School of Automation, China University of Geosciences, Wuhan 430074, China.
Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems, Wuhan 430074, China.
Sensors (Basel). 2019 Sep 23;19(19):4106. doi: 10.3390/s19194106.
Coverage and network lifetime are two fundamental research issues in visual sensor networks. In some surveillance scenarios, there are some critical locations that demand to be monitored within a designated period. However, with limited sensor nodes resources, it may not be possible to meet both coverage and network lifetime requirements. Therefore, in order to satisfy the network lifetime constraint, sometimes the coverage needs to be traded for network lifetime. In this paper, we study how to schedule sensor nodes to maximize the spatial-temporal coverage of the critical locations under the constraint of network lifetime. First, we analyze the sensor node scheduling problem for the spatial-temporal coverage of the critical locations and establish a mathematical model of the node scheduling. Next, by analyzing the characteristics of the model, we propose a Two-phase Spatial-temporal Coverage-enhancing Method (TSCM). In phase one, a Particle Swarm Optimization (PSO) algorithm is employed to organize the directions of sensor nodes to maximize the number of covered critical locations. In the second phase, we apply a Genetic Algorithm (GA) to get the optimal working time sequence of each sensor node. New coding and decoding strategies are devised to make GA suitable for this scheduling problem. Finally, simulations are conducted and the results show that TSCM has better performance than other approaches.
覆盖范围和网络寿命是视觉传感器网络中的两个基本研究问题。在某些监控场景中，存在一些关键位置需要在指定的时间段内进行监测。然而，由于传感器节点资源有限，可能无法同时满足覆盖范围和网络寿命的要求。因此，为了满足网络寿命的约束，有时需要牺牲覆盖范围来换取网络寿命。在本文中，我们研究了如何在网络寿命的约束下，通过调度传感器节点来最大化关键位置的时空覆盖范围。首先，我们分析了用于关键位置的时空覆盖的传感器节点调度问题，并建立了节点调度的数学模型。接下来，通过分析模型的特点，我们提出了一种两阶段的时空覆盖增强方法（TSCM）。在第一阶段，采用粒子群优化（PSO）算法来组织传感器节点的方向，以最大化被覆盖的关键位置的数量。在第二阶段，我们应用遗传算法（GA）来获得每个传感器节点的最佳工作时间序列。设计了新的编码和解码策略，使 GA 适用于这个调度问题。最后，进行了仿真实验，结果表明 TSCM 比其他方法具有更好的性能。
