CS-Collector: A New Approach for Data Collection in Wireless Sensor Networks Based on Two-Dimensional Compressive Sensing

Yong Wang, Zhuoshi Yang, Jianpei Zhang, Feng Li, Hongkai Wen, Yiran ShenSensors 2016 · OpenReview

Abstract

In this paper, we consider the problem of reconstructing the temporal and spatial profile of some physical phenomena monitored by large-scale Wireless Sensor Networks (WSNs) in an energy efficient manner. Compressive sensing is one of the popular choices to reduce the energy consumption of the data collection in WSNs. The existing solutions only consider sparsity of sensors’ data from either temporal or spatial dimensions. In this paper, we propose a novel data collection strategy, CS2-collector, for WSNs based on the theory of Two Dimensional Compressive Sensing (2DCS). It exploits both temporal and spatial sparsity, i.e., 2D-sparsity of WSNs and achieves significant gain on the tradeoff between the compression ratio and reconstruction accuracy as the numerical simulations and evaluations on different types of sensors’ data. More intuitively, with the same given energy budget, CS2-collector provides significantly more accurate reconstruction of the profile of the physical phenomena that are temporal-spatially sparse.
BibTeX
@inproceedings{wang2016cs,
  author = {Yong Wang and Zhuoshi Yang and Jianpei Zhang and Feng Li and Hongkai Wen and Yiran Shen},
  title = {{CS-Collector: A New Approach for Data Collection in Wireless Sensor Networks Based on Two-Dimensional Compressive Sensing}},
  booktitle = {Sensors},
  year = {2016},
  url = {https://openreview.net/forum?id=1iU6McGEgG},
}

Loading related content...

Figure