Data processing: measuring – calibrating – or testing – Calibration or correction system – Linearization of measurement
Reexamination Certificate
2002-02-26
2004-08-03
Hoff, Marc S. (Department: 2863)
Data processing: measuring, calibrating, or testing
Calibration or correction system
Linearization of measurement
C324S601000
Reexamination Certificate
active
06772079
ABSTRACT:
FIELD OF THE INVENTION
The invention relates to signal acquisition systems and, more particularly, to the compensation of errors that arise in signal acquisition systems.
BACKGROUND OF THE INVENTION
Signal acquisition systems provide for the sampling of continuous time electrical signals that may be representative of any of a very wide variety of objects and processes. The familiar PQRST of an electrocardiogram, the echo image of fetus, the blip of a radar screen, a signal waveform traced on the CRT of an oscilloscope; all are electrical signals representative of very different functions. Frequently such analog electrical signals are digitized, not only for the convenience of analysis on digital computers, but also for digital signal conditioning that enhances the display of such information. Ideally, a signal acquisition system provides a distortion-free means for sampling a signal of interest. However, as Werner Heisenberg so famously suggested, the mere act of observing disturbs the observed event. Naturally, the degree of disturbance may vary widely, depending upon the quality of the sampling process and apparatus. But, distortion-free sampling of signals is typically the ideal toward which signal acquisition systems aspire.
Digitization not only provides for convenient analysis and display of acquired signals, because digitized values are much less susceptible to corruption and various forms of degradation than analog signals, post-digitization distortion and error contributions are minimal. That is to say, a large component of signal distortion in digital signal acquisition systems is attributable to the system's analog front end; that is, to that portion of the system that lies in the signal path between the signal acquisition system's input and the system's analog to digital converter (ADC).
A system and method for the compensation of distortion attributable to the front end of a signal acquisition system would therefore be highly desirable.
SUMMARY
In accordance with the principles of the present invention, a digital signal acquisition system includes a front end, an equalizer and an output system. The front end typically includes an input amplifier and may include one or more preamplifiers, and any one of various input probes. The probes may be directed toward different applications. That is, the probes may be voltage probes or current probes, for example, and may include preamplifiers that adjust the range of input signals. The output of the probe is typically routed to an input amplifier, which may include various filters and/other signal conditioning circuitry. The system also includes an analog to digital converter that is arranged to receive a conditioned analog input signal from the input amplifier. The analog to digital converter converts the conditioned analog input signal to digitized input signal. The equalizer accepts the digitized input signal and filters the signal to produce a signal that has been compensated for inadequacies in the signal acquisition system's front end. Coefficients for the equalizer are determined and stored within the system during a calibration process. The calibration process may include the development of coefficients for a variety of signal acquisition system settings. For example, in a digital oscilloscope embodiment, the equalizer may accommodate the use of various types of probe by storing different sets of equalization coefficients corresponding to the different types of probe.
The calibration process involves sending a test signal, such as a step or impulse signal, to a model signal acquisition system and to the signal acquisition system being calibrated. The test signal is a signal of sufficiently high frequency to encompass the range of frequencies over which the signal acquisition system is intended to operate. The output of the model signal acquisition system is Fourier-transformed, as is the unequalized output of the target, that is, uncalibrated, signal acquisition system. The difference between the Fourier transforms of the model and target systems is then computed, thus yielding the amplitude frequency response and group delay for the target signal acquisition system relative to the output of the ideal system. Given the amplitude-frequency response and group delay, conventional methods may be employed to compute the coefficients of a digital filter, an equalizer, that is used to filter the digitized input signal during normal operations. These coefficients are stored and employed by the signal acquisition system during normal signal acquisition operation. The equalized signal may be stored, displayed and/or transmitted to other systems for analysis or data reduction, for example. As previously noted, the calibration process may be repeated for various types of probes and the corresponding sets of equalization coefficients stored for use with the probes in the field.
Additionally, the acquisition system may include a facility for storing the transfer function (in the form of the amplitude/frequency response and group delay) of the model acquisition system. In such an embodiment, the signal acquisition system could also include analytical tools, such as a programmable digital signal processor and accompanying software, that could be used to update the coefficients in the field. The calibration process could employ a calibration signal that would permit the comparison of the acquisition system's transfer function to the model system's stored transfer function, and the development of updated equalization coefficients thereby. The calibration signal used in the field should be closely matched to the calibration signal used to determine the initial coefficients stored within the signal acquisition system.
In an illustrative embodiment, a signal acquisition system that employs an equalizer in accordance with the principles of the present invention takes the form of a digital oscilloscope. The oscilloscope employs various probes for acquiring signals of interest, amplifies and converts those signals to digital form and displays, stores, and/or transmits for analysis or display, the data representative of the input signal. The oscilloscope equalizes the digitized signal before displaying, storing, transmitting or otherwise “outputting” the signal. Equalizer coefficients are determined during a calibration process and stored. The oscilloscope then employs those coefficients in operation in the field to enhance the operation of the oscilloscope. During the calibration process, a relatively high bandwidth continuous time signal is fed to both the oscilloscope and a model signal acquisition system.
In the case of a digital oscilloscope, the model system includes an oscilloscope of much higher bandwidth than the “target” oscilloscope (that is, the oscilloscope being calibrated) and a filter having a predetermined frequency response which mimics the desired response of target oscilloscope. For example, for a 4 Ghz bandwidth target oscilloscope, the model signal acquisition system may include a relatively high bandwidth digital oscilloscope, such as a 50 Ghz oscilloscope, that is configured to receive the calibration signal. A 4 Ghz digital filter having the frequency response desired of the target oscilloscope accepts the digital output of the 50 Ghz signal and filters it to produce a model oscilloscope output. The output from the calibration signal generator (e.g., 50 GHz scope and ideal 4 GHz filter in this example) and the output from the target oscilloscope are each Fourier transformed. The difference between the Fourier transforms is computed and, given this amplitude frequency response and group delay, the coefficients to be used in the target oscilloscope's equalizer are computed and stored within the oscilloscope. Such equalization coefficients may be computed and stored for a variety of input conditions.
REFERENCES:
patent: 5214670 (1993-05-01), Ballatore
patent: 5406429 (1995-04-01), Noguchi et al.
patent: 5418817 (1995-05-01), Richter
patent: 5675394 (1997-10-01), Choi
patent: 569898
Guzik Nahum
Stein Anatoli
Guzik Technical Enterprises
Hoff Marc S.
Le Toan M.
McDermott & Will & Emery
LandOfFree
Method and apparatus for equalization of a signal... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for equalization of a signal..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for equalization of a signal... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3286911