Feature Extraction & Data Compression System & Method For Distributed Sensor Networks

• US 20110281602A1
• Filed: 05/12/2010
• Published: 11/17/2011
• Est. Priority Date: 05/12/2010
• Status: Active Grant

• Current Assignee: Information Sports Technologies Incorporated
• Original Assignee: Information Sports Technologies Incorporated
• Inventors: McCarron, Michael V., Azimi-Sadjadi, Mahmood R., Srinivasan, SaravanaKumar
• Granted Patent: US 8,478,319 B2
• Time in Patent Office: Days
• US Class Current: 455/507
• CPC Class Codes: H04B 17/27 for locating or positioning the transmitter
• CPC Class Codes: H04B 17/309 Measuring or estimating channel quality parameters
• CPC Class Codes: H04W 84/18 Self-organising networks, e.g. ad-hoc networks or sensor networks

Abstract

A distributed sensor network has a base station and clusters of sensor nodes. In a method of locating and classifying signal sources, at each node divides a received signal into blocks, performs Fourier-based transform on the blocks, selects peaks from the transformed blocks, selects subbands with features of interest based on the frequency of occurrence of the peaks across the blocks, collaborates with other nodes in the cluster to make a final selection of the subbands, encodes the subband features of the signal, and transmits the subband features to the base station. The base station processes the received subband features to locate and classify the signal sources.

Claims

1. A distributed sensor network for locating and classifying signal sources comprising:

a base station, and

at least one cluster having a plurality of sensor nodes with each sensor node including;

a sensor,

means, connected to receive a signal from said sensor, for dividing a selected time increment of said signal into blocks,

means for performing a first Fourier-based transform on said blocks,

means for selecting a plurality of peaks in each said block,

means for selecting a plurality of subbands of said signal based on the frequency of occurrence of said peaks in said blocks,

means for performing a second Fourier-based transform on said time increment of said signal, and

means for transmitting said subbands of said time increment of said signal to said base station,

whereby said subbands include features relevant to a source of said signal and said base station locates and classifies said source by processing said subbands received from said sensor nodes.

2. The network as set forth in claim 1 wherein said sensor nodes each include means for collaborating with other sensor nodes in said cluster to select said subbands.

3. The network as set forth in claim 2 wherein said means for transmitting and said means for collaborating together include at least one wireless transceiver device.

4. The network as set forth in claim 1 wherein said sensor nodes each include an analog to digital converter for converting said time increment of said signal into a digital signal.

5. The network as set forth in claim 1 wherein said first and second Fourier-based transforms are Discrete Cosine Transforms.

6. The network as set forth in claim 1 wherein each sensor node includes a means for encoding said subbands before transmitting said subbands.

7. The network as set forth in claim 1 wherein each sensor node includes a field programmable gate array and memory.

8. The network as set forth in claim 7 wherein said means for selecting a plurality of peaks and said means for selecting a plurality of subbands include software instructions implemented in said field programmable gate array.

9. A distributed sensor network for locating and classifying signal sources comprising:

a base station, and

at least one cluster having a plurality of sensor nodes with each sensor node including;

a sensor for receiving a signal,

an analog to digital converter for converting a time increment of said signal into a digital signal,

a field programmable gate array connected to said analog to digital converter, and programmed to divide said digital signal into blocks, perform a Discrete Cosine Transform on said blocks, select a plurality of peaks in each said block, select a plurality of subbands of said signal based on the frequency of occurrence of said peaks in said blocks, perform a Discrete Cosine Transform on said digital signal, and

a wireless transceiver for collaborating, in cooperation with said field programmable gate array, with other sensor nodes in said cluster to select said subbands, and for transmitting said subbands of said digital signal to said base station,

whereby said subbands include features relevant to a source of said signal and said base station locates and classifies said source by processing said subbands received from said sensor nodes.

10. A method of feature extraction and data reduction of a signal received by a sensor node in a cluster of sensor nodes in a sensor network with said sensor network including a base station, comprising the steps of:

dividing a selected time increment of said signal into blocks,

performing a first Fourier-based transform on said blocks,

selecting a plurality of peaks in each said block,

selecting a plurality of subbands of said signal based on the frequency of occurrence of said peaks in said blocks,

performing a second Fourier-based transform on said signal for said time increment, and

transmitting said subbands of said signal for said time increment to said base station,

whereby said subbands include features relevant to a source of said signal and said subbands reduce data relative to said signal for said time increment.

11. The method as set forth in claim 10 including the step of converting said signal for said time increment to a digital signal before said step of dividing.

12. The method as set forth in claim 10 wherein said first and second Fourier-based transforms are Discrete Cosine Transforms.

13. The method as set forth in claim 10 wherein said subbands include about ten percent of said signal for said time increment.

14. The method as set forth in claim 10 wherein in said step of selecting a plurality of peaks, five peaks are selected in each block.

15. The method as set forth in claim 10 wherein said step of selecting a plurality of subbands includes computing a histogram of said peaks.

16. The method as set forth in claim 10 wherein in said step of selecting a plurality of subbands, three subbands are selected.

17. The method as set forth in claim 10 including the step of collaborating with other sensor nodes to select said subbands after said step of selecting a plurality of subbands.

18. The method as set forth in claim 10 including the step of encoding said subbands of said signal for said time increment before transmitting said subbands of said signal for said time increment to said base station.

19. A method of feature extraction and data reduction of a signal received by a sensor node in a cluster of sensor nodes in a sensor network with said sensor network including a base station, comprising the steps of:

dividing a selected time increment of said signal into blocks,

performing a Discrete Cosine Transform on said blocks,

selecting five peaks in each said block,

computing a histogram of said peaks,

selecting three subbands of said signal based on the frequency of occurrence of said peaks in said blocks,

collaborating with other sensor nodes in said cluster to select said subbands,

performing a Discrete Cosine Transform on said signal for said time increment,

encoding said subbands of said signal for said time increment, and

transmitting said subbands of said signal for said time increment to said base station,

whereby said subbands include features relevant to a source of said signal and said subbands reduce data relative to said signal for said time increment.