5th International ICST Conference on Wireless Internet

Research Article

Traffic load distribution in large-scale and dense wireless sensor networks

Download473 downloads
  • @INPROCEEDINGS{10.4108/ICST.WICON2010.8507,
        author={Qinghua Wang and Tingting Zhang},
        title={Traffic load distribution in large-scale and dense wireless sensor networks},
        proceedings={5th International ICST Conference on Wireless Internet},
        publisher={IEEE},
        proceedings_a={WICON},
        year={2010},
        month={4},
        keywords={Traffic load distribution wireless sensor network},
        doi={10.4108/ICST.WICON2010.8507}
    }
    
  • Qinghua Wang
    Tingting Zhang
    Year: 2010
    Traffic load distribution in large-scale and dense wireless sensor networks
    WICON
    IEEE
    DOI: 10.4108/ICST.WICON2010.8507
Qinghua Wang1,*, Tingting Zhang1,*
  • 1: Department of Information Technology and Media, Mid Sweden University, SE-85170 Sundsvall, Sweden
*Contact email: qinghua.wang@miun.se, tingting.zhang@miun.se

Abstract

Because of the dominating many-to-one communication pattern, traffic load is not evenly distributed over the nodes in a wireless sensor network (WSN). In a large-scale WSN, the multi-hop routing even deteriorates this traffic load imbalance by accumulating traffic flows along their paths to the sink. Understanding the traffic load distribution can guide the network-wide energy allocation, direct the design of routing algorithms, and optimize the node deployment in WSNs. In this paper, we consider a large-scale and dense WSN with nodes evenly deployed in a disk area, and find the traffic load distribution over the nodes as a function of their distances from the sink. Further, the effects of network scale and routing hop length on traffic load distribution are also investigated. The traffic loads of individual nodes are found to be in direct proportion to the radius of the network and in inverse proportion to the mean routing hop length, while independent of network density. The results presented in this paper are verified through extensive simulation experiments.