Data processing: measuring – calibrating – or testing – Calibration or correction system
Reexamination Certificate
1998-10-20
2001-08-21
Oda, Christine (Department: 2862)
Data processing: measuring, calibrating, or testing
Calibration or correction system
Reexamination Certificate
active
06278953
ABSTRACT:
TECHNICAL FIELD
The present invention relates to an automated tuning system and, more particularly, to a system that uses a priori knowledge of the general shape of the tuning curve to automate the tuning process.
BACKGROUND OF THE INVENTION
In general, there are many systems that require a first element to be “tuned” with respect to a second element in accordance with a predetermined parameter relating the two. For example, a radio receiver needs to be properly tuned to an RF signal in order to capture a particular transmitted signal. In the field of astronomy, star tracking is accomplished by controlling the image centering using a photomultiplier.
In the field of optical component assembly, it is important that the maximum optical signal power be coupled between the components. In this case, “tuning” relates to the alignment of a first piece part to a second piece part. In most conventional optical component assembly systems, a first component is often affixed to an x-y-z alignment table and a second component is held in a fixed position. The first component is then moved in all three axes, relative to the second component, until maximum coupled power is achieved. Practical, as well as experimental, limitations exist in the conventional methods of obtaining this alignment, as well as with the other “tuning” examples mentioned above. For example, “backlash” associated with the movement of the alignment table has been found to hinder the alignment process.
In more general situations, the need frequently arises where a parameter is changed (for example, relative position of two components in an optical subsystem) and a figure of merit (for example, coupled power between the components) is measured. There are many arrangements for recording the figure of merit and using the recorded data to determine when the parameter has been optimized. What is needed in the industry is a robust technique that takes into consideration the problem of backlash, as well as other sources of “noise” in the tuning process.
SUMMARY OF THE INVENTION
The need remaining in the prior art is addressed by the present invention, which relates to an automated tuning system and, more particularly, to a system that uses the a priori knowledge of the general shape of the tuning curve to automate the tuning process.
In accordance with the present invention, the a priori knowledge of the tuning curve (a “figure of merit”; for example, coupled power as a function of displacement between a pair of optical components) is used to define a mathematical approximation to the tuning curve. The mathematical expression may then be transformed (if necessary) into quadratic form with a negative coefficient. As such, the quadratic will exhibit a local maximum, where the maximum value is associated with the proper alignment between the components. That is, the local maximum can then be used to identify the proper displacement (x-, y- or z-axis) between the components required to achieve maximum coupling.
For a multi-dimensional alignment situation, the process may be repeated to achieve alignment in each dimension. For an exemplary optical x-y-z alignment situation, one process may hold “z” (the axial alignment direction) fixed and find the best x-y alignment for that “z”, then move inward along the z-axis (since the axial displacement is known, a priori, to be greater than optimal axial location) and re-do the x-y alignment at the new “z” location. This process may then be repeated until the best x-y-z alignment is achieved for all three axes.
It is an advantage of the process of the present invention that problems related to backlash are overcome by performing a series of alignment steps around the calculated maximum value.
Other and further advantages of the present invention will become apparent during the course of the following discussion and by reference to the accompanying drawings.
REFERENCES:
patent: 5431664 (1995-07-01), Ureche et al.
patent: 5822687 (1998-10-01), Bickley et al.
“Design of a multivariable self-tuning controller for a class of distributed parameter systems”, Rohal-Ilkiv et al, May 18, 1990, Slovak Technical University, Czechoslovakia.*
“Recursive algebraic curve fitting and rendering”, Shouqing Zhang et al., 1998, Nanyang Technological University, Singapore.
Duong Khoi Hung
Koba Wendy W.
Lucent Technologies - Inc.
Oda Christine
LandOfFree
Automated alignment system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Automated alignment system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Automated alignment system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2515045