On Reliable Transport and Estimation of Spatio-temporal Events Using Wireless Sensor Networks

This paper studies the efffect of channel errors on the transport and estimations of spatio-temporal events based on sensor observations monitoring a spatio-temporal process. It is observed that the quality of event estimation degrades considerably with channel fading. A new form of spatial diversity is proposed to mitigate the effect of channel fading at the sink. Dense deployment of sensor nodes and the underlying spatio-temporal nature of the observed phenomenon results in highly correlated observations in the spatial domain. The proposed algorithm exploits this feature, using supervised local function learning to achieve spatial diversity. Simulation results are provided to demonstrate the effectiveness of this approach.