Methods and apparatus for evaluating operational integrity of a data processing system using moment bounding

• US 7,089,452 B2
• Filed: 09/25/2002
• Issued: 08/08/2006
• Est. Priority Date: 09/25/2002
• Status: Expired due to Fees

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Raytheon Company (Rtx Corporation)
• Inventors: Rubin, Arthur Leonard, Schempp, Timothy Robert
• Primary Examiner(s): Beausoliel, Robert
• Assistant Examiner(s): GUYTON, PHILIP A
• Application Number:
  US10/255,029
• Publication Number:
  US 20050015680A1
• Time in Patent Office: 1,413 Days
• Field of Search: 701/214
• US Class Current: 714/32
• CPC Class Codes: G01S 19/08 providing integrity information, e.g. health of satellites or quality of ephemeris data
• CPC Class Codes: G01S 19/20 Integrity monitoring, fault detection or fault isolation of space segment
• CPC Class Codes: G06F 11/008 Reliability or availability analysis

Abstract

Mechanisms and techniques provide a system to evaluate operational integrity of a data processing system such as an area augmentation system by obtaining at least one set of distribution data related to information processed by the data processing system. The set(s) of distribution data can include corrections information and integrity bounds information related to the corrections information. The system applies a moment generating function to the set of distribution data to produce a moment bounded distribution data result and compares the moment bounded distribution data result to an integrity limit to determine if the data processing system operates within an acceptable integrity threshold. The system can be used in real-time or non-real time with sets of distribution data that are non-Gaussian, non-symmetric and/or non-unimodal.

Claims

1. A method for evaluating operational integrity of a data processing system, the method comprising the steps of:

obtaining at least one set of distribution data related to information processed by the data processing system;

applying a moment generating function to the at least one set of distribution data to produce a moment bounded distribution data result; and

comparing the moment bounded distribution data result to an integrity limit, representing a maximum value related to the collective set of errors represented within the at least one set of distribution data, to determine if the data processing system operates within an acceptable integrity threshold.

2. The method of claim 1 wherein the at least one set of distribution data includes multiple sets of distribution data, each of which is at least one of:

i) non-unimodal;

ii) non-symmetric having a non-zero mean; and

iii) distribution data collected from actual operation of the data processing system.

3. The method of claim 1 wherein:

the distribution data includes at least one integrity estimate that forms a composite error bound; and

wherein the steps of applying the moment bounding function and comparing the moment bounded distribution data result determine that every possible combination of errors represented within the distribution data is bounded by the at least one integrity estimate.

4. The method of claim 1 wherein:

the data processing system is a signal-based navigation system;

the at least one set of distribution data includes corrections information and integrity estimate information related to a location of a user of the signal-based navigation system;

wherein the steps of applying a moment generating function and comparing the moment bounded distribution data result collectively perform a convolution analysis operation on the set of distribution data such that the moment generating function produces a resultant moment bounded distribution data result that ensures that combinations of errors represented in the integrity estimates and corrections information in the set of distribution data do not exceed the integrity limit that accounts for all combined errors in the set of distribution data obtained form the signal-based navigation system.

5. The method of claim 4 wherein the signal-based navigation system is at least one of a wide area augmentation system, a local area augmentation system, and a global navigation satellite system.

6. The method of claim 5 wherein the corrections information and integrity estimate information included in the set of distribution data relate to sources of navigation error in the signal-based navigation system that includes at least one of satellite orbit error, satellite clock error, and satellite signal ionospheric delay error.

7. The method of claim 1 comprising the step of:

performing the steps of obtaining, applying and comparing during real-time operation of the data processing system; and

wherein the step of obtaining comprises the step of;

obtaining the at least one set of distribution data from real-time operation of the data processing system such that the steps of applying and comparing ensure that operation of the data processing system is within an acceptable integrity threshold.

8. The method of claim 1 comprising the steps of:

performing the step of obtaining at least one set of distribution data during a real-time operation of the data processing system to collect at least one sample set of distribution data; and

performing the steps of applying and comparing in non-real-time operation of the data processing system on the at least one sample set of distribution data to prove that the data processing system operates within an acceptable integrity threshold.

9. The method of claim 1 further comprising the steps of:

applying a Gaussian distribution function to the at least one set of distribution data to produce a Gaussian bounded distribution data result; and

comparing the Gaussian bounded distribution data result to an integrity limit to determine if the data processing system operates within an acceptable integrity threshold.

10. The method of claim 1 further comprising the steps of:

applying a overbounding distribution function to the at least one set of distribution data to produce an overbounded bounded distribution data result; and

comparing the overbounded bounded distribution data result to an integrity limit to determine if the data processing system operates within an acceptable integrity threshold.

11. The method of claim 1 wherein the step of obtaining at least one set of distribution data comprises the steps of:

operating the data processing system to produce a plurality of sets of distribution data, each having a respective associated integrity estimate;

and wherein the step of applying the moment generating function comprises the step of;

computing a respective moment bounded distribution data result based on the respective associated integrity estimate for each of the plurality of sets of distribution data by adjusting input values to the moment generating function such that the respective integrity estimate of each set of distribution data is moment-bounded;

and wherein the step of comparing the moment bounded distribution data result to an integrity limit comprises the steps of;

ensuring that each respective moment bounded distribution data result produces at least one value that is moment-bounded;

ensuring that each respective moment bounded distribution data result produces at least one value that is moment-bounded below the acceptable integrity threshold.

12. The method of claim 1 wherein the step of applying a moment generation function comprises the steps of:

selecting input values to a moment generating bound such that a first and a second condition are both satisfied, wherein;

i) the first condition is that all input distributions in the at least one set of distribution data are moment-bounded; and

ii) the second condition is that a root sum square of integrity estimates associated with each of the at least one set of distribution data are less than or equal to a probability of failure.

13. The method of claim 12 wherein the step of comparing comprises the step of:

determining that the data processing system operates within an acceptable integrity threshold if a first probability that a sum of the at least one set of distribution data is less than an integrity limit that accounts for combine errors within the at least one set of distribution data, is less than the probability of failure associated with the acceptable integrity threshold.

14. The method of claim 12 wherein the steps of applying a moment generation function and comparing the moment bounded distribution data result to an integrity limit collectively determine that:

15. The method of claim 14 wherein

16. The method of claim 14 wherein the step of applying a moment generating function to the at least one set of distribution data to produce moment bounded distribution data result comprises the step of:

applying a moment generating function of X, m_x(t) defined as;

m_x(t)=E(e<sup>t•

x</sup>)=∫

e<sup>t•

u</sup> p_x(u)du where P_x (u) is a probability density function of random variable X, such that X is (K, σ

²)-moment bounded if m_x(t) exists and is≦

K•

e<sup>t2σ

2/2</sup> for all real numbers t.

17. The method of claim 1 wherein:

there exists at least one source of error to the data processing system;

there are sets of distribution data, J_i,that represent input to the data processing system that are collected for each respective source of error to the data processing system;

each of the at least one source of error has a set of respective set of associated integrity values σ

_j, that represent an integrity limit associated with each J_i;

and wherein the step of applying a moment generation function to the at least one set of distribution data comprises the step of;

computing the moment generating function m_ji(t) of each set of distribution data and selecting input values K and α

to a moment generating bound such that;

i) all J_i are (K, (α

•

σ

²)-moment bounded; and

ii) such that 2K^Ne<sup>(KLimit/α

)2</sup>≦

P_fail, where;

K and α

are numerical input values to the moment generating bound;

K_Limit defines a size in a σ

units of a critical error;

e represents the base of natural logarithms; and

and P_fail is a value of the acceptable integrity threshold.

23. The method of claim 1 wherein the steps of applying a moment generating function and comparing the moment bounded distribution data result collectively perform a convolution analysis operation on the set of distribution data such that the moment generating function produces a resultant moment bounded distribution data result.

24. The method of claim 1, wherein applying the moment generating function includes:

applying a moment generating function, which is based on a corresponding probability density function, to the at least one set of distribution data to produce the moment bounded distribution data result.

25. The method of claim 1 wherein the step of obtaining at least one set of distribution data includes operating the data processing system to produce a plurality of sets of distribution data, each having a respective associated integrity estimate.

26. The method of claim 1, wherein applying the moment generating function includes:

computing a respective moment bounded distribution data result based on the respective associated integrity estimate for each of the plurality of sets of distribution data by adjusting input values to the moment generating function such that the respective integrity estimate of each set of distribution data is moment-bounded.

27. The method of claim 1, comparing the moment bounded distribution data result to an integrity limit includes:

ensuring that each respective moment bounded distribution data result produces at least one value that is moment-bounded; and

ensuring that each respective moment bounded distribution data result produces at least one value that is moment-bounded below the acceptable integrity threshold.

28. The method as in claim 1, wherein comparing the moment bound distribution data result includes comparing the moment bound distribution data result to an integrity limit maximum value that represents a maximum value of convolved errors in the at least one set of distribution data.

29. The method as in claim 1, wherein applying the moment generating function includes computing the moment bounded distribution data result based on a respective associated integrity estimate for the at least one set of distribution data by adjusting input values to the moment generating function such that the integrity estimate of the at least one set of distribution data is determined to be moment bounded.

30. The method of claim 1 further comprising:

initiating an integrity verification process that ensures that the moment bounded distribution data result produces at least one value that is moment bounded by selection of proper input values to the moment generating function.

31. The method of claim 1 wherein the step of comparing includes:

determining that the data processing system operates within an acceptable integrity threshold if a first probability that a sum of the at least one set of distribution data is less than an integrity limit that accounts for combined errors within the at least one set of distribution data, is less than the probability of failure associated with the acceptable integrity threshold.

18. A computer system comprising:

a communications interface;

a memory;

a processor; and

an interconnection mechanism coupling the communications interface, the memory and the processor;

wherein the memory is encoded with an integrity verification application that when performed on the processor, produces an integrity verification process that causes the computer system to evaluate operational integrity of a data processing system by performing the operations of;

obtaining at least one set of distribution data related to information processed by the data processing system;

applying a moment generating function to the at least one set of distribution data to produce a moment bounded distribution data result; and

comparing the moment bounded distribution data result to an integrity limit, representing a maximum value related to the collective set of errors represented within the at least one set of distribution data, to determine if the data processing system operates within an acceptable integrity threshold.

19. The computer system of claim 18 wherein:

the data processing system is a signal-based navigation system;

the at least one set of distribution data includes corrections information and integrity estimate information related to a location of a user of the signal-based navigation system;

wherein the steps of applying a moment generating function and comparing the moment bounded distribution data result collectively cause the processor to perform a convolution analysis operation on the set of distribution data such that the moment generating function produces a resultant moment bounded distribution data result that ensures that combinations of errors represented in the integrity estimates and corrections information in the set of distribution data do not exceed the integrity limit that accounts for all combine errors in the set of distribution data obtained form the signal-based navigation system.

20. The computer system of claim 18 wherein when the integrity verification process causes the computer system to perform the steps of applying a moment generation function and comparing the moment bounded distribution data result to an integrity limit, the integrity verification process causes the computer system to determine that:

21. A computer system comprising:

a communications interface;

a memory;

a processor; and

an interconnection mechanism coupling the communications interface, the memory and the processor;

wherein the memory is encoded with an integrity verification application that when performed on the processor, produces an integrity verification process that provides a means for the computer system to evaluate operational integrity of a data processing system, such means including;

means for obtaining at least one set of distribution data related to information processed by the data processing system;

means for applying a moment generating function to the at least one set of distribution data to produce a moment bounded distribution data result; and

means for comparing the moment bounded distribution data result to an integrity limit to determine if the data processing system operates within an acceptable integrity threshold.

22. A computer program product having a computer-readable medium including computer program logic encoded thereon that, when performed on a computer system having a coupling of a memory and a processor provides a method for the computer system to evaluate operational integrity of a data processing system by performing the operations of:

obtaining at least one set of distribution data related to information processed by the data processing system;

applying a moment generating function to the at least one set of distribution data to produce a moment bounded distribution data result; and

comparing the moment bounded distribution data result to an integrity limit to determine if the data processing system operates within an acceptable integrity threshold.

32. The computer system of 18, wherein the moment generating function is a mathematical function based at least in part on a corresponding probability distribution function and the computer system further supports operations of:

performing the steps of applying and comparing on the at least one sample set of distribution data to prove that the data processing system operates within an acceptable integrity threshold.