Joint sensor location/power rating optimization for temporally-correlated source estimation

Abstract

CitationBushnaq OM, Chaaban A, Al-Naffouri T (2017) Joint sensor location/power rating optimization for temporallycorrelated source estimation. Abstract-The optimal sensor selection for scalar state parameter estimation in wireless sensor networks is studied in the paper. A subset of N candidate sensing locations is selected to measure a state parameter and send the observation to a fusion center via wireless AWGN channel. In addition to selecting the optimal sensing location, the sensor type to be placed in these locations is selected from a pool of T sensor types such that different sensor types have different power ratings and costs. The sensor transmission power is limited based on the amount of energy harvested at the sensing location and the type of the sensor. The Kalman filter is used to efficiently obtain the MMSE estimator at the fusion center. Sensors are selected such that the MMSE estimator error is minimized subject to a prescribed system budget. This goal is achieved using convex relaxation and greedy algorithm approaches.

Sensor placement and resource allocation for energy harvesting IoT networks

“…Hence, the temporal correlation between observations that are separated by orders of a few seconds tend to be high. Scalar state estimation is studied in [2,19,14] in order to obtain a simple and optimal sensor selection solution.…”

“…• We model a practical system which takes the EH, channel gain and measurement accuracy into account. Similar considerations were taken into account in [1,2,22,16], however, all of them assumed analog communication where sensors directly amplify and forward observations.…”

“…If Ω is empty, go back to step(2).6: Define ŝ( j) l,k,b such that ŝ( j) l,k,b = 1 if ψ ( j) l = {k, b} otherwise, ŝ( j) l,k,b = 0, ∀l, k, b, j. 7:The suboptimal Boolean solution is S = arg min j∈Ω tr{Σ θ|y ( Ŝ( j) )} for the static source and arg min j∈Ω γ( Ŝ( j) ) for the dynamic source.…”