|
Claims  |
|
|
We claim:
1. A data display system comprising:
data receiving means for receiving a first series of data samples from a
first samples source, said first series of data samples having associated
therewith a hypothetical first high-fidelity plot containing every data
sample in the first series of data samples, the hypothetical first
high-fidelity plot having a waveshape and other visually recognizable
attributes associated therewith;
dividing means for dividing the first series of data samples into groups of
data samples, each group having two or more samples;
representation means for providing for each group of data samples, a
representative value representing the data samples of the group; and
plotting means for plotting the representative value of each group;
wherein the dividing means and representation means function so as to
enable substantial visual recognition of the waveshape associated with the
first hypothetical high-fidelity plot from the plot of the representative
values.
2. The data display system of claim 1 wherein the dividing means includes a
1-of-every-n sampling means for selecting a representative one data sample
from each group of n data samples, n being an integer greater than one,
and wherein the representation means provides the representative one data
sample of each group of n data samples to the plotting means.
3. The data display system of claim 1
wherein the dividing means includes means for dividing the first series of
data samples into groups each having a first number of bits and the
representation means includes data reducing means for generating the
representative value in a small-sized form having a second number of bits
less than the first number of bits;
wherein the plotting means operates at a throughput rate which is less than
a source rate at which the first samples source supplies said first series
of data samples;
and further wherein the small-sized form of the representative value is
adjusted so as to enable the plotting means to keep pace with the source
rate at which the first samples source supplies said first series of data
samples.
4. The display system of claim 3 wherein the representative value is equal
to one of an average, maximum or minimum value of the data samples in its
corresponding group.
5. The display system of claim 1 further comprising:
analog to digital (A/D) converting means for producing the first series of
data samples, said A/D converting means having a user-definable gain; and
guideline means, operatively coupled to the plotting means and responsive
to the user-definable gain, for drawing a calibration guideline on the
plot of the representative values of each group.
6. The display system of claim 1 wherein:
the plotting means is adapted to transmit a plot of said representative
values to a preselected plot display means having pixels organized as a
preselected number of display columns and display rows;
the plotting means includes data transforming means for transforming
representative values provided by the representation means into row and
column coordinated plot data which is compatible with the preselected plot
display means; and
the data transforming means operates at a throughput rate which is less
than a source rate at which the first samples source supplies said first
series of data samples.
7. The system of claim 6 wherein:
the display means includes means for displaying information in different
colors;
the data receiving means is adapted for receiving a second series of data
samples from a second samples source, said second series of data samples
having associated therewith a hypothetical second high-fidelity plot
containing every data sample in the second series of data samples, the
hypothetical second high-fidelity plot having a waveshape and other
visually recognizable attributes associated therewith;
the dividing means is adapted for dividing the second series of data
samples into second groups of data samples, each second group having two
or more samples;
the representation means is adapted for providing for each second group of
data samples, a representative value representing the data samples of the
second group, the representative values of the first and second groups
respectively forming first and second representative plots for the first
and second series of data samples;
the plotting means includes color assigning means for assigning an
individual color to each of the representative plots of the first and
second series of data samples; and
the means for displaying information displays the representative plots in
the colors assigned to the representative plots.
8. The system of claim 6 wherein the display means includes a hardcopy
printer.
9. The data display system of claim 1 wherein the number of samples per
representative value is 100 or less.
10. The data display system of claim 1 wherein the number of samples per
representative value is 50 or less.
11. The data display system of claim 1 further comprising
display ratio adjusting means for allowing a viewer of a plot produced by
the plotting means to adjust the number of samples per representative
value while viewing the plot.
12. The data display system of claim 1 wherein the data receiving means
includes:
buffer means for temporarily storing the received data samples, said buffer
means being operatively coupled to the plotting means for transferring
data to be plotted to the plotting means by way of the dividing means and
the representation means;
mass storage means for more permanently storing the received data samples;
and
control means, operatively coupled to the buffer means and the mass storage
means, for controlling transfer of data between the buffer means and the
mass storage means, said control means also controlling transfer of data
from the buffer means to the plotting means by way of the dividing means
and the representation means;
and wherein the data display system further comprises:
automatic display ratio adjusting means for automatically adjusting the
number of samples per representative value so that the controlling by the
control means of data transfer between the buffer means and the mass
storage means is not hampered by the controlling by the control means of
data transfer from the buffer means to the plotting means by way of the
dividing means and the representation means.
13. The data display system of claim 1 wherein the dividing means divides
the first series of data samples into groups of data samples, each group
having a same number of samples.
14. The data display system of claim 1 wherein the data receiving means is
adapted to receive the first series of data samples at a first rate and
the plotting means is adapted for plotting values at a substantially
slower, second rate.
15. The data display system of claim 1 wherein the first series of data
samples are produced in real time and at a first sampling rate from
sampled measurements of physical quantities and the plotting means is
adapted for plotting values at a second rate that is substantially slower
than the first sampling rate.
16. The data display system of claim 1 wherein
the data receiving means is adapted for further receiving second through
Nth series of data samples from second through Nth samples sources, said
second through Nth series of data samples having associated therewith a
hypothetical second through Nth high-fidelity plots each respectively
containing every data sample in the respective second through Nth series
of data samples, the hypothetical second through Nth high-fidelity plots
having respective waveshapes and other visually recognizable
characteristics associated therewith;
the dividing means is adapted for dividing the second through Nth series of
data samples into respective second through Nth groups of data samples,
each of the second through Nth groups having two or more samples;
the representation means is adapted for providing for each of the second
through Nth groups of data samples, a representative value representing
the data samples of the respective second through Nth group, the
representative--values of the first through Nth groups respectively
forming first through Nth representative plots for the first through Nth
series of data samples; and
the plotting means is adapted for further plotting the representative
values of each of the second through Nth groups;
wherein the dividing means and representation means function so as to
enable substantial visual recognition of the waveshapes associated with
the first through Nth hypothetical high-fidelity plots from the plots of
the first through Nth representative values.
17. The data display system of claim 1 wherein:
the data receiving means has a throughput rate of at least 12000 samples
per second per sample source; and
the number of samples per representative value is adjusted to allow the
plotting means to plot the representative values substantially in pace
with said throughput rate of the data receiving means.
18. The data display system of claim 1 wherein the plotting means plots the
representative value of each group on a plot surface of limited area, the
plot surface having first and second plotting axes of limited extent, said
data display system further comprising:
display gain control means for adjusting a gain applied in plotting the
representative values against at least one axis of the plot surface so
that, with appropriate adjustment of the gain, the plot of the
representative values can be spread across the full extent of the at least
one axis.
19. A computer system comprising:
CALIBRATE means for generating a calibration file having first information
defining units of measurement of a physical quantity for association with
raw data numbers that are to be stored in a raw data file and that are to
represent samples of a measured physical quantity, and having second
information defining calibration equations for converting the raw data
numbers into calibrated physical quantity measurement values;
COLLECT means for collecting raw data produced by a raw data producing
means and displaying a plot representative of such raw data;
REPLAY means for reproducing the raw data collected by the COLLECT means;
CHART means for producing a hard copy plot of the collected data, said hard
copy including one or more guidelines indicating preselected levels of
physical quantity measure conforming with the calibration equations of
said CALIBRATE means.
20. A plot generating system for generating one or more real-time plots
showing characteristic waveforms of incoming digital data points sampled
from physical signals, said plot generation occurring during substantially
simultaneous transfer of the incoming data points to, or from, permanent
storage, where the incoming digital data points can be represented by
hypothetical high-fidelity plots containing every one of the incoming
digital data points, said system comprising:
representative value generating means for generating for each plurality of
incoming data points, a single representative point providing
representative information about the plurality of points, each
representative value consuming fewer bits than the digital data points it
represents; and
plotting means for plotting representative values generated by the
generating means;
wherein the combination of the representative value generating means and
the plotting means function to enable substantial visual recognition of
the waveshape associated with the hypothetical high-fidelity plots from
the plots of the representative values.
21. The plot generating system of claim 20 wherein the plotting means
includes a printer having a fixed sheet feeding speed, and the number of
incoming data points represented by each representative point is varied to
simulate a variable sheet-feeding speed.
22. The plot generating system of claim 21 wherein the plotting means is a
laser printer.
23. A method for compressing points plotted on a graph having first and
second coordinate axes, the method comprising the steps of:
generating a first number N of data points each having associated therewith
a unique one of a corresponding number N of unique first axis points and a
predetermined one of a corresponding number N* of second axis points where
N* is equal to or less than N;
dividing the set of N unique first axis points to define a smaller number
of N/M first axis subsets where each first axis subset contains a
plurality of first axis points and where M is a number greater than unity;
assigning a unique identifier point from a succession of identifier points
to each of the N/M first axis subsets;
dividing the set of N data points into N/M data point subsets each
containing a plurality of data points;
linking each of the N/M data point subsets with the corresponding unique
identifier point which identifies the subset of first axis points with
which the data points of the data point subset are associated; and
plotting the data points of each data point subset over the corresponding
unique identifier point which was linked to that data point subset,
thereby plotting the N data points over a number of N/M identifier points,
the number of said identifier points being less than N.
24. The method of claim 23 wherein the first coordinate axis is a time
line.
25. The method of claim 24 wherein the second coordinate axis is a voltage
line.
26. The method of claim 24 wherein the data points are sample points
collected in real time from measurements of a time varying physical
quantity.
27. The method of claim 23 wherein M is equal to or greater than 2 and each
data point subset contains M data points.
28. The method of claim 23 wherein the N data points belong to a series of
successively generated data points, the method further comprising the step
of skipping over a predetermined fraction of the successively generated
data points and not including these skipped over data points in the data
point subsets.
29. A method for compressing the display of a large number N of data
points, said N data points having associated therewith a hypothetical
high-fidelity plot containing every data point of the N data points, the
method comprising:
dividing the set of N data points into N/M subsets, each subset containing
a plurality of data points, M being a value greater than unity;
for each subset, generating a representative data point whose value is
representative of the data points in the subset; and
plotting only the representative data points rather than all N data points;
wherein the dividing and representation steps enable substantial visual
recognition of the waveshape associated with the first hypothetical
high-fidelity plot from the plot of the representative values; and
wherein said N data points represent real time measurements of a
time-varying physical quantity.
30. The method of claim 29 including the step of selecting N to be a number
greater than 999.
31. The method of claim 30 including the step of selecting M to be a number
equal to or greater than 10.
32. The method of claim 29 wherein the generating step comprises selecting
one of the data points in the subset as representative of the other data
points in the subset.
33. The method of claim 29 wherein the generating step comprises averaging
over the plural data points of the subset to generate the representative
data point.
34. A method for compressing the display of a large number N of data
points, said N data points having associated therewith a hypothetical
high-fidelity plot containing every data point of the N data points, the
method comprising:
dividing the set of N data points into N/M subsets, each subset containing
a. plurality of data points, M being a value greater than unity;
for each subset, generating a representative data point whose value is
representative of the data points in the subset;
plotting only the representative data points rather than all N data points;
wherein the dividing and representation steps enable substantial visual
recognition of the waveshape associated with the first hypothetical
high-fidelity plot from the plot of the representative values; and
varying the value of M during said plotting to thereby provide a desired
degree of fidelity in the plot of the representative values. |
|
|
|
|
Claims  |
|
|
Description  |
|
|
BACKGROUND OF THE INVENTION
1. Cross Reference To Microfiche Appendix
This application includes a computer program listing in the form of
Microfiche Appendix A which is being filed concurrently herewith as 655
frames (not counting target and title frames) distributed over 7 sheets of
microfiche in accordance with 37 C.F.R. .sctn.1.96. The disclosed computer
program listing is incorporated into this specification by reference but
it should be noted that the source code and/or the resultant object code
are subject to copyright protection. The copyright owner has no objection
to the facsimile reproduction by anyone of the patent document (or the
patent disclosure as it appears in the files or records of the U.S. Patent
and Trademark Office) for the purpose of studying the disclosure but
otherwise reserves all other rights to the disclosed computer program
including the right to reproduce said computer program in
machine-executable form.
2. Field of the Invention
This invention relates generally to a data acquisition system having a
capability to display in plotted from one or more continuous streams of
digital data samples and more specifically to a display system which uses
a combination of standard computer graphic display adaptors and graphic
monitors together with specialized software for the real-time production
of sample point plots.
3. Description of the Relevant Art
The need to visualize or display in real time a graphic image of continuous
streams of data emanating from one or more measurement instruments during
data acquisition is well known. By way of example, medical experiments or
tests tend to produce voluminous amounts (i.e., more than 1000 points or
more than 1,000,000 points) of result data and it is desirable to know
from the outset of data collection whether the produced results are within
a generally expected range. It is also preferable to know immediately when
unusual results are being produced so they can be immediately noted,
corrected and/or investigated if so desired.
The preferred display mode is a real-time graphic one. That is, waveshape
recognition or recognition of other graphic information relating to the
collected data should occur simultaneously or almost simultaneously with
the production of data. Preferably, the display image should be of a form
that can be permanently retained for re-production and study at later
times.
Traditionally, real-time graphic display and permanent storage of result
information have been achieved in unison through the use of a conventional
paper chart recorder. A voltage signal or other signal of interest is
amplified as necessary and applied to move a recording pen across the
surface of a continuously streaming paper chart (or "strip chart"). A
waveshape of the received information is thereby continuously inscribed
onto a paper record and supplied cut of the recorder for immediate
viewing. The paper trace provides a nearly real-time visual display of the
time versus voltage characteristics of the signal or of other signal
characteristics that are to be studied, and the paper record stores the
image for later re-examination.
Because traditional chart recorders employ ink pens or other mechanical
parts to write the signal image to paper, they cannot faithfully display,
in real time, rapidly changing signals such as for example signals having
a frequency component of 200 Hz or more.
Recently, paper recording devices have been devised which digitize incoming
analog voltage signals and inscribe the digitized form to paper using a
rapidly responding printing means such as a thermal writing micro-array or
by using electro-static or photographic methods of recording on paper.
Examples of such devices include the Gould Electronics Model TA2000 or ES
2000 Electrostatic Recording stem, by Gould Inc. Recording Systems, 3631
Perkins Avenue, Cleveland, Ohio, 44114; and the Model ME-9500 8-channel
recorder by Astromed Inc., West Warwick, R.I., U.S.A.
These newer recorders can reliably display in real-time signals having a
maximum frequency of approximately 5 to 8 kHz. The new digitizing
recorders are provided with features such as adjustable "paper speeds",
i.e. the rate at which signals are inscribed and paper is rolled out is
adjustable, so as to allow for either compression of long durations of
slow-changing signals into a single short length of paper, or
alternatively by using rapid paper speed, to allow brief events to be
expanded over long segments of rapidly streaming paper for high event
resolution.
But despite such features, the newer recorders still have disadvantages.
Aside from having a restriction on the maximum signal frequency which can
be reliably displayed, (no higher than about 8 kHz), these newer forms of
digitizing recorders suffer from the fact that both the real-time nature
of the display and the storage of the display depends on paper or some
other write-once recording medium which has the disadvantages of being
non-reusable, expensive to purchase, inconvenient to review and bulky to
store.
Recently, attempts have been made to use the data acquisition and storage
capabilities of so-called personal computers (low-cost microcomputers) to
collect, and store for later study, sample values of analog signals. A
typical IBM PC or compatible microcomputer is used in combination with a
suitable magnetic medium type of storage device (e.g. a "hard" disk), an
analog to digital signal adaptor board, and a video display driver (e.g.
color graphics adaptor (CGA), enhanced graphics adaptor (EGA), or virtual
graphics array (VGA)). The analog to digital adaptors usually come in the
form of a single printed circuit board, which includes an analog to
digital converter (A/D) system. The adaptor board is inserted into a
standard computer bus slot in the personal microcomputer to provide the
necessary hardware components for digitizing analog signals. Single board
A/D adaptor systems are manufactured and marketed, for example, by
companies such as Data Translation, Inc., 100 Locke Drive, Marlborough,
Mass., 01752, Scientific Solutions, 6225 Cochran Road, Solon, Ohio, 44139,
and National Instruments, 12109 Technology Boulevard, Austin, Tex.,
78727-6204.
When driven by suitable software, these combinations of a personal computer
plus one or more A/D adaptor boards have the capability of receiving
continuous analog voltage signals on plural A/D channels, digitizing the
voltage signals to generate a representative stream of sequential sample
point values, passing the digital sample point values to a random access
memory (RAM) buffer section of the computer and from there to permanent
storage in the files of a magnetic disk.
Some A/D adaptor boards are provided with relatively high performance
components to allow direct memory access (DMA) transfer of binary data
from RAM to disk so that signal digitization and transfer can take place
at the peak rate of the A/D converter system and thus faithful recording
of signals having relatively high frequency content is made possible.
Currently, there are several software programs on the market which provide
total throughput rates (sampling, digital data transfer to RAM, and
permanent hard disk storage) which roughly match the maximum rate of
digitization of the A/D converter system taken alone. For example, using
an ASYST.TM. software package (McMillan Software Company, 866 Third
Avenue, New York, N.Y., 10022), together with Data Translation's "high
speed" A/D board Model No. 2821F, one can continuously sample and store to
permanent disk storage at rates as high as approximately 130,000
samples/second (thereby allowing for faithful reproduction of signals
having a maximum frequency of 65 KHz or less). But this high speed
performance applies only for retention of signal information and not to
systems where the sampled information is to be simultaneously displayed as
a graphic plot during data collection.
Display systems tend to have a bit rate limitation, independent of the A/D
system bit rate limitation, which prevents sample points from being
displayed in plotted form at rates higher than a fixed number of sample
points per second. Accordingly, even as A/D digitizing rates rise and
maximum computer to storage data transfer rates increase, the display rate
limitation acts as a bottleneck which prevents real-time display on a
point-by-point basis of the incoming signal samples. Heretofore, the
operation of sampling and storage had to be conducted in the "blind" if
one wished to avoid the display rate limitation. But this mode of blind
collection is undesirable in that it deprives the experimenter of valuable
information. Blind data collection leaves the user without the option to
take immediate corrective action should something go awry.
When currently available A/D boards sample incoming voltage signals at
standard, moderate to high sampling rates, the number of digital data
points generated by the A/D system can easily exceed the ability of the
host computer's video display driver and/or monitor to immediately display
the entirety of the continuous stream of data points onto screen. The
computer display subsystem cannot keep pace with the A/D subsystem, and
therefore cannot display in real-time the waveshapes of signals being
received by the A/D subsystem.
In existing software packages, when a voltage signal is digitized by the
A/D board, each successively digitized sample point value is converted by
the computer into X and Y screen coordinates, as it is received, and it is
then processed sequentially via the video display driver and plotted to
the screen. Since a finite time interval is required for mapping into
screen coordinates and writing each successive point into a video buffer,
a display rate limit is imposed by both the computer hardware and the
application software upon the rate of continuous data sampling that can be
visualized. That is, there is a limit on the maximum signal frequency that
can be adequately displayed in real-time, while sample data is being
produced. Additionally, when real time storage is required, time has to be
allotted in each processing cycle for the computer to store the incoming
data samples in appropriate locations on the magnetic disk. This storage
time further limits the maximum signal frequency that can be displayed in
real-time.
Ideally, users would like to view the waveshapes of incoming signals, even
those having relatively high frequency, and to simultaneously record the
waveshapes permanently for later review by suitable analysis means (i.e.
software types of signal analysis tools) utilizing low-cost computers, A/D
board and software. But heretofore, a low-cost combination of hardware and
software has not been available for realizing such a goal.
A computer program sold under the name Lab Tech Notebook.TM. by Laboratory
Technologies Corporation, 400 Research Drive, Wilmington, Mass., 01887, is
representative of the peak performance available from prior art data
acquisition programs. The maximum rate of continuous sampling and
permanent storage of incoming data that is possible while simultaneously
providing a real-time display is limited in prior art systems to a maximum
total rate of sampling not greater than approximately one thousand samples
per second. When a typical number of (i.e., 8) separate signals are to be
processed in a time multiplexed manner on a corresponding number of (i.e.,
8) channels, the maximum rate at which each individual signal can be
sampled, visualized, and stored is divided down to a restrictively low
rate of approximately 100-200 samples per second (SPS). Thus, previously
existing programs cannot successfully display in real-time and
simultaneously store incoming signals, except at very slow rates of
sampling (i.e., 100-200 SPS per channel) that are only a fraction of the
the maximum sampling rates (i.e., 16,000 SPS per channel) available with
many popular A/D boards. At higher rates of sampling, either the display
points have to be visualized on a non real-time basis, i.e., they lag in
time behind the newest burst of incoming data points, or the display
capability is simply not available and data collection takes place in the
"blind".
Besides this inability to display data in real-time, previously existing
data acquisition software packages have not been able to quickly provide
any permanent printed record equivalent to the output of the traditional
paper recorder. Thus, if a disk "crash" occurs well after the time of an
in-the-blind collection session but before there is time to make a
hard-copy print-out of the data collected, the entirety of the results
from the experiment may be lost without giving the experimenter a chance
to visualize or review even gross aspects of the results. The fruits of
all efforts put into conducting the experiment are completely destroyed.
With respect to visualizing the experimental results, post-collection
"playback" or "replay" of stored digital data could satisfy the
requirement for a reviewable, permanent record of the experimental
results, but this does not allow for the same kind of immediate feedback
and optional reaction made possible by displaying incoming data in
real-time. While a subsequently printed hard-copy plot of the results
might be provided as a reviewable record after the facts, the production
of a visually pleasing and/or useful record usually requires some degree
of a priori knowledge. Certain functional features found in the
conventional chart recorders, such as the ability to select different
paper speeds "on the fly" (as the experiment is proceeding) turn out to be
extremely useful. Most digital printers feed their paper at a fixed speed
making such "on the fly" adjustments to paper speed impossible. If an
eye-pleasing plot is to be produced, it is advantageous to give the
experimenter the ability to adjust the characteristics of the data
acquisition system in real time (at the time the experiment is being
carried out) rather than waiting till later and relying only on
post-experiment data-editing processes so that a visually pleasing and
thereby more useful record is immediately produced. It is undesirable to
wait until after the experiment has been completed and to realize only
then that perhaps changes should have been made to the data collecting
process, i.e., some segments of a hard-copy print-out whose specific
locations are now forgotten should have been compressed by using slow
paper speed and others should have been expanded by using a faster paper
speed so as to emphasize key portions of the result data.
Even before an experimental run begins, it is useful to have knowledge of
the magnitude range and frequency range of the initial waveshapes
generated during the set-up of the experiment. The experimenter should be
able to dynamically use this knowledge to adjust the recording
instrumentation during set-up, before the experiment begins, in order to
produce a visually useful and more pleasing record. Ideally, the
experimenter should be able to stretch the amplitude of each waveform plot
so that it occupies the full vertical span of the display screen and thus
provides maximum magnitude resolution on screen if so desired. And the
experimenter should also be able to alter the horizontal spacing between
plot points on screen in real-time so as to optimize visual appreciation
of time related characteristics (i.e., slope or frequency) of one or more
plotted waveforms. It is difficult for a human being to produce an
acceptable plot (video or hardcopy) having such effects after a "blind"
digital collection session because the only information available prior to
beginning such a plot is a digital data file which is stored on a computer
hard disk in the form of a series binary-coded values. Direct print-out of
such a file provides only a list of numbers in text form, not a graphic
plot whose attributes are adjustable on the fly.
The best of the previously available combinations of personal computers,
single board A/D systems, printers and data acquisition software fail to
fulfill the need for a low-cost data acquisition system which provides
continuous real-time display of visually pleasing plots and reviewable
storage of high frequency input signals. Even the capabilities of the
earlier paper chart recorders are not matched by these personal computer
forms of data acquisition systems.
Attempts have been made to circumvent the inability of the low-cost
personal computer systems (video graphics adaptors, monitors and software)
to successfully "keep up with" and plot each successive incoming data
point as it is digitized by the A/D board. But these attempts have
directed themselves to the use of high cost and very specialized computer
hardware. Dedicated hardware has been designed by some to supplement or
accelerate the capability of the video adaptor which drives the
microcomputer monitor and thereby enable a limited form of real-time data
acquisition and display. This hardware attempt does not address, however,
a limitation which arises primarily from the software control program,
namely that it takes the microprocessor a finite amount of time to map
each raw binary number representing a signal level into a memory address
associated with specific screen coordinates and to transfer such mapped
(transformed) data to a video buffer. The software limitation remains
despite hardware augmentation to the system video driver board.
The above mentioned use of specialized computer hardware has been
popularized by the CODAS system available from DATA Instruments Inc., 825
Sweitzer Avenue, Akron, Ohio, 44398-6140. A proprietary and dedicated
videoboard ("scrolling" board) is inserted into a predetermined slot in
the microcomputer to increase the real-time display rate of the computer
monitor in a specific "scrolling" format. That is, the addition of the
CODAS, "scrolling" board allows the microcomputer monitor to more rapidly
write to screen each successive digital data point as it is generated by
exploiting additional hardware plus software to continuously and
automatically "scroll" the display, i.e., to continuously roll all of the
visual information on the screen with the latest points appearing at one
edge of the screen while the oldest points pass off and disappear at the
opposite side of the screen. When this CODAS board plus accompanying
software is used with standard personal computer systems (A/D boards and
monitors), a modest increase in real-time display and storage of digital
data samples is realized, typically to a maximum of 4,000 samples per
second. This means that when the more powerful of standard microcomputers
are used (i.e. an Intel 80386 system) in combination with a high
throughput rate (50-130 kHz) A/D board system, the maximum rate of display
sampling and storage available for a typical set of 8 incoming voltage
signals would be a maximum of 500 samples per second per channel. Such
performance still compares unfavorably against the maximum display and
storage throughput of the older chart recorders, and this display solution
does not yet provide a permanent reviewable log since the waveshapes are
shown only momentarily on a scrolling video display.
The attempts to substitute or add specialized hardware to work around the
failings of standard computer systems (A/D board, and software) has
recently been carried a step further in another type of CODAS system by
the same DATA Instruments company. Additional, more specialized, hardware
has been created in the form of an advanced "scrolling board" which
removes the standard A/D card from the computer system and replaces it
with a proprietary combination board which includes both a videographic
scrolling capability and analog/digital conversion circuitry. When a
computer is outfitted with this special "scrolling" board, a maximum of
50,000 digital data samples per second (approximately 6,000 SPS per
channel in an 8 channel system) can be passed to hard disk storage while
continually being displayed in a special "scrolling" display format.
Thus, by supplanting the commercially standard computer video adaptors
previously used, abandoning the standard A/D board design, and turning to
a custom scrolling screen display format requiring a special item of
hardware, the newer CODAS system is able to provide a moderate rate of
real-time display and store the displayed points for later review on
screen.
While this display ability does finally rival the ability of the
traditional paper recorder to display moderate frequency signals of
several kHz, it is realized with the penalty of severe restrictions. A
first penalty arises in that the real-time scrolling display does not
allow for a full vertical resolution across the video screen as provided
across the paper width of contemporary electrostatic paper recorders
because the size of the scroll buffer has to be limited. And the only
record made available for later review by the "scrolling" board is the
same scrolling display created during real-time acquisition. Another
penalty comes from the fact that the standard video graphics adaptors have
been supplanted by proprietary single-vendor hardware, and accordingly,
many common A/D boards which could offer greater sampling performance
and/or lower cost are not compatible. Thus, it can be seen that such
hardware augmentation only circumvents but does not overcome an inherent
limitation of the data acquisition process, namely, the time it takes a
CPU to map into display coordinates and transfer to a screen or printer
the raw sample point data produced by the A/D board.
Data acquisition software needs to be designed which can overcome the
mapping limitation and successfully generate a real-time display of
incoming data at the highest rates of sampling of contemporary A/D boards,
or accommodate even higher rates of future planned A/D boards. The
real-time display of incoming data should appear relatively continuous
despite simultaneous transfer of bursts of incoming data to permanent
storage (i.e., hard disk) and despite improvements to the rate at which
the hard disk or other medium can physically accept the flow of incoming
data. That is, the displayed plots should not lag in time by any
appreciable amount behind the latest salvo of result data that is being
produced by the measurement equipment. The maximum sampling rate and
permanent storage rate of incoming data points should be limited only by
the speed of the A/D subsystem and the ability of the storage medium to
accept the flow of data, not by the limitations of the display system. All
this should preferably be achieved through software modification rather
than hardware addition so that the necessity or cost of adding specialized
hardware to enhance performance or supplant any usual component of a
standard microcomputer-A/D system is avoided. And the successful mimicking
of the ability of traditional chart recorders to perform real-time display
and storage of continuous voltage signals will not be truly complete
unless the data acquisition software also provides a mechanism for quickly
producing a high resolution hard copy record of the collected data as
would be done by the traditional paper chart recorder.
SUMMARY OF THE INVENTION
It is an object of the present invention to overcome the above described
drawbacks of the prior art.
In accordance with a first aspect of the invention, a first sampled data
stream which is produced by a first data sampling means of a data
acquisition system is again sampled by a second data sampling means at an
adjustable rate to thereby produce a second sampled data stream having a
sample point flow rate (number of samples transferred per second) less
than or equal to the sample point flow rate of the first sampled data
stream. The second sampled data stream is preferably coupled at the very
time of its production to a plot display means such as a computer-driven
video monitor or a printer for immediate display of its | | |