Neuronal sensor networks

• US 7,429,923 B2
• Filed: 10/22/2004
• Issued: 09/30/2008
• Est. Priority Date: 10/22/2004
• Status: Expired due to Fees

• Current Assignee: HENEYWELL INTERNATIONAL INC.
• Original Assignee: Honeywell International Inc.
• Inventors: McKitterick, John B., Burne, Richard A.
• Primary Examiner(s): Swarthout; Brent
• Application Number:
  US10/970,933
• Publication Number:
  US 20060097862A1
• Time in Patent Office: 1,439 Days
• Field of Search: 340/540, 340/541, 340/545.4, 340/545.5, 340/550, 340/552, 340/561, 340/562, 340/565, 340/567, 340/949, 340/968, 250/341, 256/338.1, 119/708, 119/908
• US Class Current: 340/541
• CPC Class Codes: G06N 3/02 Neural networks
• CPC Class Codes: H04L 67/12 specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]

Abstract

A method for detecting significant events using neuronal sensor networks is provided. The method includes monitoring an environment surrounding a plurality of sensors for the presence of an event, when one or more events are detected, integrating the detected events over space and time using one or more collectors responsive to the plurality of sensors, determining when the one or more events are significant events and identifying and tracking significant events using processors responsive to the one or more collectors.

Claims

1. A system for detecting events using neuronal sensor networks, the system comprising:

a plurality of sensors that produce event detection signals when an event is detected and exceeds a minimum event threshold level, wherein each of the plurality of sensors comprises a respective controller that determines when an event exceeds a minimum event threshold level, and a respective transmitter that is coupled to the respective controller and that transmits the event detection signals according to a simple communication protocol, wherein each of the plurality of sensors has a source of power derived from the event;

one or more collectors adapted to receive one or more of the event detection signals from an associated subset of the plurality of sensors that are in close proximity to the one or more collectors, wherein each of the one or more collectors integrates the received event detection signals transmitted from the associated subset of the plurality of sensors over space and time to produce a stored summation, wherein the one or more collectors produce a threshold detection signal when a predetermined collector threshold level is exceeded by the stored summation, and wherein each sensor of the plurality of sensors communicates directly with at least one collector; and

one or more processors, adapted to receive threshold detection signals from the one or more collectors.

2. The system of claim 1, wherein the sensors are deployed over a wide area.

3. The system of claim 1, wherein each event detection signal provides the location of the event and strength of the event above a minimum event threshold level.

4. The system of claim 1, wherein each of the plurality of sensors runs on ultra low power.

5. The system of claim 1, wherein each of the one or more collectors transmits a threshold detection signal when a predetermined collector threshold level is exceeded by the stored summation.

6. The system of claim 1, wherein each of the one or more processors uses threshold detection signals to track and identify significant events.

22. The system of claim 1, wherein the detected event is a target stepping on one or more of the plurality of sensors.

7. A method for detecting significant events using neuronal sensor networks, the method comprising:

monitoring an environment surrounding a plurality of sensors for the presence of an event, wherein monitoring the environment surrounding a plurality of sensors for the presence of an event comprises (i) monitoring the surrounding environment for an event, (ii) determining whether the strength of the event exceeds a minimum event threshold level, and (iii) transmitting an event detection signal from one or more sensors to one or more associated collectors and nearby sensors, wherein each of the plurality of sensors has a source of power derived from the event, and wherein the event detection signals are ultra short-range detection transmissions;

when one or more events are detected, integrating the detected events over space and time to produce a stored summation using one or more collectors responsive to each collector'"'"'s associated subset of the plurality of sensors that is in close proximity to the associated collector;

determining when the one or more events are significant events, wherein a significant event occurs when the stored summation exceeds the collector'"'"'s minimum threshold level; and

identifying and tracking significant events using one or more processors responsive to the one or more collectors.

8. The method of claim 7, wherein the sensor lowers its minimum event threshold level upon receiving an event detection signal from a nearby sensor.

9. The method of claim 8, wherein the sensor lowers its minimum event threshold level for a set period of time.

10. A system for detecting events using neuronal sensor networks, the system comprising:

a plurality of sensors that produce and receive event detection signals when an event is detected, wherein each of the plurality of sensors comprises a respective controller that detects when an event exceeds a minimum event threshold level, a respective transmitter that is coupled to the respective controller and that transmits the event detection signals according to a simple communication protocol, and a receiver that receives event detection signals transmitted by nearby sensors, and wherein each of the plurality of sensors has a source of power derived from the event;

one or more collectors wherein each collector receives event detection signals from an associated subset of the plurality of sensors that are in close proximity to the one or more collectors, wherein each of the one or more collectors integrates the received event detection signals transmitted from the associated subset of the plurality of sensors over space and time to produce a stored summation, wherein the one or more collectors produce a threshold detection signal when a predetermined collector threshold level is exceeded by the stored summation, and wherein each sensor of the plurality of sensors communicates directly with at least one collector; and

one or more processors adapted to receive threshold detection signals from the one or more collectors.

11. The system of claim 10, wherein the plurality of sensors are deployed over a wide area.

12. The system of claim 10, wherein the event detection signal provides the location of the sensor event and the strength of the event above a minimum event threshold level.

13. The system of claim 10, wherein each of the plurality of sensors will lower its minimum event threshold level when it receives event detection signals from nearby sensors.

14. The system of claim 10, wherein each of the plurality of sensors runs on ultra low power.

15. The system of claim 10, wherein each of the plurality of sensors runs on power obtained from the environment.

16. The system of claim 10, wherein the detected event is a target stepping on one or more of the plurality of sensors.

17. The system of claim 10, wherein each of the one or more collectors transmits a threshold detection signal when a predetermined collector threshold level is exceeded by the stored summation.

18. The system of claim 10, wherein each of the one or more processors uses threshold detection signals to track and identify significant events.

21. The system of claim 10, wherein each of the plurality of sensors will lower its minimum event threshold level for a set period of time when it receives event detection signals from nearby sensors.

19. A method for detecting events using neuronal sensors, the method comprising:

monitoring the surrounding environment for an event;

determining whether the strength of the event exceeds the minimum event threshold level; and

transmitting an event detection signal from one or more sensors to one or more associated collectors and nearby sensors when the strength of the event exceeds the minimum event threshold level, wherein the event detection signal comprises a series of ultra-short pulses, and wherein the number of pulses in the series of ultra-short pulses corresponds to the strength of the event, wherein each of the plurality of sensors has an independent source of power and runs on ultra low power, and wherein a given sensor lowers its minimum event threshold level when it receives an event detection signal from a nearby sensor.

20. The method of claim 19, wherein the given sensor lowers its minimum event threshold level for a set period of time.