Fu Pengcheng, Cheng Yongbo, Tang Hongying, Li Baoqing, Pei Jun, Yuan Xiaobing
Science and Technology on Microsystem Laboratory, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 201800, China.
University of Chinese Academy of Sciences, Beijing 100049, China.
Sensors (Basel). 2017 Mar 20;17(3):639. doi: 10.3390/s17030639.
In this paper, we propose an effective and robust decentralized tracking scheme based on the square root cubature information filter (SRCIF) to balance the energy consumption and tracking accuracy in wireless camera sensor networks (WCNs). More specifically, regarding the characteristics and constraints of camera nodes in WCNs, some special mechanisms are put forward and integrated in this tracking scheme. First, a decentralized tracking approach is adopted so that the tracking can be implemented energy-efficiently and steadily. Subsequently, task cluster nodes are dynamically selected by adopting a greedy on-line decision approach based on the defined contribution decision (CD) considering the limited energy of camera nodes. Additionally, we design an efficient cluster head (CH) selection mechanism that casts such selection problem as an optimization problem based on the remaining energy and distance-to-target. Finally, we also perform analysis on the target detection probability when selecting the task cluster nodes and their CH, owing to the directional sensing and observation limitations in field of view (FOV) of camera nodes in WCNs. From simulation results, the proposed tracking scheme shows an obvious improvement in balancing the energy consumption and tracking accuracy over the existing methods.
在本文中，我们提出了一种基于平方根容积信息滤波器（SRCIF）的有效且稳健的分布式跟踪方案，以平衡无线摄像机传感器网络（WCNs）中的能量消耗和跟踪精度。更具体地说，针对WCNs中摄像机节点的特性和约束，在该跟踪方案中提出并集成了一些特殊机制。首先，采用分布式跟踪方法，以便能够高效且稳定地实现跟踪。随后，考虑到摄像机节点能量有限，通过基于定义的贡献决策（CD）采用贪婪在线决策方法动态选择任务簇节点。此外，我们设计了一种高效的簇头（CH）选择机制，将这种选择问题转化为基于剩余能量和到目标距离的优化问题。最后，由于WCNs中摄像机节点视场（FOV）的方向感知和观测限制，我们还对选择任务簇节点及其CH时的目标检测概率进行了分析。从仿真结果来看，所提出的跟踪方案在平衡能量消耗和跟踪精度方面比现有方法有明显改进。
