Wave transmission lines and networks – Automatically controlled systems
Reexamination Certificate
2001-05-22
2003-07-15
Ham, Seungsook (Department: 2817)
Wave transmission lines and networks
Automatically controlled systems
C333S174000
Reexamination Certificate
active
06593828
ABSTRACT:
FIELD OF THE INVENTION
The present invention generally relates communications systems wherein signal information is processed both in digital and analog forms. More specifically, the invention relates to the tuning and adjustment of frequency-selective filters that have finite transmission zeros.
BACKGROUND OF THE INVENTION
In communications systems, the need for frequency-selective filtering of analog signals frequently arises. In this regard, a receive filter is an example of a frequency-selective filter of analog signals in a communications system. A receive filter operates to remove signals with frequencies outside of a determined frequency range so that only a desired signal (i.e., the “receive signal”) is received by a receiver. The signals that are removed by the receive filter have frequencies that are, depending on the filter type (e.g., low-pass or high-pass), above or below a determined cut-off frequency (“cutoff”).
A transmit filter is another example of a frequency-selective filter of analog signals in a communications system. A transmit filter operates to ensure that a transmitter only transmits signals in the frequency range allotted to the transmitter (i.e., the “transmit signal”) by removing other, spurious signals that may be introduced into the transmit signal, for example, due to typical imperfections in the transmitter circuitry. These removed signals also have frequencies that are above or below a cutoff frequency.
Frequency-selective filters can be implemented in many ways, as is known in the art. For example, such filters may be implemented using components such as resistors, capacitors, inductors, transconductances, or controlled (i.e., dependent) sources. Although the following description of the present state and needs in the art mainly discusses an integrated circuit (IC) implementation of frequency-selective filters, it should be understood that the description also applies to any other frequency-selective filter implementations, for example, gm-C (transconductance-C), gyrator-based, MOS-R (metal-oxide semiconductor-R), etc.
As is known in the art, frequency-selective filters perform a filtering function, for example as described above, in a signal processing system such as a communications system transmitter. In this regard, filters should meet, among other requirements, three accuracy-related requirements concerning the filter cutoff. These three requirements are: 1) the general nominal cutoff frequency; 2) the accuracy with which the general nominal cutoff is achieved; and 3) the accuracy with which the cutoff is maintained over time and temperature variations. Further, the strictness of these requirements is dependent on the system that the filter is used in.
In regard to the first requirement, i.e. the general nominal cutoff, the filter should operate with a cutoff that is sufficiently close (i.e., within an acceptable tolerance range) to the required cutoff for the filter application. In this regard, the need may arise to tune or adjust the filter cutoff depending on the components utilized to implement the filter. For example, if a filter design is implemented utilizing IC resistors and capacitors, the filter cutoff may vary significantly depending on the manufacturing batch of IC components utilized to construct the filter. Typically, the variation will be too significant for the filter to be useable as designed unless it is tuned (i.e., the cutoff is adjusted). Tuning of the filter in this regard, which will be referred to as “initial tuning”, is conducted for the purpose of selecting an operating mode of frequency band and correcting the filter performance due to manufacturing tolerances of the IC components. Typically, initial tuning is accomplished by adjusting the filter cutoff based on the observed filter output of known frequency test signals that are transmitted through the filter.
In regard to the second accuracy-related requirements of the filter cutoff, i.e. the accuracy with which the general nominal cutoff is achieved, this requirement concerns the available accuracy for the initial tuning. In order to satisfy the second requirement, the filter should be capable of adjustment during initial tuning that is fine enough to set the cutoff sufficiently close to the general nominal cutoff described above as the first requirement. In this regard, if the filter cutoff variations due to manufacturing tolerances are small, only a narrow tuning range is needed to meet the second requirement. But, if the filter cutoff variations due to manufacturing tolerances are large, such as when the filter is implemented using IC components, a considerably larger tuning range is needed.
In regard to the third requirement, i.e. the accuracy with which the cutoff is maintained over time and temperature variations, this requirement concerns the stability of the frequency cutoff and it is distinct from the initial tuning requirement (i.e., the second requirement). In order to satisfy the third requirement, the filter should be capable of providing sufficient drift compensation. Drift occurs, for example, when one or more components of the filter (e.g., a resistor or capacitor) has a significant temperature coefficient such that a change in the ambient or operating temperature of the component causes a change in its operating characteristic (e.g., an increase/decrease in resistance or capacitance). This operating characteristic change in the filter component causes the filter cutoff to drift, and this resulting drift may be significant enough to interfere with or disrupt the operation of the signal processing system that the filter is integrated in to. Drift compensation provides for the correction of the filter cutoff in response to drift caused by temperature variations or other conditions. In comparison to the adjustment fineness required for initial tuning, the adjustment fineness required for adequate temperature drift compensation is generally much higher. Additionally, adjustment of the filter cutoff while the system is operating may cause significant transients which severely disrupt the system operation. Therefore, the drift compensation provided by the filter must be designed such that the adjustment transients are sufficiently small enough to avoid system operation disruption.
Presently in the art, the implementation of filters that satisfy the three accuracy-related cutoff requirements described above is generally addressed in one of two ways. In the first way, the need for initial tuning and drift compensation is avoided altogether by implementing the filter such that it is highly accurate and drift-stable. As discussed above, the cutoff accuracy and drift performance of a filter are dependent on the type of components used to implement it. Thus, in order to satisfy the three accuracy related requirements in this first manner, the implementation of the filter must be limited to certain types of components, and this restriction may impose cost inflation and overall application limitations for the filter. For example, meeting the requirements in this manner generally restricts the use of IC components in the filter implementation because of the variation issues discussed above that affect the filter cutoff accuracy and drift. Thus, the low cost and high compactness benefits of IC components are unavailable for filter implementation in this manner.
In the second way of addressing the cutoff accuracy requirements, the system that the filters operates in is designed such that the filter cutoff accuracy requirements are loose enough to accommodate the particular filter implementation. For example, if a filter is implemented in the system using IC components, the filter cutoff accuracy requirements of the system are made relatively loose to accommodate the variation issues discussed above that affect the filter cutoff accuracy and drift. Addressing the cutoff accuracy requirements in this manner may impose cost inflation and application limitations for the entire system that the filter is integrated in to.
There are many situations where addressing
Helfenstein Markus
Lim Drahoslav
Moschytz George S.
GlobespanVirata Inc.
Ham Seungsook
Thomas Kayden Horstemeyer & Risley
LandOfFree
System and method for filter tuning does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for filter tuning, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for filter tuning will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3051781