Detecting sensor malfunctions using compression analysis of binary decision diagrams

• US 9,176,819 B2
• Filed: 09/23/2011
• Issued: 11/03/2015
• Est. Priority Date: 09/23/2011
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Stergiou, Stergios, Jain, Jawahar, Nakata, Tsuneo
• Primary Examiner(s): Kundu, Sujoy
• Assistant Examiner(s): Ngon, Ricky
• Application Number:
  US13/242,253
• Publication Number:
  US 20130080843A1
• Time in Patent Office: 1,502 Days
• Field of Search: 702/116, 714/49
• US Class Current: 1/1
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G06F 11/0736 in functional embedded systems, i.e. in a data processing system designed as a combination of hardware and software dedicated to performing a certain function testing or monitoring of automated control systems G05B23/02
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/14 Error detection or correction of the data by redundancy in operation G06F11/16 takes precedence

Abstract

In particular embodiments, a method includes analyzing a binary decision diagram (BDD) representing a data stream from a sensor to determine a compression rate of the BDD and indicating a sensor malfunction in the sensor if the compression rate of the BDD deviates from a specified compression rate range.

Claims

1. A method for detecting malfunctions in medical sensors comprising:

by one or more computing devices,

accessing, from a first medical sensor, a first binary decision diagram (BDD) representing medical sensor data from the first medical sensor, the medical sensor data measuring physiological stimulus from a person'"'"'s body;

determining a first compression rate of the first BDD accessed from the first medical Sensor;

indicating a malfunction in connection with the first medical sensor if the first compression rate is outside of a predetermined compression-rate range; and

size optimizing storage of the medical sensor data by;

if a malfunction is not indicated, then storing the first BDD in a data store associated with the first medical sensor; and

if a malfunction is indicated, then deleting the first BDD from the data store associated with the first medical sensors.

2. The method of claim 1, wherein the predetermined compression-rate range has a lower bound equal to a predetermined constant.

3. The method of claim 1, wherein the predetermined compression-rate range has an upper bound equal to a predetermined constant.

4. The method of claim 1, wherein the predetermined compression-rate range has a lower bound approximately equal to 2×

compression rate.

5. The method of claim 1, wherein the predetermined compression-rate range has an upper bound approximately equal to 100×

compression rate.

6. The method of claim 1, wherein the predetermined compression-rate range has a lower bound approximately equal to a compression rate achieved when compressing a BDD that represents random noise.

7. The method of claim 1, wherein the predetermined compression-rate range has an upper bound approximately equal to a compression rate achieved when compressing a BDD that represents a single continuously repeating number.

8. The method of claim 1, wherein indicating the malfunction in connection with the first medical sensor comprises indicating that the medical sensor data from the first medical sensor are invalid.

9. The method of claim 1, wherein indicating the malfunction in connection with the first medical sensor comprises indicating that the first medical sensor receives invalid stimulus.

10. The method of claim 1, wherein indicating the malfunction in connection with the first medical sensor comprises indicating a software error in the first medical sensor or in a sensor network comprising first medical sensor.

11. The method of claim 1, wherein indicating the malfunction in connection with the first medical sensor comprises indicating a hardware error in the first medical sensor.

12. The method of claim 1, further comprising if the first compression rate is within the predetermined compression-rate range, then indicating that the medical sensor data from the first medical sensor represented by the first BDD are valid.

13. The method of claim 1, further comprising:

determining a second compression rate of a second BDD, wherein the second BDD represents medical sensor data from a second medical sensor and has been compressed; and

if the second compression rate is outside of the predetermined compression-rate range, then negating the indication of the malfunction in connection with the first medical sensor.

14. The method of claim 1, further comprising:

constructing the first BDD for representing the medical sensor data from the first medical sensor;

initializing the first BDD to be empty; and

iteratively receiving the medical sensor data from the first medical sensor and adding the medical sensor data to the first BDD as they are received.

15. The method of claim 1, further comprising:

if the first compression rate is outside of the predetermined compression-rate range, then deleting the first BDD; and

if the first compression rate is within the predetermined compression-rate range, then storing the first BDD.

16. The method of claim 1, wherein the first medical sensor is affixed to a person'"'"'s body.

17. An apparatus for detecting malfunctions in medical sensors comprising:

one or more processors; and

a memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to;

access, from a first medical sensor, a first binary decision diagram (BDD) representing medical sensor data from the first medical sensor, the medical sensor data measuring physiological stimulus from a person'"'"'s body;

determine a first compression rate of the first BDD accessed from the first medical sensor;

indicate a malfunction in connection with the first medical sensor if the first compression rate is outside of a predetermined compression-rate range; and

size optimize storage of the medical sensor data by;

if a malfunction is not indicated, then store the first BDD in a data store associated with the first medical sensor; and

if a malfunction is indicated, then delete the first BDD from the data store associated with the first medical sensors.

18. The apparatus of claim 17, wherein the predetermined compression-rate range has a lower bound equal to a predetermined constant.

19. The apparatus of claim 17, wherein the predetermined compression-rate range has an upper bound equal to a predetermined constant.

20. The apparatus of claim 17, wherein the predetermined compression-rate range has a lower bound approximately equal to 2×

compression rate.

21. The apparatus of claim 17, wherein the predetermined compression-rate range has an upper bound approximately equal to 100×

compression rate.

22. The apparatus of claim 17, wherein the predetermined compression-rate range has a lower bound approximately equal to a compression rate achieved when compressing a BDD that represents random noise.

23. The apparatus of claim 17, wherein the predetermined compression-rate range has an upper bound approximately equal to a compression rate achieved when compressing a BDD that represents a single continuously repeating number.

24. The apparatus of claim 17, wherein indicating the malfunction in connection with the first medical sensor comprises indicating that the medical sensor data from the first medical sensor are invalid.

25. The apparatus of claim 17, wherein indicating the malfunction in connection with the first medical sensor comprises indicating that the first medical sensor receives invalid stimulus.

26. The apparatus of claim 17, wherein indicating the malfunction in connection with the first medical sensor comprises indicating a software error in the first medical sensor or in a sensor network comprising the first medical sensor.

27. The apparatus of claim 17, wherein indicating the malfunction in connection with the first medical sensor comprises indicating a hardware error in the first medical sensor.

28. The apparatus of claim 17, wherein the processors are further operable when executing the instructions to, if the first compression rate is within the predetermined compression-rate range, then indicate that the medical sensor data from the first medical sensor represented by the first BDD are valid.

29. The apparatus of claim 17, wherein the processors are further operable when executing the instructions to:

determine a second compression rate of a second BDD, wherein the second BDD represents medical sensor data from a second medical sensor and has been compressed; and

if the second compression rate is outside of the predetermined compression-rate range, then negate the indication of the malfunction in connection with the first medical sensor.

30. The apparatus of claim 17, wherein the processors are further operable when executing the instructions to:

construct the first BDD for representing the medical sensor data from the first medical sensor;

initialize the first BDD to be empty; and

iteratively receive the medical sensor data from the first medical sensor and adding the medical sensor data to the first BDD as they are received.

31. The apparatus of claim 17, wherein the processors are further operable when executing the instructions to:

if the first compression rate is outside of the predetermined compression-rate range, then delete the first BDD; and

if the first compression rate is within the predetermined compression-rate range, then store the first BDD.

32. The apparatus of claim 17, wherein the first medical sensor is affixed to a person'"'"'s body.

33. One or more computer-readable non-transitory storage media embodying software instructions for detecting malfunctions in medical sensors that is operable for causing a computer processor when executed to:

access, from a first medical sensor, a first binary decision diagram (BDD) representing medical sensor data from the first medical sensor, the medical sensor data measuring physiological stimulus from a person'"'"'s body;

determine a first compression rate of the first BDD accessed from the first medical sensor;

indicate a malfunction in connection with the first medical sensor if the first compression rate is outside of a predetermined compression-rate range; and

size optimize storage of the medical sensor data by;

if a malfunction is not indicated, then store the first BDD in a data store associated with the first medical sensor; and

if a malfunction is indicated, then delete the first BDD from the data store associated with the first medical sensors.

34. The media of claim 33, wherein the predetermined compression-rate range has a lower bound equal to a predetermined constant.

35. The media of claim 33, wherein the predetermined compression-rate range has an upper bound equal to a predetermined constant.

36. The media of claim 33, wherein the predetermined compression-rate range has a lower bound approximately equal to 2×

compression rate.

37. The media of claim 33, wherein the predetermined compression-rate range has an upper bound approximately equal to 100 ×

compression rate.

38. The media of claim 33, wherein the predetermined compression-rate range has a lower bound approximately equal to a compression rate achieved when compressing a BDD that represents random noise.

39. The media of claim 33, wherein the predetermined compression-rate range has an upper bound approximately equal to a compression rate achieved when compressing a BDD that represents a single continuously repeating number.

40. The media of claim 33, wherein indicating the malfunction in connection with the first medical sensor comprises indicating that the medical sensor data from the first medical sensor are invalid.

41. The media of claim 33, wherein indicating the malfunction in connection with the first medical sensor comprises indicating that the first medical sensor receives invalid stimulus.

42. The media of claim 33, wherein indicating the malfunction in connection with the first medical sensor comprises indicating a software error in the first medical sensor or in a sensor network comprising the first medical sensor.

43. The media of claim 33, wherein indicating the malfunction in connection with the first medical sensor comprises indicating a hardware error in the first medical sensor.

44. The media of claim 33, wherein the software is further operable when executed to, if the first compression rate is within the predetermined compression-rate range, then indicate that the medical sensor data from the first medical sensor represented by the first BDD are valid.

45. The media of claim 33, wherein the software is further operable when executed to:

determine a second compression rate of a second BDD, wherein the second BDD represents medical sensor data from a second medical sensor and has been compressed; and

if the second compression rate is outside of the predetermined compression-rate range, then negate the indication of the malfunction in connection with the first medical sensor.

46. The media of claim 33, wherein the software is further operable when executed to:

construct the first BDD for representing the medical sensor data from the first medical sensor;

initialize the first BDD to be empty; and

iteratively receive the medical sensor data from the first medical sensor and adding the medical sensor data to the first BDD as they are received.

47. The media of claim 33, wherein the software is further operable when executed to:

if the first compression rate is outside of the predetermined compression-rate range, then delete the first BDD; and

if the first compression rate is within the predetermined compression-rate range, then store the first BDD.

48. The media of claim 33, wherein the first medical sensor is affixed to a person'"'"'s body.

49. A system for detecting malfunctions in medical sensors comprising:

means for accessing, from a first medical sensor, a first binary decision diagram (BDD) representing medical sensor data from the first medical sensor, the medical sensor data measuring physiological stimulus from a person'"'"'s body;

means for determining a first compression rate of the first BDD accessed from the first medical sensor;

means for indicating a malfunction in connection with the first medical sensor if the first compression rate is outside of a predetermined compression-rate range; and

means for size optimizing storage of the medical sensor data by;

if a malfunction is not indicated, then means for storing the first BDD in a data store associated with the first medical sensor; and

if a malfunction is indicated, then means for deleting the first BDD from the data store associated with the first medical sensors.

50. A method for detecting malfunctions in medical sensors comprising:

by one or more computing devices,

accessing, from a first medical sensor, a binary decision diagram (BDD) representing medical sensor data from the first medical sensor, the medical sensor data measuring physiological stimulus from a person'"'"'s body;

determining a compression change rate of the BDD accessed from the first medical sensor;

indicating a sensor malfunction in the medical sensor if the compression change rate is outside of a predetermined compression-change-rate range; and

size optimizing storage of the medical sensor data by;

if a malfunction is not indicated, then storing the first BDD in a data store associated with the first medical sensor; and

if a malfunction is indicated, then deleting the first BDD from the data store associated with the first medical sensors.