Abstract
Distributed estimation is examined for energy-harvesting wireless sensor networks, where the energy available at the sensors are converted entirely from ambient sources. Here, each sensor takes a local measurement of the common parameter of interest and forwards the information to the fusion center (FC) using a digital forwarding (DF) transmission strategy. In the DF system under consideration, local observations are first quantized using a uniform quantizer and transmitted to the FC using BPSK modulation. In this work, two cases are considered: the case with no batteries at the sensors and the case with finite-capacity batteries at the sensors. {In the case with no batteries, each sensor transmits with whatever energy it has accumulated in that transmission period; in the case with finite-capacity batteries, {each sensor transmits with a constant power} whenever its accumulated energy is sufficient to do so. The maximum-likelihood estimator (MLE) is adopted at FC to combine the information received from all sensors and is derived based on the statistics of the energy arrivals. In the case with no batteries, an approximate MLE is derived by treating the transmission noise as Gaussian. The optimal number of quantization levels (i.e., the number of bits to be transmitted in each observation period) and the transmit power in the case with batteries} are determined by maximizing the Fisher information. The effectiveness of our proposed schemes is demonstrated through Monte Carlo simulations.