Data dissemination to mobile sinks in wireless sensor networks: an information theoretic approach

Abstract

Uncertainty is an inherent characteristic of wireless sensor networks (WSNs) due to uncertainty in wireless communication links, limited resources (e.g., energy, storage, CPU, bandwidth), mobility and topology, to name a few. Thus, resource-efficient data dissemination between source sensors and a sink becomes a challenging task, particularly if the sink is moving in a wireless sensor field. This paper proposes an energy-aware protocol, called weighted entropy data dissemination (WEDAS) for disseminating data to the mobile sink in WSNs using an information theoretic approach. The proposed protocol attempts to quantify the uncertainty of position of a mobile sink and the remaining energy uncertainty of static sensors in order to select the most appropriate ones that will act as data disseminators between static sources and the mobile sink. The WEDAS protocol favors sensor nodes whose weighted entropy with respect to their location and remaining energy is the minimum to participate in building dissemination paths between sources and the mobile sink. Specifically, we introduce the concepts of relative mobility zone of a sink, which includes its most probable future positions, and coordinator node set, which restricts the search space of candidate data disseminators, to conduct the selection process. The analytic results show that the selection of sensors with minimum weighted entropy as data disseminators depends on not only their remaining energy but also their relative positions with respect to the mobile sink, which meets our goals of extending the lifetime of WSNs by minimizing and balancing energy utilization of sensor nodes