Dynamic duty-cycle control for wireless sensor networks using artificial neural network (ANN)

Department

Department of Computer Science

Was this content written or created while at IBA?

Yes

Document Type

Conference Paper

Publication Date

7-1-2017

Author Affiliation

  • Anwar Ahmed Khan is Ph.D. Scholar at the Faculty of Computer Science, Institute of Business Administration, Karachi
  • Mohammad Shoaib Jamal is Assistant at Institute of Business Administration, Karachi
  • Shama Siddiqui is Ph.D. Scholar at the Faculty of Computer Science, Institute of Business Administration, Karachi

Conference Name

2017 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery (CyberC)

Conference Location

Nanjing, China

Conference Dates

12-14 Oct. 2017

ISBN/ISSN

85050185712 (Scopus)

First Page

420

Last Page

424

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Abstract / Description

Efficient management of duty-cycle in Wireless Sensor Networks (WSN) results in considerable performance improvement of MAC protocols by catering to the issues of overhearing and idle listening. The researchers' interest has recently shifted to adaptive control of duty-cycle which offers an opportunity to conserve energy of WSN in highly dynamic environments, without external intervention. Machine learning techniques such as Artificial Neural Network (ANN) have been deployed for embedding artificial decision making capabilities in WSN. This paper integrated ANN with WSN to achieve dynamic duty-cycle control in WSN, at the receiver end. ANN is used to predict the data arrivals instants at which the receiver should wake up. The scheme has been implemented in MATLAB and the results revealed significant improvement in terms of delay and energy as compared to if the fixed wakeup and sleep intervals are used.

Link to Full Text

Find in your library

Share

COinS