CDC : Compressive Data Collection for Wireless Sensor Networks

摘要: Data collection is a crucial operation in wireless sensor networks. The design of data schemes challengingdue to the limited energy supply and hot spot problem. Leveraging empirical observations that sensory possess strongspatiotemporal compressibility, this paper proposes novel compressive scheme for We adopt power-law decaying model verified by real sets then propose random projection-based estimation algorithm model. Our requires fewer compressed measurements, thus greatly reduces consumption. It allowssimple routing strategy without much computation control overheads, which leads strong robustness practical applications. Analytically, we prove it achieves optimal error bound. Evaluations on (from GreenOrbs, IntelLab NBDC-CTD projects) show compared with existing approaches, new prolongs network lifetime $1.5 \times$ $2 5-20 percent.