WSN target tracking sleep scheduling method based on fault tolerance

Abstract

The invention relates to a WSN (Wireless Sensor Network) target tracking dormancy scheduling method based on fault tolerance, which comprises the following steps that: a node sensing the appearance of a target first competes a cluster head based on time delay broadcast and forms a cluster, and nodes in the cluster perform target sensing work; the cluster head calculates the position and the motion trail of the target after collecting the sensing data of the nodes in the cluster to the target; constructing a prediction fault-tolerant domain according to the calculated position of the target in combination with a prediction error; the nodes in the prediction fault-tolerant domain are scheduled through a scheduling algorithm, the scheduling algorithm constructs an objective function according to the node set energy consumption, the sensing coverage area and the redundant overlapping area, and the objective function value is made to be minimum by adjusting the sensing radius and the active state of each node in the prediction fault-tolerant domain. According to the method, the influence of prediction errors on the tracking precision can be reduced, the loss rate of the target is reduced, and relatively small energy is consumed.

Description

technical field [0001] The invention relates to the technical field of wireless sensor networks, in particular to a fault-tolerant WSN target tracking sleep scheduling method. Background technique [0002] Wireless Sensor Network (WSN), as a new way of information perception and acquisition, has been widely used in environmental detection, medical care, intelligent transportation and other fields. Among them, the target tracking technology supported by large-scale WSN can be directly applied to many scenarios, including location finding of missing persons, vehicle navigation of intelligent transportation systems, and so on. [0003] Object tracking algorithms usually organize the network into a logical structure (such as tree, cluster or face) to achieve data fusion. These algorithms often use Kalman Filter (KF), Particle Filter (PF) and other methods to estimate the future position of the target. Then, according to the predicted position, combined with the sleep schedulin...

AI Technical Summary

Problems solved by technology

Although this mechanism reduces the loss rate of the target to a certain extent, it has two main defects: first, activating a large number of nodes to participate in retrieving the target will cause additional energy consumption; Before the target is retrieved, its related sensory data will be lost

Images