The authors study the performance of the randomized scheduling algorithm and explore the impact of the size of intrusion object on the sensor network’s configuration.Since energy efficiency and reasonable sensing coverage can be achieved by exploiting the sensing spatial redundancy, redundant sensors may be turned off to save energy [16, 17, 18, 19]. However, the network connectivity is not considered in those schemes. In order to further reduce energy and computational overhead, some scheduling schemes [2, 16, 19, 20] operate without the location information or time synchronization. Although the joint problem of coverage and connectivity is considered in [21, 22, 23, 24, 25], the optimization of the sensing spatial redundancy is not taken into account.

A survey of energy-efficient scheduling mechanisms in sensor networks is detailed in [26].In contrast, the approaches of this paper consider coverage, connectivity, and sensing spatial redundancy simultaneously in order to improve energy efficiency in a hierarchical network structure. For the CASA approach, the clusterhead collects local topology information to manage the sensing schedule centrally. By approximating the network behavior throughout the neural network learning process, the clusterhead may be able to roughly predict the performance of the scheduling management. For the DASA approach, the setting of the random waiting timer allows each sensor to exploit the information about coverage, connectivity, and sensing spatial redundancy such that a balance of network resources can be maintained.

Due to the randomized property of the waiting timer, the probabilistic model is proposed to abstract global network behavior. The comparison of the proposed approaches and the other cluster-based schemes [10][11] is further discussed in Section 5.3.?Dynamic Sensor Scheduling AlgorithmsThis section describes two scheduling management schemes for organizing the sensing tasks, the Centralized Adaptive Scheduling Algorithm (CASA) and the Distributed Adaptive Scheduling Algorithm (DASA). The main assumptions of the network are: (1) All sensors are homogeneous with the same transmission range; (2) The sensors are fixed without location information; (3) Symmetric communication channel: all links between sensors are bidirectional; (4) All sensors perform the sensing task periodically. Note that there are no base stations to coor
Conventional ultrasonic inspection of large structures is very time-consuming because the transducer needs to be scanned over Dacomitinib each point of the structure to be tested. The use of guided waves is potentially a very attractive solution to this problem since they can be excited at one point of the structure and can be propagated over considerable distances [1].