Amplifiers – With semiconductor amplifying device – Including differential amplifier
Reexamination Certificate
2001-02-22
2003-05-27
Mottola, Steven J. (Department: 2817)
Amplifiers
With semiconductor amplifying device
Including differential amplifier
C330S254000, C330S258000, C330S283000
Reexamination Certificate
active
06570446
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to methods and apparatus for improving the performance of electronic devices, and more specifically to methods and apparatus using cross-degeneration techniques for improving the linearity, noise, gain and power consumption characteristics of transconductance cells, and of amplifiers, mixers, continuous-time filters and other active elements based upon such cells.
2. Description of Related Art
As is well known, linear amplification is a very important function in many electronic devices including those designed for use in signal processing applications. Under ideal conditions, the transconductance, g
m
, of an amplifying circuit remains constant for all input values. Disadvantageously, practical amplifiers are typically implemented with devices that are fundamentally non-linear. As a consequence, the transfer functions of the non-linear devices vary greatly and depend upon the magnitude of the applied input signal. The prior art is replete with attempts at improving the performance characteristics of electronic devices using these non-linear components. Prior art circuit design techniques have been proposed that compensate for variations in the transfer functions of non-linear devices and thereby increase the input range of circuits using these devices, without also adversely affecting noise characteristics. In amplification applications, for example, prior art circuit design techniques have been proposed that compensate for these non-linearities in order to maximize the magnitude of the input signal to be amplified.
In one exemplary prior art circuit design technique, the supply current is increased in order to increase the linear range overall gain of the circuit. Using another design technique commonly referred to as “degeneration”, feedback is introduced into the design of the electronic circuit in an attempt to force the overall gain of the circuit to behave in a more linear manner. Disadvantageously, this approach substantially increases noise and power consumption while reducing gain and therefore may not be attractive in applications where noise or power considerations play an important role. Another technique combines signals produced by a plurality of interconnected devices in an attempt to more precisely shape and configure the overall device transfer function to meet specific design parameters. Disadvantageously, this technique requires use of an increased number of interconnected devices to proportionally increase the input range of the device. The performance of devices made in accordance with this technique disadvantageously depends upon the precision to which the devices are matched and upon the tolerance values of each circuit element. In addition, because an increased number of devices are needed, the technique also suffers from the disadvantages of increased overall circuit area, increased circuit complexity, increased power consumption, and reduced speed. Further, the applicability of this particular prior art technique is restricted to use in bipolar device technologies and sub-threshold MOS technologies.
FIG. 1
illustrates a circuit design technique referred to as a “hybrid doublet” (HD) technique that combines degeneration with a multi-tanh signal shaping approach. A traditional Hybrid Doublet circuit
10
is shown in FIG.
1
. The multi-tanh concept is a well known technique for extending the capacity of a transconductance g
m
cell (implemented in bipolar technologies), or an amplifier, mixer, continuous-time filter, or other active element based on such a cell, by using at least two differential transistor pairs operating in parallel. Each of the differential transistor pairs has a base offset voltage that splits the individual g
m
functions along the input-voltage axis. This allows the cell to process larger input voltage swings while allowing the overall transconductance of the cell to be more linear. The HD technique, and the description of the circuit
10
of
FIG. 1
, is described in more detail in a paper authored by Barrie Gilbert, entitled “The Multi-tanh Principle: A Tutorial Overview”, published in the IEEE Journal of Solid-State Circuits, Volume 33, No. 1, in January, 1998, which is incorporated herein by reference for its teachings on HD circuit design techniques.
The HD approach exemplified in the circuit of
FIG. 1
suffers from many shortcomings. For example, the HD technique does not provide for high-order derivative cancellation. As a consequence, distortion cannot be carefully controlled or reduced. Another consequence of this limitation is that third-order distortion in the HD topology does not decrease monotonically with decreasing input power. In addition, as noted above in connection with the other prior art signal shaping approaches, the HD technique relies upon a precise matching of component values given a fixed power consumption specification. This restriction fixes the gain and the input range of the circuit. This, in turn, limits a circuit designer's flexibility in tailoring the circuit for a desired application. For example, a circuit designer may wish to improve the input range and linearity of a circuit by allowing the circuit to consume more power. Disadvantageously, when selecting a topology based upon the prior art HD technique, the designer does not have this tradeoff flexibility. Thus, the components used in circuits designed in accordance with the HD approach are not easily scaleable.
Another disadvantage of the prior art HD approach is that circuit performance is quite sensitive to circuit parameters. As a consequence, variations in integrated circuit processes yield substantial degradations in signal distortion. A further disadvantage of circuits designed in accordance with the HD approach is that degeneration resistors (i.e., R
e1
25
and R
e1
/A
e
50
of
FIG. 1
) result in a loss in current “headroom” in the circuit. In addition, the choice of degeneration resistor values relies upon complex empirical mathematical formulae derived from the tanh characteristics of the hybrid doublet. Disadvantageously, this significantly constrains the flexibility and usefulness of the circuit. Consequently, circuit performance tradeoffs cannot be readily designed using the prior art HD approach. Because arbitrary input signal ranges cannot be accommodated, practical circuit performance parameters such as noise, gain, and linearity are not easily implemented using the HD approach. Finally, the HD approach disadvantageously is restricted to use in bipolar technologies.
Another exemplary design technique used in the prior art in an attempt to “linearize” electronic circuits utilizes sophisticated feedback circuits having high gain characteristics. An example of such a feedback circuit is an operational amplifier feedback circuit. However, similar to the other prior art approaches described above, this approach disadvantageously substantially increases the complexity of circuit design, reduces the overall speed and bandwidth of the device, increases the circuit area required to implement the device, and does so with limited enhancement of device linearity.
Therefore, the need exists for a method and apparatus that improves the performance of electronic devices, and more specifically, that enhances the linearity of electronic devices. The need exists for a compact circuit topology that provides enhanced device linearity as compared with traditional approaches. The enhancement in linearity should be accomplished: (a) without unduly increasing noise generated by a device; (b) without substantially increasing device complexity; (c) without substantially reducing the speed or bandwidth of the device; and (d) without substantially increasing the device power consumption. The power consumed by a device using this technique should be comparable to devices using the prior art “linearization” approaches.
The method and apparatus for enhancing the linearity of electronic devices also should be easily “scaleable.” That is, the circuit should be able to acco
Jaquez & Associates
Jaquez, Esq. Martin J.
Mottola Steven J.
Silicon Wave Inc.
LandOfFree
Method and apparatus for improving the performance of... 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 improving the performance of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for improving the performance of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3057803