DD284318A5 - ARRANGEMENT AND METHOD FOR DATA REDUCTION IN THE DATA COLLECTION OF DIGITIZED MEASUREMENTS - Google Patents

Abstract

The invention relates to an arrangement and a method for data reduction in the data acquisition of digitized Meszwerten that are in a defined relation to each other, so that there is a Y value for each X value (XY operation, Yt operation). The invention can be used anywhere where many analog-to-digital conversions must be stored quickly. By using standard components, a simple and cost-effective construction is possible. Auszerdem can be done pre-processing, so dasz not only the saving of memory locations, but also time savings are achieved in the evaluation. Figure 1 shows the arrangement of the assemblies. Figure {Data reduction; Data collection; digitized mesz values; Analog-to-digital converter; digitization; defined reference; XY mode; Yt mode}

Description

For this 3 pages drawings

Field of application of the invention

The invention relates to an arrangement and a method for data reduction in the data acquisition of digitized measured values, wherein two measuring words have a defined relation to each other. The digitization is preferably done with analog-to-digital converter (ADC).

The application extends to data acquisition / measurement systems where many values need to be stored quickly.

In particular, dia invention allows already during the detection of a simple Meßwertvorverarbeitung, so that reduce the cost of storage and evaluation and the transmission distances to the connected digital computer.

Characteristic of the known state of the art

When acquiring data with ADUs, large amounts of data are often generated, which are usually written into memory with the aid of sequential control systems or microcomputers in order to evaluate them after the measurement, if the time for processing during acquisition is insufficient. This requires large and expensive data storage or even external mass storage devices (with their poor time response). With the simultaneous acquisition of two measured values, the storage space requirements still double. For this reason, some measuring tasks can not be realized with reasonable effort, or large circuit-related expenditures or restrictions have to be made in order to reduce the number of resulting data values (see, for example, methods for adaptive analog-digital conversion, DD-WP 25 ^ 154). That two input variables have a defined relationship to one another is taken into account in known solutions only for the Y-t-Botrieb, in which for each time point (X value) only a data value (Y value) is formed. Otherwise, each input variable is treated and stored separately and the mutual reference is only considered during processing (see, for example, Reinhart, F., Elektronische Meßwertverarbeitung, Hüthig-Verlag 1977, pp. 197-200, Sheingold, DH, Analog-Digital-Conversion Notes. Norwood: Analog Device ', 1977). A convenient form of data storage for processing is thus not possible. A circuit-technical and therefore fast preprocessing during the data acquisition is not taken into account or only in the form of characteristic curve corrections.

For a direct storage of all data values would have to be provided according to known technical solutions, a large and because of the time requirements fast memory, which would be associated with considerable effort and high costs.

Object of the invention

It is the object of the invention to prepare digitized measured values whose time sequence is such that they can not be taken over directly by a digital computer in such a way that they can be further processed by the connected digital computer. The effort for the necessary caching is to be kept as low as possible.

-2- 284 318 Explanation of the nature of the invention

The invention has for its object to achieve by a switching and procedural solution, a data reduction in the data acquisition of digitized measurements, so that an evaluation in the connected digital computer is possible or simplified.

According to the invention, the object is achieved by switching a digitized measured value (Y value) directly or via a selection logic to the data input and the associated other digitized measured value (X value) to the address input of a read-write memory with random access. As a result, one of the two measured values is stored in the memory and the second used as the address of the memory location for the first. In the realization of the object, the property of the digitized measured values is utilized in that they have a defined relation to one another, that is to say that for each abscissa value there is only one ordinate value which is necessary for the exact evaluation. The exercises; It is not necessary to save and save the X values because the offset from the start of the memory represents the X value. Since the memory can be read out in an ordered sequence, the XY value pairs are in X-ordered form, which is favorable for processing in the digital computer. The digitization of the measured values is preferably carried out with analog-to-digital converters (ADCs).

Figure 1 shows the arrangement of the assemblies. A sampling cycle has the following structure, wherein all control and clock signals are generated by the sequence control 3:

First, the analog values Ue 1 and Ue2 are sampled with the ADUs ADU 1 and ADU 2. The conversion method has influence on the conversion time, the effort and the accuracy.

When the data is available at the ADU2, the selection of the corresponding memory location in the memory 4, which is designed as a random access memory (RAM). The address data arrive via an address multiplexer 5 to the memory 4. The multiplexer 5 is necessary in order to switch the addresses from the address bus of the connected digital computer to the memory when reading.

When the address is applied to the memory 4, after the access time of the memory 4, the data appearing at its output written from a previous sampling cycle or from the initialization appears. This data can optionally be linked in a selection logic or arithmetic logic unit (ALU) 6 with the data from the ADU1 in a predetermined manner. Then the writing of the result of the ALU 6 takes place on the selected memory location of the memory 4.

The structure of the ALU 6 and thus the combination of the two data values depends on the respective measurement problem. For example, it can be done so that the larger value always comes into the memory. This is a maximum value storage realized. Another simple example is the averaging which is realized by adding the two data values in binary full adders and connecting the outputs of the full adder one bit to the right of the data inputs of the memory 4 so that a division by 2 is achieved , Thus, averaging (a + b) / z is possible without additional expenditure of time. If it is guaranteed that only one data value is written to each address, or if only the respectively last value is relevant, the ALU 6 can be dispensed with.

When using a buffer between memory output from the memory 4 and ALU6, a link to the data values of any other memory cell is possible. In this case, if necessary, a second read access with likewise cached address is necessary.

The combination of different selection criteria is possible.

After completion of the measurement, which may consist of any number of samples, the data from the connected digital computer 8 via the bus buffer 7 are read from the memory.

After initializing the memory, a new measuring cycle can be started. By separating the data acquisition from the Digi'alrechner 8, the calculator during the measurement other tasks, for example, the evaluation of the previous measurement, e ^ fill.

By the invention it is possible to write digitized measurement values quickly in a memory, de ^r is coupled to a digital computer.

The inventive arrangement according to Figure 1 is a fast preprocessing possible, so that simplifies the evaluation in the digital computer. By taking into account the mutual reference of two Meß.verten a memory space saving and a favorable for processing form of storage is possible.

The invention can be realized in most cases with standard components.

The invention will be explained in more detail by exemplary embodiments.

embodiments

Example 1:

The described method is used in the evaluation of the data of a sector field mass spectrometer in conjunction with a digital computer.

The spectrometer provides two analog values (X, Y), where X represents the magnetic field strength corresponding to the mass value and Yden represents the ion current selected by the separation system at this magnetic field and thus an intensity information. This results from the measurement method and the resolution of the instrument an Intunsitätsverteilung according to Figure 2.

The magnetic field is changed with an analog control circuit and measured with a magnetic field sensor. As a result, the mass value changes non-linearly in time over the entire measuring time. The resolution of the measuring device does not ideally decrease towards higher masses, as shown in Fig. 3.

In AD conversion, the task now is to determine the associated maximum intensity value for each integer mass value. However, since atomic and molecular masses are not even, a window of ± 0.5 mass units should be searched for the intensity maximum. It must be so often measured for reliable detection of the maximum that it is determined at all points with sufficient accuracy. Since the peaks in the lower part of the mass range are only a fraction of a mass wide, about 200 samples per mass must be made for a 12-bit resolution. This results in a mass range of 1 to 1000 to 20,000 samples. An average of about 10ms is available for sampling a mass.

The arrangement is realized with standard low-power TTL circuits for the sequential control and multiplexer. The sequencer generates the start, stop, read and write clocks for all modules in the correct sequence as shown in Figure 4. AIsADUs are 2ICe AD574 with a preset sample and hold circuit. Only 1ObIt are used for the Y value. The ADCs work according to the method of successive approximation with a conversion time of max. 40ms (including S / H circuitry) at a resolution of 12 bit.

As memory three 1K x 4bit static RAMs of type U 214 are used. The ALU is built with three cascaded 4-bit CMOA size comparators V4585. For the memory, a write pulse is only released by the comparator if the value coming from the ADU is greater than the value read from the memory. As a result, a maximum value storage is realized.

After the end of a measurement, which is signaled by the spectrometer, the data block comprising the 1000 values, with a clear assignment between mass number and intensity, is read from the memory and processed further by the connected digital computer.

Claims (6)

1. Arrangement and method for data reduction in the data acquisition of digitized measured values, which are in a defined relationship, characterized in that a digitized measured value (Y value) directly or via a selection logic on the data input and the associated other digitized measured value ( X value) switches to the address input of a read-write memory with random access, whereby one of the two measured values is stored in the memory and the second is used as the address of the storage location for the first measured value. 2. Arrangement and method according to claim 1, characterized in that by the arrangement of a selection logic before the data input of the memory from the digitized measured values to be written in the memory, and the data that are already in memory, a selection in the Type is met that predetermined conditions met and only the selected data are stored. 3. The method according to claim 2, characterized in that only the maximum value is stored from belonging to a memory data values. 4. The method according to claim 2, characterized in that only the minimum value is stored from belonging to a memory data values. 5. The method according to claim 2, characterized in that formed from the belonging to a memory data values of the average and only this is stored. 6. Arrangement and method according to claim 3,4 or 5, characterized in that one enables a connection with the data value of any other memory cell by d'e arrangement of a buffer between memory output and selection logic.