SYSTEM AND METHOD FOR STATISTICALLY SEPARATING AND CHARACTERIZING NOISE WHICH IS ADDED TO A SIGNAL OF A MACHINE OR A SYSTEM

• US 20090248767A1
• Filed: 05/22/2009
• Published: 10/01/2009
• Est. Priority Date: 02/10/2005
• Status: Active Grant

• Current Assignee: Mineral Lassen LLC (Intellectual Ventures LLC)
• Original Assignee: Nemor Properties LLC (Intellectual Ventures LLC)
• Inventors: Moriya, Netzer
• Granted Patent: US 8,285,764 B2
• Time in Patent Office: Days
• US Class Current: 708/200
• CPC Class Codes: G06F 17/18 for evaluating statistical data , e.g. average values, frequency distributions, probability functions, regression analysis forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04

Abstract

Method for finding the probability density function type and the variance properties of the noise component N of a raw signal S of a machine or a system, said raw signal S being combined of a pure signal component P and said noise component N, the method comprising: (a) defining a window within said raw signal; (b) recording the raw signal S; (c) numerically differentiating the raw signal S within the range of said window at least a number of times m to obtain an m order differentiated signal; (d) finding a histogram that best fits the m order differentiated signal; (e) finding a probability density function type that fits the distribution of the histogram; (f) determining the variance of the histogram, said histogram variance being essentially the m order variance σ²_(m) of the noise component N; and (g) knowing the histogram distribution type, and the m order variance σ²_(m) of the histogram, transforming the m order variance σ²_(m) to the zero order variance σ²₍₀₎, σ²₍₀₎ being the variance of the pdf of the noise component N, and wherein the histogram type as found in step (e) being the probability density function type of the noise component N.

Claims

1-14. -14. (canceled)

15. A method for an electronic noise estimation unit to characterize noise in a signal, the method comprising:

numerically differentiating the signal within a window at least m number of times to obtain an m-order differentiated signal;

providing a histogram having a distribution that approximates the m-order differentiated signal;

providing a probability density function type that approximates the distribution of the histogram; and

outputting the probability density function type.

16. The method of claim 15, wherein the probability density function type that approximates the distribution of the histogram is selected from a plurality of probability density function types.

17. The method of claim 15, wherein the distribution of the histogram best fits the m-order differentiated signal and the probability density function type best fits the distribution of the histogram.

18. A method for an electronic noise estimation unit to characterize noise in a signal, the method comprising:

numerically differentiating the signal within a window at least m number of times to obtain an m-order differentiated signal;

providing a histogram having a distribution that approximates the m-order differentiated signal;

providing a probability density function type that approximates the distribution of the histogram;

determining a variance of the histogram;

determining a zero-order variance using the variance of the histogram and the probability density function type; and

outputting at least one of the probability density function type or the zero-order variance.

19. The method of claim 18, wherein the signal comprises a noise component and a raw signal component.

20. The method of claim 19, wherein the variance of the histogram is the m-order variance of the noise component from the signal.

21. The method of claim 19, wherein the zero-order variance is a variance of the probability density function of the noise component from the signal.

22. The method of claim 18, wherein the window of the signal is recorded.

23. The method of claim 18, wherein the probability density function type that approximates the distribution of the histogram is selected from a plurality of probability density function types.

24. The method of claim 18, wherein the distribution of the histogram best fits the m-order differentiated signal and the probability density function type best fits the distribution of the histogram.

25. An apparatus for estimating noise in a signal, the apparatus comprising:

a differentiation module configured to receive the signal and numerically differentiate the signal within a range of a window at least m number of times to obtain an m-order differentiated signal;

a first module configured to determine a histogram that approximates the m-order differentiated signal;

a second module configured to determine a probability density function type that approximates the distribution of the histogram;

a third module configured to determine a variance of the histogram;

a fourth module configured to determine a zero-order variance using the variance of the histogram and the probability density function type; and

an interface module configured to output at least one of the zero-order variance or the probability density function type.

26. The apparatus according to claim 25, wherein the apparatus is further configured to determine another probability density function type and another zero-order variance for a noise component of the signal within another range of another window.

27. The apparatus of claim 25, wherein the probability density function type is determined from a plurality of probability density function types.

28. The apparatus of claim 25, wherein the first, second, third, and fourth modules are the same module.

29. The apparatus of claim 25, wherein the zero-order variance estimates a variance of the probability density function type of a noise component from the signal.

30. A system for estimating a pure signal component of a raw signal that includes the pure signal component and a noise signal component, the system comprising:

an electronic noise estimation unit including;

a differentiation module configured to receive a raw signal and numerically differentiate the raw signal within a range of a window at least m number of times to obtain an m-order differentiated signal;

a first module configured to determine a histogram that approximates the m-order differentiated signal;

a second module configured to determine a probability density function type that approximates the distribution of the histogram;

a third module configured to determine a variance of the histogram; and

a fourth module configured to determine a zero-order variance using the variance of the histogram and the probability density function type; and

a filter coupled to the noise estimation unit, wherein the filter is configured to receive the zero-order variance and the raw signal and to output the estimated pure signal component of the raw signal.

31. The system of claim 30, wherein the filter is an adaptive filter.

32. The system of claim 30, wherein the filter is a Kalman Filter.

33. The system of claim 30, wherein the electronic noise estimation unit is configured to determine a zero-order variance on the raw signal in real time.