Method and apparatus for random-number generator

• US 7,472,148 B2
• Filed: 07/23/2004
• Issued: 12/30/2008
• Est. Priority Date: 07/23/2004
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Simon, Harris S., Van Pelt, Kenneth Andrew, Sharp, Dale Ogden
• Primary Examiner(s): Do; Chat C
• Application Number:
  US10/897,589
• Publication Number:
  US 20060020647A1
• Time in Patent Office: 1,621 Days
• Field of Search: 708250-256
• US Class Current: 708/250
• CPC Class Codes: G06F 7/58 Random or pseudo-random number generators

Abstract

Method and apparatus for generating random numbers are disclosed. In one aspect, a method for generating random numbers for use in a wireless communication device provides for generating random numbers, gathering a sample of the generated random numbers, and computing at least one metric, such as mean value, standard deviation, and/or entropy. The method further provides for comparing the metric with a corresponding reference value and adjusting the metric based on a result of said comparison so that the generated random numbers achieve a desired distribution. In another aspect, an apparatus for generating random numbers includes an analog noise generator and hardware components for generating random numbers and feedback values to adjust the random numbers. The apparatus further includes a processor capable of executing instructions to carry out control algorithms for adjusting the random numbers.

Claims

1. A method for generating random numbers for use in a wireless communication device, the method comprising:

generating random numbers having an adjustable distribution based on at least one adjustable input value;

gathering a sample of the generated random numbers;

computing at least one metric based on the sample;

comparing the metric with a corresponding reference value; and

adjusting the adjustable input value based on a result of said comparing so that the generated random numbers achieve a desired distribution;

wherein adjusting the adjustable input value based on said comparing comprises;

adjusting a dc offset value for generating an analog noise voltage to cause the generated random numbers to achieve a desired numeric mean; and

adjusting a reference voltage value to cause the generated random numbers to achieve a desired numeric range.

2. The method of claim 1, wherein said metric includes a mean value.

3. The method of claim 1, wherein said metric includes a standard deviation value.

4. The method of claim 1, wherein said metric includes an entropy value.

5. The method of claim 1, wherein said adjusting includes adjusting the adjustable input value through a linear algorithm.

6. The method of claim 1, wherein said adjusting includes adjusting the adjustable input value through a nonlinear algorithm.

7. The method of claim 1, wherein said adjusting includes adjusting the adjustable input value through an adaptive algorithm.

8. The method of claim 1, wherein computing at least one metric based on the sample comprises:

computing a first metric representing a mean value of the random sample; and

computing a second metric representing a standard deviation of the random sample.

9. The method of claim 1, wherein the desired distribution is a Gaussian distribution having a mean value corresponding to a center value of a range for the random numbers.

10. An apparatus for generating random numbers in a wireless communication device, comprising:

means for generating random numbers having an adjustable distribution based on at least one adjustable input value;

means for gathering a sample of the generated random numbers;

means for computing at least one metric based on the sample;

means for comparing the metric with a corresponding reference value; and

means for adjusting the adjustable input value based on a result of said comparing so that the generated random numbers achieves a desired distribution;

wherein the means for adjusting the adjustable input value based on said comparing comprises;

means for adjusting a dc offset value for generating an analog noise voltage to cause the generated random numbers to achieve a desired numeric mean; and

means for adjusting a reference voltage value to cause the generated random numbers to achieve a desired numeric range.

11. The apparatus of claim 10, wherein said metric includes a mean value.

12. The apparatus of claim 10, wherein said metric includes a standard deviation value.

13. The apparatus of claim 10, wherein said metric includes an entropy value.

14. The apparatus of claim 10, wherein said means for adjusting includes means for adjusting the adjustable input value through a linear algorithm.

15. The apparatus of claim 10, wherein said means for adjusting includes means for adjusting the adjustable input value through a nonlinear algorithm.

16. The apparatus of claim 10, wherein said means for adjusting includes means for adjusting the adjustable input value through an adaptive algorithm.

17. The apparatus of claim 10, wherein the means for computing at least one metric based on the sample comprises:

means for computing a first metric representing a mean value of the random sample; and

means for computing a second metric representing a standard deviation of the random sample.

18. The apparatus of claim 10, wherein the desired distribution is a Gaussian distribution having a mean value corresponding to a center value of a range for the random numbers.

19. A computer-readable medium embodying means for implementing a method for generating random numbers in a wireless communication device, the method comprising:

computing at least one metric based on a sample of randomly generated numbers having an adjustable distribution based on at least one adjustable input value;

comparing the metric with a corresponding reference value; and

adjusting the adjustable input value based on a result of said comparing so that the generated random numbers achieves a desired distribution;

wherein adjusting the adjustable input value based on said comparing comprises;

adjusting a dc offset value for generating an analog noise voltage to cause the generated random numbers to achieve a desired numeric mean; and

adjusting a reference voltage value to cause the generated random numbers to achieve a desired numeric range.

20. The medium of claim 19, wherein said metric includes a mean value.

21. The medium of claim 19, wherein said metric includes a standard deviation value.

22. The medium of claim 19, wherein said metric includes an entropy value.

23. The medium of claim 19, wherein said adjusting includes adjusting the adjustable input value through a linear algorithm.

24. The medium of claim 19, wherein said adjusting includes adjusting the adjustable input value through a nonlinear algorithm.

25. The medium of claim 19, wherein said adjusting includes adjusting the adjustable input value through an adaptive algorithm.

26. The medium of claim 19, wherein computing at least one metric based on the sample comprises:

computing a first metric representing a mean value of the random sample; and

computing a second metric representing a standard deviation of the random sample.

27. The medium of claim 19, wherein the desired distribution is a Gaussian distribution having a mean value corresponding to a center value of a range for the random numbers.

28. A processor for implementing a method for adjusting randomly generated numbers, the method comprising:

computing at least one metric based on a sample of randomly generated numbers having an adjustable distribution based on at least one adjustable input value;

comparing the metric with a corresponding reference value; and

adjusting the adjustable input value based on a result of said comparing so that the generated random numbers achieves a desired distribution;

wherein adjusting the adjustable input value based on said comparing comprises;

adjusting a dc offset value for generating an analog noise voltage to cause the generated random numbers to achieve a desired numeric mean; and

adjusting a reference voltage value to cause the generated random numbers to achieve a desired numeric range.

29. The processor of claim 28, wherein said metric includes a mean value.

30. The processor of claim 28, wherein said metric includes a standard deviation value.

31. The processor of claim 28, wherein said metric includes an entropy value.

32. The processor of claim 28, wherein said adjusting includes adjusting the adjustable input value through a linear algorithm.

33. The processor of claim 28, wherein said adjusting includes adjusting the adjustable input value through a nonlinear algorithm.

34. The processor of claim 28, wherein said adjusting includes adjusting the adjustable input value through an adaptive algorithm.

35. The processor of claim 28, wherein computing at least one metric based on the sample comprises:

computing a first metric representing a mean value of the random sample; and

computing a second metric representing a standard deviation of the random sample.

36. The processor of claim 28, wherein the desired distribution is a Gaussian distribution having a mean value corresponding to a center value of a range for the random numbers.