without the insertion of the execution program relocation possible. This process, which is partially carried out in

209848/0983

~ 30-2U9717

6C could be referred to as a two-level connection. The first time it connects, the executor pushed out of the core memory portion 116 into the checkpoint memory position 126 during the first selected program, the Module operation K, or program block 122, into the core memory section 112 is pushed in order to execute or effect the data processing instruction »Then a program instruction can be added to a next selected program or the module operation H of the program block 124 in the core memory portion 116 of the computer to push without the executor returning.

A connection from other operating or work modules must be specially programmed and in this case is located the module operation N in the core memory part 116 changes to another To effect processing of operational data. If the execution program returns from checkpoint location 126 If the core memory part 116 is brought into this memory part 116, the module operation program is first destroyed N the program of the module operation K and then the execution program, with the destruction of the module operation K, pushed. The programs for module operation Bf and K remain, however in disk storage, i.e. in program blocks 122 and 124. A connection from the execution program to subsequent ones Modular operations are better known through parameter waiting Kind of controlled in a known manner. A connection from other operating modules to subsequent modular programs must be special programmed.

Fig. 7 shows a magnetic core memory of the type as he

is used in .XDS-Sigma-7 computers with which the here described Procedure is carried out. It is a 64K core memory with an entangled order in one half of the storage locations. This means that 32K storage spaces are initially available in equal halves are divided, which, differentiated according to odd and even, are alternately accessible, while to the rest

209848/0989

-31- 2U9717

Storage locations can be accessed in regular order. The size and type of the core memory is irrelevant since the initial programming ensures correct data handling, however, it must have a sufficiently large capacity of at least approximately 48K in order for it to be suitable for the present Invention is suitable.

Figure 8 shows a flow diagram of the generalized module connection process. The connection process arises from monitor functions of the computer. The corresponding monitor data are located in the monitor part 114 of the core memory 110 of the calculator. The linking process flowchart shown in FIG is per se an illustration of the process shown in FIGS. 6A and 6B. First, the computer monitor is underloading Use of a control card (a [! RUN (MN, EXEC) J in the case of of the Sigma-7 computer) the EXEO program module from the disk storage and initiates the execution of the EXEC module. The monitor part 114 of the core memory thus contains the correct addressing for the start of execution, whereupon the execution program initiates the next step 150 in order to execute the execution program (in the core memory section 116) to be copied to disk, d «, h. to write to one of the disk stores 42 (FIG. 2) «.

Thereafter, the program begins to bring any module to be connected into the core storage part 116, as with the stage 132 of the flow chart is shown «, the next stage 134 controls the update (update) the input / output control blocks based on information or data changes on ASSIGN cards (assignment cards) are located. According to step 136, the data status is then changed in such a way that that the execution of the selected module is initiated, i.e. all data settings for the start of the module operation be made ready «,

209848/0989

2H9717

«32-

Figure 9 shows the flow stages for the typical operation a program module, i.e. other data operations than the execution module 22 in the central control sector 20. At stage 140 the program module is entered, whereby the program input data are located in the core storage part 116. In the next step 142 of the flow chart, they take place in accordance with the special computer system sub-routine calls to all necessary registers for the special modular operation. Then the remaining input / output devices become put into operation.

In the step 144, the global variables are obtained from the global memory part 112 of the core memory 110. Then the program continues with data processing as requested by the particular module. After completing the During the program run, the program reconnects to the execution program at step 146, which is shown below Destruction of the previously used module program (which is still located in the disk memory 42) back into the core memory section 162 is pushed.

Figs. 10 and 11 show a basic flow chart of the EXEC- Module 22 (FIG. 1), which exercises the central control and general organizational function in the computer system 10. As input information the EXEG module determines the size of the to be processed Map and other critical parameters »All these parameters are programmed in the core memory so that they are accessible from other program modules. From the EXEC module all cards are localized and all information for input / output routines set or programmed in the various modules. He also ensures the maintenance of the Sequence files for output cards, controls the test according to the Presence of all required modules, executes the entry of all control cards and creates a connection to the required ones Modules in the prescribed order.

209848/0 989

The EXEC program is loaded into core memory at stage 150 of the schedule. At this point the computer monitor is in control, since it addressed the program in the core memory of the computer. Then, at step 152, the execution program takes over control of the data processing. The program reserves at least one page of the core memory for receiving the global variables compiled above. The reservation is made in level 154 «. Space is provided for 512 data, which corresponds to a conventional page ₀

In the next step 156, the invariant global variables are also brought into the core memory. After that will be information read from the job control card at step 158 is brought into core memory, while at step 160 the various Global job constants are inserted "

The job control card carrying this information is in the standard form of a code card. It carries the following information:

Columns 1-3: JOB; this identifies the sequence when creating the program modules.

Columns 5-8: GSIZE; this represents the network size in inches for all output cards · This value can be left out (blank) and separately in individual output modules are fed.

Columns 9-12: no code; if this field is not empty, it is assumed that all remaining fields are on the job card Contain information in the form of decimal degrees (latitude or longitude). They are from an input module in the same Way converted like the input data.

Columns 13-20: XOR; this represents the x-origin in selected units

209848/0989

-34- 2H9717

Columns 21-28: JCLM; this represents the x-limit in selected Units.

Columns 29-36: YOR; this represents the y origin in selected units.

Columns 37-44: YIM; this represents the y-limit in selected units, _etc.

Columns 45-b2: YGRDB; this means map units per network in the x-direction and, if negative, it is assumed that the number of columns takes up the space of the field ₀

Columns 53-60: YGRDS; this represents map units per network in the y-direction, even if this is negative is assumed that the number of lines occupies the space of the field.

It should be noted that if XOR, XLM, YOR, YLM are empty, an input module (which inputs control point information) will return values equal to the lower left point (i.e. XOR, YOR) and the upper one or right boundary point (d _o h _o XLM, YLM) are.

If the information has been entered correctly and the EXEC program is being executed, according to the square brackets 162 are used to produce the operation tables, as well as the card table, the number of control cards per operation table and all other table forms that are used when printing The control card data is then stored in disk memory 42. The control card reading process is continued for each subsequent control card

In addition to the job control card, a large number of other control cards will be included in a typical job or system operation _o In a typical job chosen as an example, the following control types can be used in sequence:

209848/0989

2U9717

JOB; this card would set the size and be special Data regarding XOR, XIM, YOR, YlM, XG-S and YGS (origins, Limits or network size).

HDRI; (a map); it contains descriptive data that identify a map of a geographic area (e.g. of the "Alaskan North Slope").

INPS (one or more cards); this module reads. Checkpoints and bring them to the control point tape.

GRID; this map or maps calculate the network of the individual networks from the control points.

ONTR; the contour of the network is formed using this map.

HDR2; a second description card for identifying a geographic map (e.g. "Map of the Foundation CI-200 Meter").

NSFL; this program filters those trained as a network Map after the creation of a particular elliptical filter.

ONTR; this program forms the contour of the filtered Data.

END; this card marks the end of all control cards: it terminates the control card input and directs the various ones Connection executions.

After all the control cards have been processed in step 162 of the flowchart according to Figo 10, determines a decision step 164, if all modules are available ₀ If this is not the case, an abort (ABORT) is signaled, and is determined at the step 166, a corresponding Print out «If all modules are available, however, a YES output signal is applied to stage 168 in FIG.

209848/0989

-36- 2H9717

Prepare fixtures.

After the preparation in step 168, the computations run away. The functions in square brackets 170 in Mg. 11 are carried out under the control of the execution program module. The EXEG program reads and checks all control cards they used to determine the number of control cards for each Operation and the various modules required to complete the job. The control cards have the following format:

Columns 1-4: module name

Columns 5-8: Input card name (can be empty) Columns 9-12: Output card name (can be empty)

Columns 13-16: If secondary input card (optional)

Columns 17-80: determined by user »

Any number of control cards can be used for each program module according to the designation in columns 1 to 4 of its corresponding control card or cards. For all cards with the exception of the start card of a module, columns 5-80 can be used in any way according to the preprogrammed by the user. »A modification" END "in columns 1-3 will indicate the end of all control cards. Any number of files can be placed after the end card. Names for new program modules that are executed under the control of the computer system 10 must begin with a $ sign {$) if two input cards are required for the process, i.e. $ SAM, $ FFT etc. "

If an irretrievable or irrecoverable error occurs, stage 172 provides a YES command to the output stage 174 to exit the system and stop operation o If there is no error displayed, the Random access disk storage is checked for overflow in a decision stage 176, whereupon it is required in stage 176

209848/0989

in case the overflow belt will call. Become at level 178 then the space of the direct access disk storage released and all input / output devices turned off. After completing the required for the present work process Operations of the stages contained in square brackets 170 become control again according to stage 180 returned to the monitor »

To explain the manner in which the various cards for the particular programs or modular operations are stored, FIG. 12 shows a card table. Each card is identifiable by a header or header block and a sequence file (the meaning given above), consisting of forty words of the header information, fourteen parameters describing the card, and all control cards that have contributed to the construction of the card.The forty words of the Header information contains the general header information determined by the other parameters. They are determined by additional code words and represent, for example, the number of rows, the number of columns, x, y and z scale values, the origin, limits, etc. After the map has first been generated by its corresponding program module the card information is stored in core memory as an integral part of the EXEC program. J) The corresponding key specifications for the columns or for the sequence file are each stored in the form of a hexadecimal code for each card. In addition, the coding contains an output identifier that indicates whether the respective card has already been written or not.

jj'ig. 13 explains the GRID module (network module) according to Invention · This program is used to calculate networks of predetermined Big ones that contain checkpoints depending on the circumstances or not. After the GrRID program module 50 (Fig. I) has been loaded into the core memory, whereby the EXEC module located there is pushed out to the disk storage, the data processing continues according to the G-RID program

209848/0989

-38- 2U9717

Access to all global variables and input control parameters. The data processing flow passes through from the input 190 preparatory stages, ie preparation of input / output devices, Iese control card, reading checkpoint header, sets the convergence criteria, etc. ₀ according to the successive steps 192β If data is present, which belong to a fault condition, are entered at decision stage 194. At step 196, the preparation or initiation of column storage o occurs

A decision stage 198 then determines whether the network | The formation of irregularly spaced control point data must be based. If the answer is no, it is assumed that the Entry for network formation according to level 199 of regular spaced control point data with missing values. The flow of data from stage 199 then continues directly to decision stage 200 where the available core storage space is determined relative to the map size. At stage 202, if segmentation is required, the necessary Information stored in the global area · The next Connection is made to the GRID-2-B program, which will be explained later will be programmed and the column segmentation operations will be programmed. Then the computer control is connected to the EXEÖ program module returned.

If the decision stage 198 reports irregularly spaced control points for the formation of a network, the Checkpoint data at stage 206 directly into core memory loaded, for the subsequent input in a GRID1 phase at stage 208. This stage 208 for the GRID1 phase is more precise shown in FIG. The GRID4 phase of the GRID module is calculated all mesh points for nets that contain at least one control point. If a fault crosses the network, only those mesh points are calculated that are in the same flaw or error block as the control point. the Mesh points are calculated by having a "plane of the

209848/0989

2H9717

least squares "is determined by each of eight sectors to the evaluated control point around by the next control points. All of the control points, d _e h. usable points must be in the same block error.

If there are fewer than three control points, the corresponding mesh points are averaged calculated. Execution of the GRID1 phase requires that the process has space for two columns of nets in the computer at a given time. If not all control points in core memory A remnant of control points around the calculated mesh is retained. Because the control points are sorted by columns and in an ascending order of columns can be processed, columns (control) can be dropped for the above-mentioned purpose and new columns can be added later Can be added to the course of the process.

According to the flow chart shown in FIG. 13, the data processing result is passed from stage 208 via decision stage 200 in order to determine whether segmentation is necessary _9, as already mentioned. If all columns fit in the core memory, the data passes from a second, opposite output of the decision stage 200 to a stage 210 in which all columns of data are read from the disk memory and the No-Z values (Z-NiI (nothing) values) are filled in will. The data then enter the GRID2 phase corresponding to stage 212, which is shown in more detail in FIGS

According to the flow charts 212a, b and c of FIGS. 15 to 13, the GRID2 module accepts output data from the GRID 1-phase module or data of any other form if the map contains values at regularly spaced mesh points between the No-Z values, i.e. missing values interspersed at mesh points sin,:. The purpose of the process is to provide a value for the no-Z steep to be calculated based on the surrounding calculated mesh points. In this process, all irregularly spaced inspection

209848/0989

-40- 2U971.7

values deleted so that only the control point values at the regularly spaced mesh points remain. After new points have been calculated, which are the no-Z steeps replace them, they can be used as control point values to calculate other no-z digits as well. The process enables i.e. the recursive filling of blanks or no-Z-places, with the area * that is closer to the valid control points are to be filled in first.

When all of the mesh points for the map, including the no-Z digits, fit into the core memory, a single process is performed. However, if a card is too big to fit the core memory, the process must have a connection to another module (GRID2B level 204, as will be explained later) in order to carry out the necessary calculations «. Both processes perform the same basic operations, but the GRID2B process must control the amount of overlap required between core memory segments when pulling the mesh points _e

The basic process controlled by the GRID2 module for calculating a no-Z digit runs as follows:

1 * An orderly search is initiated for a no-Z digit with sufficiently close control values, doh. at least three neighboring nets that are within a two-nets distance either above, below, to the left or to the right of the nothing position. All control values must also be in the same fault block as the calculated nothing position.

2e After a job has been found, whichever satisfies the above condition becomes an area consisting of eight sectors around the mesh point-nothing point Initially formed, at least six of the eight sectors must contain control values, excluding the nearby control points,

209848/0989

-41- 2H9717

Again, all control values must be in the same error block.

3 If these two conditions are met, the sector control is combined with the proximity control, to calculate a value for the nothing position. If one of the conditions is not met, the nothing position remains uncalculated.

4 · The process is repeated until all possible networks are calculated using the six sector criteria. If desired, the sector limit can be reduced to five and an even smaller number for the repetitions of the process will.

The processed data of the GRID2 phase is then fed to a stage 214 which contains the completed columns and causes control to return to the EXEC program ο The output of the data is initiated by stage 214, that at decision stage 200 is a positive or affirmative one Decision was made, that is, in the event that all columns of data of the processed card in the core memory matched and it was not necessary to perform an automatic segmentation when outputting the generated map

The opposite decision of decision stage 200 requires the execution of the GRID2B program, so that connected is that it controls the segmentation of the columnar data output «This means that it is necessary because it is not all columns of map data can be brought into memory at the same time, the data in vertical segments or groups of columns to process and output information separately one after the other

209848/0989

2U9717

According to the flow chart shown in FIG applied a column group iteration method "in which Data from global storage area 202 is processed in stage 216 to prepare the column data relative to the required global parameters. Then the available core memory filled with columns of partially calculated data. Fault or imperfection coordinates stored in the global area become the Data is added prior to treatment in stage 218 to collect all data in column form and the data as completed Output columns. When all columns are completed, decision stage 220 allows a final output so that with changing the number range (scaling) of the columns to the correct value and storing the required variables can continue in the header of the card (steps 222). You then return to the EXEC program module, such as shown in FIG.

If all of the columns are not complete, as indicated by the discrimination level 220, the data is converted into a cycle through steps 224 again to derive the column overlap zone and insert a predetermined Number of new or additional columns in core memory for data processing through stages 218 to effect. The renewed circumnavigation of the columnar Sementes and the displacement take place until all segments the overall card has been processed and the output information has been reassembled to form the overall card.

An existing program can be used in the computer system 10 according to the invention by inserting two cards into the main program be put into operation · The first executable instruction in a module of the system must be "CALL SVS;" be a standard version, which consists of a sub-routine that maintains or maintains the general registers, the data for the Fijcpunkt- and restart operations. To return to the EXEC module the "CALL RELNK" instruction must be executed. This is

209848/0989

-43- 2H9717

a sub-routine that executes the program saved by the checkpoint process as determined by the general registers reloads or enters in the SVSRI subroutine. If the new program module handles a card created by a previous program module operation, i.e. a of the modules shown in Fig. 1, the required global variables must be available, and the card must go through the general card entry routines are entered, which are shown here as COLIN De »If the new module has a Map generated that will be handled by other existing modules of the system, it must be created using the here as a COLOT designated general output routines in the auxiliary memory of the system. New modules can also be created using the RTBLK and WTBLK routines as well as COLIN, COLOT, SVSRI, RELNK a control point baad storage repeat or generate »The global storage and recovery Find routines are a sequence of subroutines that are available in the disk storage of the computer system and from any modular operation that requires them, can be loaded and used. These routines are integral Part of the computing system.

New operations that are entered into the computing system 10 require a control card assignment in which columns 1 to 4 bear the module name, columns 5 to 8 or 9 to 12 refer to the input or output card name, and columns 13 up to 80 can be determined by the user. If control cards are then required, they must also have a module name in columns 1 to 4, while the rest of the card is used for subdivision, depending on the requirements Global variables No ₀ 1 specified unit can be read. It should be noted that this only applies to control cards and that large data sets

209848/0989

-44- 2U9717

after the END card should be arranged so that it is directly can be read by the card reader.

New programs can be downloaded from the card reader, from magnetic tapes etc. loaded or entered »You can be either a temporary or a permanent part of the system, what depends on the respective program. These new programs can do lORTAN translations, SYMBOLIC sentences (assembler programs), Binary map sets etc. included. It goes without saying that the computing system described here is in no way essential for operation needs to be changed with the new processes. New modules can t in the system with any number of combinations of inputs and outputs work from cards. A module can either enter and issue one card, two cards, or no card at all it can be a map, a checkpoint tape record, or no output at all.

The method and computing system described here, which currently works with a SIGMA-7 computer, as already mentioned can perform many desired operations with maps, cross-sections, and other three-dimensional data arrangements in a coordinated manner carry out. Although the system is not currently in practice for another computer system such as type IBM 360 is provided, it can easily be adapted to this or P another type of computer. The computing system can process two-dimensional arrangements of practically any size. Since it is open, new operations can easily be added to the overall program.

Briefly summarized, the invention thus creates a computing system, which is formed by a modular sequence of computer processes that can be accessed in a preselected sequence and deliver a desired calculation result. The system uses an execution module that is foreign to the computer memory (i.e. normally not included) and all data processing of other selected modules which have certain special processing functions have coordinated The system carries out the input

2098A8 / 0989

-45- 2U9717

Processing and displaying many forms of data that can be expressed in three-coordinate values. Ss is particularly suitable for contour processing, for comparison processes and for the mapping of terrestrial Data obtained from various geophysical processes. It is an "open ended" system in which one or more new data processes can be determined and processed under the control of the existing central system. These new processes are added to the system as specially programmed modular components. «The system is practical unlimited in the number of coordinate values that can be processed. Furthermore, by the here described Procedure Removed restrictions on the size of the computer memory.

It goes without saying that various changes and modifications of the exemplary embodiment described are possible within the scope of the invention.

209848/0989

2H9717

Continental Oil ... ^ 8 ^ 7-71 / H _{/ 4}

einoegongen on.

NEW DESCRIPTION PART (to be inserted after page 45)

A typical data processing system 250 in which the program or calculation control process can be carried out according to the invention is shown in Fig. 19 from which the ψ relationship between the programming and the program control is apparent. It is a Sigma 7-8 computer from Xerox Data Systems, but any other conventional data processing system could be used, which has a core memory, a central processing unit to carry out the various linking and arithmetic functions * large, serially accessible external memories such as e.g. contains tape units as well as a random access memory. For example, systems of the JJ6Q / 37Q series from IBM or the 3000 to 6000 series from Control Data Corporation are suitable.

In the initial stages, the mapping computing system uses an external random access memory, such as a stack of magnetic disks 252 or another random access memory 254. Instead of this, a serially accessible storage device such as a magnetic tape unit could also be used, but one would have to in In this case, expect a slightly lower degree of efficiency.

At the next stage of operation, the execution module is taken from disk pack 252 or from store 254 via a multiplexer input / output stage 256 is input directly into the core memory, i.e. into one of the core memory units 258, 250 or, if available, additional core storage units 262 and 264.

information regarding the storage locations in units 258 to 264 and the size of the module to be processed is provided by the

- p—

209848/0989

2H9717

received m £ ^ ^?

Central processing unit 2ό6.

The control of the central processing unit 266 and all other hardware is then carried out by the EXEC module. In the The next phase of processing receives card samples from a card reader 268 or equivalent devices such as magnetic tape units 270a, 270b, 270c, etc., from plate stack 252 or read from memory 254.

The information registered on the cards is transferred directly to the core storage units via the multiplexer stage 256 258 to 264 brought, under the control of the central processing unit 266, the location in the core memory and the amount of to the reading data.

All peripheral devices, i.e. the magnetic tape units 270a to 270c, the disk pack 252, the card reader 268, the random access memory 254 and so on, usually use the interrupt hardware 272 in direct connection with the central unit 266, so that this is signaled when the peripheral devices have completed their functions. This enables the simultaneous operation of several peripheral devices. However, the completion of the data transmission can be achieved by time monitoring or by special puncturing of the hardware parts to be established.

The central unit 266 now runs through the commands specified in the execution module and examines the information coming from the card reader 268. After all the required core memory cells have been brought to the correct value, the EXEC module is stored at an addressed location in the disk stack 252, together with information such as the addresses of the core memory locations for the further execution of the EXEC module from the core memory units 258 to 2t> 4 or the memory 254 is written. The central unit 266 then controls the input of the required processing module from the disk stack 252 or memory 2 ¹ ^ H via the multiplexer stage 2po directly into the

209848/0989

? 2U9717

received on. clioju ^

Core memory, i.e. units 258 to 264. The controller the central unit 2ö6 and the entire hardware is then transferred to the special module, which was previously located in the core storage units 258 to 264.

The four basic module types and their relationship to the different ones Parts of the data processing system will be explained below.

EINGAEEFHASE

Input to the system can be from any of the peripheral devices take place, i.e. from the magnetic tape units 27'O * one Teletype keyboard 274, the disk stack 252, the card reader 268 and memory 254. Selected from one of these Devices are the data via the multiplexer stage 256 in the Transfer core storage units 258 to 264, where they (namely the X, Y data) by linking arithmetic functions of the central unit 266 can be converted into networks. Using disk pack 252 or memory 254, the data is then stored for their use in connection with the central processing unit 266 sorted. The resulting sorted output information is then written into the magnetic tape units 2 '/ Oa to 2YOc, whose Puncture also replaced by stack 252 or storage 254 can be.

PROCESSING PHASE

The processing phase can involve two general types of data. First of all, it is the one in the initial phase generated, now stored in the magnetic tape units 2? 0a to 270c Data. The data from & en magnetic tape units (or the disk stacks) selectively read and entered into the core storage units 258 bis, where the central processing unit 266 through logical and arithmetic Functions, for example through surface calculations from the data, a regular network is constructed. To

209848/0989

-4-

2H9717

received on.

the required arithmetic operations are the data in column format from the core storage units 258 to 264 via the multiplexer stage 256 transferred to memory 254, from where they are selectively can be accessed.

FILTER PHASE (optional)

After the data has been standardized in the column form, spatial filters (two dimensions) can now be generated, which are used for the standardized data. This is done by adding the required data columns for so many lines are read as will fit into core storage units 258-264. The data is transmitted through the multiplexer stage 256 transferred directly to the core memory. The desired spatial filter can then use an associated convolution filter unit 276 to which the data is applied. The unit 276 enables maximum efficiency in the pilter operation. Your puncture can can also be simulated by the central unit 266, but this happens at a much lower speed. The filtered Data is then stored in memory 254 in column format or written to disk pack 252.

OUTPUT FUNCTIONS

Upon output, the data is written from memory 254 or stack 252 to some permanent storage device such as a line printer 278 or the magnetic tape units 270a to 270c. It is also possible to transfer data from core storage units 258 to 264 Transfer intermediate processing to a random access facility, i.e., to disk pack 252 or memory 254. Likewise, can Data can be fed directly to the tape units 270 and / or the line printer 278. All data transfer takes place via the Multiplexer stage 256. Those in the magnetic tape units 270a to 270c Stored data can now be retained or printed or also subsequently using an associated "Calcomp" facility or another commercially available peripheral output device will.

209848/0989

-K-

2U9717

Received on £ i ±.: il. ^

RESTORING THE EXECUTION MODULE

After the selected modular processing function has been completed, the EXEC module must pass through the multiplexer stage 256 from disk stack 232 or memory 254 be re-stored in the core memory. After this transfer, the control of all parts of the system is transferred to the EXEC module passed, whereupon the execution of the program is continued in the specified manner until an end-of-file signal from card reader 268 appears, whereupon control is taken over by the monitor program.

209848/0989

Claims (1)

2U9717 Claims . Method for the automatic processing of three-dimensional input data, with special data processing operations are each carried out by special modular operations, all of which are controlled by a central execution program, characterized in that a processing operation is carried out by entering processing commands is initiated into the execution program, which is then executed this instruction performs special modular processing operations in sequence; that input data in the form of x ~, y and z coordinates are introduced (assembled), which at each processing module are available; that successively the processing of the introduced three-dimensional data by selected ones the special processing module is controlled under the control of the execution module; and that from everyone The processed data resulting from the special processing module operations are output and as a map in three-dimensional form Values are saved. 2. The method according to claim 1, characterized in that the processing modules Definden itself in program sections that come from an input program section, an output program section, a general processing program section and a special processing program section can be selected, and that each of these four program sections from a plurality of individual, different Processing modules consists, each of which is able to carry out its special data processing operation through the execution program to be carried out with the three-dimensional input data. 3ο method according to claim 2, characterized in that that the three-dimensional data map has a predetermined x and y coordinate plane in the form of z-value control points processed with irregular spacing 209848/0989 2H9717 S3 and that the irregularly spaced z-value control points interpolated and extrapolated in such a way that a uniform network of control point values in pontoon z-values is determined, which are evenly over the predetermined x and y coordinates «extend data card © 4 · The method according to claim 3 »characterized in that that a distinction is made as to whether the data of the i, y and z ^ values with regard to suitability are regularly or irregularly spaced for a uniform agitation, and that a network of z-values on x- and y-coordinates are formed if these data are uniform, while the three-dimensional data is preserved if they irregular distances with regard to uniform network dimensions in the x and y planes. 5. The method according to claim 4 »characterized in that that, through further network processing, z-values are irregularly spaced relative to known ones three-dimensional data control points are interpolated and extrapolated and thereby a uniform field of z-values is formed over the determined x- and y-coordinate plane. * ' 6. Program system for the automatic processing of three-dimensional input data in an electronic computer, which consists of a central processing unit which contains an internal core memory formed integrally with it, the required input and output units and an external random access memory, characterized by an input Program sector that contains several different input program modules, each of which can be called up separately to carry out a special input function; a general processing & program sector, which contains a plurality of different processing program modules, each of which can be called up separately to carry out a special processing function; an output program sector comprising a plurality 209848/098 9 contains various output brogram modules, each for Execution of a special output function can be called; a central control program sector with an execution program module for controlling the input, processing and output functions through continuous selection and control of the respective operational program modules in the input program sector, in general Processing program sector as well as in the output program sector; and data input means for receiving three-dimensional Input data and parameters required for processing this data and for effecting a predetermined sequence of modular operations, which are carried out under the control of the execution program module. 7 »Program system according to claim 6, characterized characterized in that the execution program in the internal core memory of the processing unit is foreign and itself is only initially located in this and is pushed out into the outer memory when a program module effective at this time is entered until the function of the effective program module is completed, whereupon the execution program is destroyed the previously stored program module is pushed back into the inner core memory; and that each of the individual Input, processing and output program modules with a callable address is stored in the external memory and under control of the execution program for input into the inner core memory is called. 8. Program system according to claim 7t, characterized in that that the input data is three-dimensional data in the form of x, y and z values; that a network formation program ~ module located in the general processing program sector under the control of the execution program by evaluating the three-dimensional data determines whether the data in the form of the x and y coordinates are regular or irregular are; that the network formation program module is a Hetz-Eins program contains, through which those three-dimensional data are recorded, which regularly and to a 209848/0989 - ** 2H9717 uniform representation in χ and y coordinates in the position are; and that the network formation program module is a network two program contains, with which irregular three-dimensional data are processed in such a way that missing z-values interpolated and extrapolated to form data that is uniform in terms of y and x coordinates. 9 · Program system according to claim 8, characterized in that the output program sector Contains a contouring program module that is under control of the execution program all z-values of the x- and y-coordinate map classified and predetermined values for output as contour display values relative to the z-value data on the x and y coordinate maps selects. tO.Programmsystem according to claim 8, characterized characterized in that an output program module, which can be controlled by the execution program, a map from the uniform z-y values of the x and y coordinates printed out and these card values are stored in the external direct access memory for selective retrieval. 11. Program system according to claim 10, characterized in that the output program sector W contains a contouring program module which, under the control of the executive program, causes all z-values of the x and y coordinate maps to be classified so that predetermined values for output as contour display values relative to the z-value data on the x and y coordinate maps can be selected. 12. Program system according to claim 10, characterized marked that the execution program selectively controls the output program module in such a way that a data card are output in x, y and z values by one or more processing program modules and that data values separately addressable positions for output data cards are stored in the external direct access memory 209848/0989 "2H9717 ■ • 50— 13 · Program system according to. Claim 8, characterized characterized in that the three-dimensional data for regularly spaced data are outside of the inner Core memory of the processing unit are located and in the external direct access are stored in the memory. 209848/0989