Data processing: measuring – calibrating – or testing – Measurement system in a specific environment – Electrical signal parameter measurement system
Reexamination Certificate
2001-10-09
2003-10-21
Hoff, Marc S. (Department: 2857)
Data processing: measuring, calibrating, or testing
Measurement system in a specific environment
Electrical signal parameter measurement system
C702S058000, C702S085000, C702S124000, C331S018000, C331S023000, C331S025000, C331S034000, C324S076410, C324S076440, C324S085000, C367S043000, C367S045000, C367S049000, C367S080000, C367S100000
Reexamination Certificate
active
06636816
ABSTRACT:
TECHNICAL FIELD OF THE INVENTION
The present invention relates to the field of signal measurement. More specifically, the present invention relates to the field of integral and simultaneous signal measurement and measurement device calibration.
BACKGROUND OF THE INVENTION
For accurate instrumentation, it is desirable to fully understand the characteristics of the devices used. With any given one of these devices, i.e., a “device-under-test” or DUT, the specific characteristics of that DUT are not fully understood. As a simplistic example, a given “47 k&OHgr;” resistor would rarely have a value of exactly 47,000&OHgr;. To properly understand the operation of a circuit containing such a resistor, acknowledge of the actual resistance (e.g., 46,985.42&OHgr;) would be helpful. It should be understood that a DUT may encompass a wide range of electrical components, equipment, and systems. A typical DUT may be a filter, an amplifier, a transmitter, a receiver, or any component, group of components, circuit, module, device, system, etc.
The drift of components, filters, amplifiers, and other signal-conditioning circuitry typically limits the accuracy of electronic measurements. A measurement system may advertise a large dynamic range and very-high resolution. However, the full dynamic range and resolution may not be realizable because of errors inherent in the system. Current measurement systems have not been demonstrated to have verifiable, full-scale uncertainties of better than 0.1 percent over all types of errors.
Real-world measurement instruments tend not to be perfect. This imperfection will affect the accuracy of the resulting measurements. This accuracy is dependent upon measurement errors. Such errors may be classed as static (systematic) errors and dynamic errors.
Static errors are repeatable, time-invariant system errors. That is, static errors do not vary over time. Static errors result from the nonideal aspects of a system. These errors are repeatable as long as no changes are made to the system. Static errors include directivity errors, source-mismatch errors, load-mismatch errors, reflection and transmission tracking errors, isolation or cross-talk errors, etc.
Static errors may be reduced through the use of precision components and circuits. However, no matter how precisely a circuit is designed, there will still be some level of static error present. Since static errors are repeatable, they can be suppressed using various static error suppression techniques, such as twelve-term error modeling, known to those skilled in the art. Twelve-term error modeling, typically employed with standard network analyzers, can account for directivity errors, source-mismatch errors, load-mismatch errors, tracking errors, and isolation errors.
Before error modeling may be employed, the error coefficients of the requisite equations must be calculated by making a set of measurements on a set of known loads meeting precise standards. A sufficient number of precise standards must be used in order to determine the various error coefficients in the error model. A common static-error calibration technique is the Short, Open, Load, and Through (SOLT) technique. The SOLT calibration technique yields better than 0.1 percent accuracy for static errors. However, dynamic errors limit the actual accuracy to less than this. An alternative static-error calibration technique, also well-known to those skilled in the art, is the Through, Reflection, and Load (TRL) technique. The TRL calibration technique yields significantly better static-error accuracy than the SOLT technique at the cost of calibration complexity. There are a number of other well-known static-error calibration techniques that may be used to advantage in specific instances.
However, removing a significant amount of static error achieves little if other types of errors are not also addressed. Dynamic errors are time-varying errors. That is, dynamic errors change over time, often in an unpredictable manner. Such errors may be attributable to a number of different sources. For example, component-drift, physical-device (e.g., cables, connectors, etc.) errors, phase-noise errors, random noise, etc.
Measurement systems exhibit several types of dynamic errors. Phase-noise errors and random errors, while inherently dynamic (i.e., time-variant) are special cases independently discussed hereinafter.
Component-drift errors may be either long-term or short-term. Long-term component-drift errors, however, are typically due to aging of the components, with resultant variations in component specifications.
One type of short-term component-drift error, source drift error, is usually attributable to thermal or mechanical variations within the system, and may include both amplitude and phase fluctuations of the output wave.
Another type of short-term component-drift error, receiver drift error, is associated with a data receiver. This may be due to drift in components such as amplifiers, filters, and analog-to-digital converters. Receiver-drift error may also appear as time-varying gain and phase drift in the received signals.
Thermal variations may also lead to physical expansion of passive components within the system. At high frequencies, such expansion may lead to appreciable phase errors. In applications where the DUT is located at a considerable distance from a transmitter and/or receiver, there may be a number of time-varying errors associated with the connections between the DUT and the transmitter and receiver. These may include amplitude and phase-drift errors in the amplifiers or errors associated with the modulation and demodulation circuitry. Systems in which such errors become significant include systems utilizing propagation media other than traditional cables (e.g., the atmosphere, space, the earth, railroad tracks, power transmission lines, the oceans, etc.).
Dynamic physical errors result from physical changes in the test setup. One example of a physical error is connector repeatability. As one connects and disconnects the DUT, there will be reflection and transmission errors associated with any nonrepeatability of the connectors. The severity of the connector repeatability error is related to the type of connector, the condition of the connector, and the care with which the user makes the connection.
Another type of dynamic physical errors are cable-flexure errors. Cable-flexure errors appear as one moves the cables to connect or disconnect a DUT or perform a calibration. Time-variant phase errors associated with the relaxation of the cable can occur for a period of time after the cable has been flexed.
Phase noise (jitter) is directly related to the frequency stability of a signal source. In a perfect sinusoidal oscillator, all the energy would lie at a single frequency. Since oscillators are not perfect, however, the energy will be spread slightly in frequency. This results in a pedestal effect. This effect, referred to as phase noise, is more severe at higher frequencies. Phase noise is a performance-limiting factor in applications where a weak signal must be detected in the presence of a stronger, interfering signal.
Random or white noise is common in measurement systems. Random noise includes thermal noise, shot noise, and electromagnetic interference. Random noise may appear as random data errors. Traditional and well-known techniques of random error suppression utilize some form of oversampling to determine the correct data and suppress the random errors.
Calibration frequency is also a problem in conventional signal measurement systems. Typically, high-accuracy measurement systems employing manual calibration are calibrated periodically. The interval between calibrations may be hourly, daily, weekly, monthly, quarterly, or even yearly. This technique produces a steadily decreasing accuracy that progresses over the inter-calibration interval. Additionally, drift errors occurring during the inter-calibration interval are uncompensated. Such drift errors tend to accumulate. Therefore, measurements taken shortly before calibration m
Dvorak Steven L.
Sternberg Ben K.
Desta Elias
Gresham Lowell W.
Hoff Marc S.
Meschkcw Jordan M.
Meschkow & Gresham P.L.C.
LandOfFree
Vector signal analysis method and apparatus therefor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Vector signal analysis method and apparatus therefor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Vector signal analysis method and apparatus therefor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3133536