Miscellaneous active electrical nonlinear devices – circuits – and – Signal converting – shaping – or generating – Amplitude control
Reexamination Certificate
2001-11-01
2003-04-22
Le, Dinh T. (Department: 2816)
Miscellaneous active electrical nonlinear devices, circuits, and
Signal converting, shaping, or generating
Amplitude control
C327S331000, C327S350000, C327S553000, C330S278000, C333S014000
Reexamination Certificate
active
06552591
ABSTRACT:
BACKGROUND OF THE INVENTION
This invention relates to processing single channel or single-ended analog signals in a manner which accommodates a wide dynamic range of signal amplitude.
Radio and other wireless receivers must accommodate a wide range of signal levels. This wide dynamic range involves many factors such as path loss and variation in receiver gain. Automatic gain control (AGC) circuits have conventionally been employed for automatically controlling the gain of the receiver to narrow the dynamic range within the receiver. Typically, these gain control circuits have employed feedback circuitry to adapt the gain. Input signal level changes typically cause gain changes in these feedback loops, which require time to damp out, since each current output value is affected by the preceding values. Hence, the feedback loop introduces delay in the response which causes a transient in the desired output.
Logarithmic amplifiers have been used extensively in processing signals with wide dynamic range. Logarithmic amplifiers that are well known are described in detail in two books by Richard Smith Hughes Logarithmic Amplification and Logarithmic Video Amplifiers, cited here for general background in the field. A true logarithmic amplifier accurately reflects at its output the logarithm of the input signal, including sign, over the useful range of the amplifier. This is not to be confused with amplifiers which provide as output the logarithm of the envelope of the input signal.
In U.S. Pat. No. 4,551,688, entitled “DELAY COMPENSATED AUTOMATIC GAIN CONTROL” in the name of Craiglow,
FIGS. 2 and 3
show prior art implementations of feedback-based AGC circuits. U.S. Pat. No. 5,283,536, in the name of Wheatley et al., incorporates feedback to implement the same function.
A circuit used as automatic gain control is described in U.S. Pat. No. 5,014,013, entitled “ANTILOG CIRCUIT WITH AUTOMATIC GAIN CONTROL” in the name of Kotzian et al. In this patent, the circuit described uses feedback circuitry utilizing a bipolar transistor to linearize and normalize the amplitude modulation on a high frequency carrier.
A gain controlled amplifier with temperature compensation circuit used in automatic gain controlled circuits with feedback is described in U.S. Pat. No. 5,408,697, entitled “TEMPERATURE-COMPENSATED GAIN-CONTROLLED AMPLIFIER HAVING A WIDE DYNAMIC RANGE” in the name of Price et al. Another circuit with “linear in dB” gain controlled characteristics is described in U.S. Pat. No. 5,572,166 entitled “LINEAR IN DB VARIABLE GAIN AMPLIFIER” in the name of Gilbert.
In U.S. Pat. No. 5,627,857 entitled “LINEARIZED DIGITAL AUTOMATIC GAIN CONTROL” in the name of Wilson describes a digital implementation of an automatic gain control circuit with feedback. U.S. Pat. No. 5,838,194, entitled “CONSTANT SETTLING TIME AUTOMATIC GAIN CONTROL CIRCUITS” in the name of Khoury, describes a similar feedback-type circuit with constant response for all amplitude levels.
What is needed are techniques for extracting modulation characteristics from a signal without undesirable performance characteristics of feedback-based AGC circuits.
SUMMARY OF THE INVENTION
According to the invention, method and apparatus are provided for processing a signal without using feedback along the signal path. The input is typically a single-channel or single-ended wide dynamic range analog signal wherein processing makes use of a dynamic range compressor stage and subsequent operations on the compressed signal. Typically, the compressor stage makes use of a polarity sensitive, compressive nonlinear transfer function, typically implemented with a true logarithmic amplifier, namely an amplifier whose magnitude response to the input signal is substantially logarithmic and which is additionally sensitive to the sign of the input. This type of amplifier is effective for bipolar signals. According to the invention, a characteristic related to the average amplitude level of the compressed signal is detected and is used to produce an output signal whose amplitude is nominally equal to a chosen value. A frequency selective filter is employed to select the frequencies used in this averaging and which will not appear at the output; that is, allowing modulation characteristics to pass while setting the nominal value of the output signal to a desired level, even in the presence of signals with non-constant envelope. The characteristics of this filter are optionally controllable by means of a response control input rather than feedback. This controllability is desirable in accommodating signals with differing modulation characteristics. The characteristics of the frequency selective filter are used as a control input to a polarity sensitive exponentiation circuit.
In one embodiment, the polarity sensitive exponentiation circuit is implemented using an emitter-coupled transistor pair that implicitly detects and averages the amplitude of the compressed input signal. The exponentiating transistor pair also serves as the amplitude detector.
In a second embodiment, a separate signal path is used to determine the amplitude of the compressed signal. A synchronous or full-wave detector is employed to determine the amplitude of the compressed signal. One method of implementing a synchronous detector is to multiply the signal by its sign. The sign of the compressed signal may be produced by a hard-limiter. The output of the synchronous detector is applied to the frequency selective filter that may have controllable filter characteristics. To compensate for the group delay introduced by the filtering operation, a delay element may be inserted in the compressed signal path to the exponentiator. The amount of delay may nominally be chosen to equal the group delay introduced by the filtering operation. The output of the filter is used as an additional input to the exponentiation process, the first input being the output of the delay element. The nominal output level of the exponentiation circuit may be adjusted by means of an amplitude control input and a summer placed between the filter and the exponentiation block. The exponentiation process may be implemented as the difference between two exponential responses that produces an arc sine hyperbolic response. This is seen to be an accurate approximation to the polarity sensitive exponentiator.
In a third embodiment, a synchronous detector is used together with an emitter-coupled transistor exponentiation circuit, representing a specific implementation of the second embodiment.
In a fourth embodiment, a polarity-sensitive exponentiator is used. This requires that the subtraction of the average amplitude be done in a more complex manner. Specifically, the output of the frequency selective filter, as modified by the amplitude control, is modulated onto a square wave. The result of this modulation step is subtracted from the compressed signal, optionally delayed by a delay element, and applied to a polarity sensitive exponentiator.
The present invention is highlighted by the use of a true logarithmic amplifier in conjunction with filtering and exponentiation circuits to extract modulation characteristics from a signal without undesirable performance characteristics of feedback-based AGC circuits. In particular, several novel circuit topologies are disclosed which are useful to illustrate various aspects of the present invention.
This invention represents a further development in the field of wide dynamic range amplifiers of the present inventors, as represented by co-pending patent application Ser. No. 09/715,395 entitled METHOD AND APPARATUS FOR PROCESSING A MULTIPLE-COMPONENT WIDE DYNAMIC RANGE SIGNAL filed Nov. 16, 2000, the content of which is incorporated herein by reference.
REFERENCES:
patent: 4551688 (1985-11-01), Craiglow
patent: 4944024 (1990-07-01), Leveque
patent: 5014013 (1991-05-01), Kotzian
patent: 5283536 (1994-02-01), Wheatley, III et al.
patent: 5408697 (1995-04-01), Price et al.
patent: 5572166 (1996-11-01), Gilbert
patent: 5627857 (1997-05-01), Wilson
patent: 5838194 (1998-11-01), Khoury
Abadi Kamran Khorram
Lorenz Robert Gustav
Walker James T.
Allen Kenneth R.
Le Dinh T.
PiRadian, Inc.
Townsend and Townsend / and Crew LLP
LandOfFree
Method and apparatus for processing a wide dynamic range 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 processing a wide dynamic range signal, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for processing a wide dynamic range signal will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3088960