Local reliability aware random access for correlated sources in WSNs

In this paper, we study a random access scheme for distributed estimation to estimate a target signal through multiple sensors in wireless sensor networks (WSNs). Since the random access scheme is based on (slotted) ALOHA, it might be possible to implement a WSN for distributed estimation using machine-type communications within cellular systems. Unfortunately, in conventional ALOHA, since the signal at each node is assumed to be independent, when the signals of nodes or sensors are correlated, it may not be efficient. Thus, in order to take into account the correlation of sensors' observations, we propose a transmission rule for each sensor based on local reliability of sensor's local estimate in ALOHA, which can improve the performance of distributed estimation. A generalization with multiple subchannels is also considered and it is shown that the performance can be improved in terms of mean squared error.