Energy-Efficient Distributed Data Storage for Wireless Sensor Networks Based on Compressed Sensing and Network Coding
Xianjun Yang,Xiaofeng Tao,Eryk Dutkiewicz,Xiaojing Huang,Y. Jay Guo,Qimei Cui
DOI: https://doi.org/10.1109/twc.2013.090313.121804
IF: 10.4
2013-01-01
IEEE Transactions on Wireless Communications
Abstract:Recently, distributed data storage (DDS) for Wireless Sensor Networks (WSNs) has attracted great attention, especially in catastrophic scenarios. Since power consumption is one of the most critical factors that affect the lifetime of WSNs, the energy efficiency of DDS in WSNs is investigated in this paper. Based on Compressed Sensing (CS) and network coding theories, we propose a Compressed Network Coding based Distributed data Storage (CNCDS) scheme by exploiting the correlation of sensor readings. The CNCDS scheme achieves high energy efficiency by reducing the total number of transmissions N{t_{tot}} and receptions N{r_{tot}} during the data dissemination process. Theoretical analysis proves that the CNCDS scheme guarantees good CS recovery performance. In order to theoretically verify the efficiency of the CNCDS scheme, the expressions for N{t_{tot}} and N{r_{tot}} are derived based on random geometric graphs (RGG) theory. Furthermore, based on the derived expressions, an adaptive CNCDS scheme is proposed to further reduce N{t_{tot}} and N{r_{tot}}. Simulation results validate that, compared with the conventional ICStorage scheme, the proposed CNCDS scheme reduces N{t_{tot}}, N{r_{tot}}, and the CS recovery mean squared error (MSE) by up to 55%, 74%, and 76% respectively. In addition, compared with the CNCDS scheme, the adaptive CNCDS scheme further reduces N{t_{tot}} and N{r_{tot}} by up to 63% and 32% respectively.