Coded data generation or conversion – Analog to or from digital conversion – Analog to digital conversion
Reexamination Certificate
1999-11-11
2001-01-23
JeanPierre, Peguy (Department: 2819)
Coded data generation or conversion
Analog to or from digital conversion
Analog to digital conversion
C341S158000
Reexamination Certificate
active
06177898
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to a method and apparatus for encoding and decoding a waveform, and, in particular, to a method and apparatus for representing a waveform as digital messages which are formed independently of a specific sampling rate.
2. Prior Art
In a conventional system for converting an analog signal into a digital signal, a reference clock is used to generate a periodic train of pulses. Upon each pulse, the analog signal is sampled and an analog to digital conversion is carried out.
At the end of each conversion a binary code is produced which is proportional to the amplitude of the sampled analog signal. The conversion process is repeated within each tick of the clock resulting in a staircase-like amplitude approximation of the original waveform. The accuracy of this reconstruction depends on a number of factors including the rate at which the waveform is sampled, the time required to complete each A/D conversion and the frequency content of the original signal.
In most conventional systems a sampling speed or rate is selected and fixed, as determined by the reference clock, for use during the entire A/D conversion of the original waveform. In selecting the optimum sampling frequency, a sampling theorem commonly known as the “Nyquist Theorem” is most often used. According to the Nyquist Theorem, if a waveform is sampled at a speed or rate that is approximately greater than twice the highest frequency component present in the waveform then in theory it is possible to accurately reconstruct the original waveform from these samples.
As might be expected, in the case of high frequency analog waveforms, A/D conversion methods which employ a Nyquist Theorem based sampling rate tend to generate a high number of data samples. It is preferable, whenever possible, to minimize the total number of samples necessary to reconstruct a waveform. Fewer data samples require a shorter period of time to transmit and require smaller memory reserves for storing and saving the samples.
A high number of samples is particularly undesirable when only a fraction of the samples are needed to accurately reconstruct the original waveform. For example, in an analog waveform having a combination of low and high frequency components it is not necessary and in fact undesirable to sample the low frequency components at the sampling speed necessary for accurately reproducing the high frequency components. Using conventional techniques the entire waveform would be sampled at a sampling rate best suited for reproducing the high frequency components of the multi-component waveform. This would result in an unnecessarily high number of data samples being generated.
In an effort to address these undesirable consequences of employing a fixed conventionally derived sampling rate, there are known in the prior art a number of systems and methods which incorporate a frequency monitoring stage for dynamically varying the sampling rate according to the changing frequencies of a waveform. Upon a change in the frequency component of a sampled waveform, the dynamic stage is designed to vary the sampling rate to a rate which is more appropriate for reconstructing the particular frequency component being sampled. Depending on the waveform, the sample rate may be increased for high frequency components or decreased for low frequency components. As a result of dynamically controlling the sampling rate the conversion is made more efficient and the number of samples taken for accurately reproducing the waveform is optimized.
Podalak U.S. Pat. No. 4,763,207 is an example of a system and apparatus where varying the sampling rate in accordance with the changes in frequency of the sampled waveform is proposed. Although Podalak represents a satisfactory approach for dealing with the potential consequences of utilizing a fixed sampling rate derived using conventional sampling theorems, the Podalak system proposes the use of expensive hardware in the form of a logic analyzer to determine the optimum sampling frequency. In addition to being prohibitively expensive, the Podalak system continues to implement an analog waveform encoding approach which is tied to a sampling frequency.
Kitamura U.S. Pat. No. 4,370,643 discloses yet another system where the sampling frequency may be adjusted to reflect changes in the frequency of the waveform being sampled. Although Kitamura proposes an apparatus and method which can be implemented at a far less cost than that of Podalak, Kitamura remains a method and apparatus which reflects a marriage to the conventional way of thinking in waveform reproduction.
Although the prior art apparatuses and methods have been effective in reducing the total number of data samples necessary to reproduce a waveform of varying frequency, these methods and apparatuses nevertheless represent only an extension or adaptation of conventional methods. That is, although they manage to circumvent some of the drawbacks of the prior art systems, they continue to rely on the basic theory underlying the prior art systems, namely, A/D conversion of a waveform utilizing an “ideal” sampling rate. As a result, the converted analog waveform data is still represented as a series of amplitude points corresponding to the sampled original waveform which must, at some point, be passed through a smoothing filter in order to approximately reconstruct the original waveform.
Furthermore, the problems common to conventional A/D conversion techniques which rely on a sampling rate including quantization errors and aliasing have not been overcome. As is known in the art, quantization noise and aliasing can significantly deteriorate the quality and accuracy of the reproduced waveform.
Thus a need continues to exist for a method and apparatus for compressively approximating an analog waveform which does not rely on a fixed or varying sampling rate.
Still a further need exists for a system and apparatus which can be used to compressively approximate an analog waveform without the need for an inordinate amount of memory for storing derived samples.
The object of the present invention, like the prior art systems, is to encode or represent an original analog waveform as a series of digital messages which can be used to precisely reconstruct the original analog waveform. It is expected that the benefits of the present invention will extend to applications in the field of Digital Audio, data transmission over low to medium bandwidth networks such as the Internet and to the field of data acquisition in general.
To that end, it is a general object of the present invention to provide a method and apparatus that has intrinsic advantages over conventional conversion techniques in that there is no one to one relationship between the number of data samples and a system or reference clock and as a consequence does not have a fixed data rate per second. Instead, the number of sample data dynamically fluctuates based on the frequency and amplitude content of an original analog waveform.
It is another object of the present invention to provide a method and apparatus which is not based on the sampling theorem or variations thereof.
It is a further object of the present invention to provide a method and apparatus which automatically adjusts to varying frequency components in a sampled analog waveform.
It is yet another object of the present invention to provide a method and apparatus which performs dynamic data compression of the previously sampled data.
It is yet a further object of the present invention to provide a method and apparatus which permits the reconstruction of an original waveform from the sample data without using smoothing circuitry.
SUMMARY OF THE INVENTION
In accordance with the stated objectives and other objectives which will hereinafter become apparent, there is provided a method and apparatus for representing an analog waveform as a plurality of digital messages describing a segment or sample of the analog waveform at a particular moment in time. In particular, each dig
Amster Rothstein & Ebenstein
Jean-Pierre Peguy
LandOfFree
Method and apparatus for representing an analog waveform as... 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 representing an analog waveform as..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for representing an analog waveform as... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2522288