System and method for power effective participatory sensing

• US 9,769,305 B2
• Filed: 03/13/2014
• Issued: 09/19/2017
• Est. Priority Date: 04/01/2013
• Status: Expired due to Fees

• Current Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Original Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Inventors: Banerjee, Rohan, Sinha, Aniruddha, Saha, Arindam
• Primary Examiner(s): Acolatse, Kodzovi
• Application Number:
  US14/772,928
• Publication Number:
  US 20160021237A1
• Time in Patent Office: 1,286 Days
• Field of Search: None
• CPC Class Codes: H04M 1/72454 according to context-related or environment-related conditions
• CPC Class Codes: H04W 4/02 Services making use of location information
• CPC Class Codes: H04W 4/024 Guidance services
• CPC Class Codes: H04W 4/027 using movement velocity, acceleration information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 4/38 for collecting sensor information
• CPC Class Codes: H04W 4/44 for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
• CPC Class Codes: H04W 4/50 Service provisioning or reconfiguring
• CPC Class Codes: H04W 52/0254 detecting a user operation or a tactile contact or a motion of the device
• CPC Class Codes: Y02D 30/70 in wireless communication networks

Abstract

Disclosed is a method and system enabling power effective participatory sensing. The hand held device of the system is equipped with plurality of sensors, and is configured to enable the power effective sensor to monitor operation of the power intensive sensors. In one embodiment, a participatory sensing approach is used for traffic condition. A methodology for triggering power hungry sensors (audio) with the help of low power sensors (accelerometer) is presented which is able to reduce the overall power consumption of the mobile device. Further, a decision tree based approach is used to classify the level of congestion by measuring the horn density in a particular location.

Claims

1. A computer implemented method for power effective participatory sensing, the method comprising:

enabling at least one power effective sensor to monitor operation of one or more power intensive sensors, wherein the power effective sensor is in an active state to capture and analyze contextual information related to an object area in response to receiving analyzed contextual information from the power effective sensor;

triggering the one or more power intensive sensors, based on the analyzed contextual information, to initiate audio recording and extracting at least one feature from the audio recording, and concurrently generating a corresponding metadata file, wherein the metadata file comprises information related to the power effective sensor and the contextual information; and

transmitting the at least one feature and the metadata file for analysis to a backend server, wherein said at least one feature and the metadata file are analyzed for tracking the object area condition.

2. The method of claim 1, wherein the contextual information comprises of location based information, timestamp information, traffic conditions, weather conditions, area estimations and the like.

3. The method of claim 2, wherein the location based information is captured using received signal strength indicator (RSSI) signals sensed from one or more hand held devices present within or in close proximity of the object area.

4. The method of claim 1, wherein the analysis of the contextual information is based on vehicular static or dynamic state interpreted from accelerometer data, and vehicle honking rate.

5. The method of claim 4, wherein variance of the accelerometer data is computed and compared against a threshold limit prior to trigger the power intensive sensors into an active state.

6. The method of claim 1, wherein the audio feature extraction is accomplished using a modified Mel Frequency Cepstral Coefficient (MFCC) obtained by combining the conventional Mel filter structure and reverse structure of conventional Mel filter.

7. The method of claim 1, further comprising compressing the features and corresponding metadata at the hand held device and decompressing the compressed features and the corresponding metadata at the backend server.

8. The method of claim 1, wherein the analysis at the backend server involves, mapping the features and the corresponding metadata file with one or more pre-stored training models of the backend server.

9. The method of claim 8, wherein the training models are created using a Gaussian Mixture Model classifiers based on audio analysis of horn and non horn sound detected from the object area.

10. The method of claim 9, wherein the non horn sound includes non vehicular noise sound emanating from engine idle noise, tire noise, air turbulence noise, human speech, vehicle music noise and the like.

11. The method of claim 1, wherein the object area condition refers to traffic condition thereof.

12. A participatory sensing system, comprising:

a hand held device equipped with plurality of sensors, and configured to enable at least one power effective sensor to monitor operation of one or more power intensive sensors, wherein the power effective sensor is in an active state to capture and analyze contextual information related to an object area in response to receiving analyzed contextual information from the power effective sensor;

trigger the one or more power intensive sensors, based on the analyzed contextual information, to initiate audio recording and extract at least one feature from the audio recording and concurrently generate a corresponding metadata file, wherein the metadata file comprises information related to the power effective sensor and the contextual information; and

a backend server loaded with one or more training models that are mapped against the at least one features and the metadata file received from the hand held device to track and monitor the object area condition.

13. The participatory sensing system of claim 12, wherein the power effective sensors include an accelerometer, orientation sensor and the like, or a combination thereof.

14. The participatory sensing system of claim 12, wherein the power intensive sensors include microphones;

location providing sensors such as Global Positioning System (GPS);

gyroscope and the like, or a combination thereof.

15. The participatory sensing system of claim 12, wherein the contextual information comprises of location based information, timestamp information, traffic conditions, weather conditions, area estimations and the like.

16. The participatory sensing system of claim 12, wherein the power effective sensor analyses the contextual information based on vehicular static or dynamic state interpreted from an accelerometer data, and vehicle-honking rate.

17. The participatory sensing system of claim 12, wherein the power intensive sensors accomplishes feature extraction by using a modified Mel Frequency Cepstral Coefficient (MFCC) obtained by combining conventional Mel filter structure and reverse structure of conventional Mel filter.

18. The participatory sensing system of claim 12, wherein the hand held device is further configured to compress the features and corresponding metadata, while the backend server is further configured to decompress the compressed features and the corresponding metadata obtained from the hand held device.

19. The participatory sensing system of claim 12, wherein the training models are created using Gaussian Mixture Model classifiers based on audio analysis of horn and non horn sound detected from the object area.

20. The participatory sensing system of claim 12, wherein the object area condition refers to traffic condition thereof.