Image transmissions with security enhancement based on region and path diversity in wireless sensor networks

Honggang Wang, Dongming Peng, Wei Wang, Hamid Sharif, Hsiao Hwa Chen

Research output: Contribution to journalArticle

21 Scopus citations

Abstract

Transmissions of large sized images can be a bottleneck for a wireless sensor network (WSN) due to its limited resources. Security can be another concern. This paper proposes a collaborative transmission scheme for image sensors to utilize inter-sensor correlations to decide the transmission and security sharing patterns based on the path diversities. Our proposed approach for secret image sharing on multiple node-disjoint paths for image delivery is to achieve high security without any key distribution and management, and thus the key management related problems do not exist. The energy efficiency is another major contribution made in this paper. This scheme does not only allow each image sensor to transmit optimal fractions of overlapped images through appropriate transmission paths in an energy-efficient way, but also provides unequal protection to overlapped image regions by path selections and adaptive bit error rate (BER) requirement. The simulation results show that the proposed scheme can achieve considerable gains in terms of network lifetime extension, image transmission security enhancement, image quality improvement, and energy efficiency for wireless sensor networks.

Original languageEnglish (US)
Article number4786437
Pages (from-to)757-765
Number of pages9
JournalIEEE Transactions on Wireless Communications
Volume8
Issue number2
DOIs
StatePublished - Feb 2009

Keywords

  • Collaborative image transmission
  • Network lifetime
  • Network security
  • Wireless sensor network

ASJC Scopus subject areas

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Image transmissions with security enhancement based on region and path diversity in wireless sensor networks'. Together they form a unique fingerprint.

  • Cite this