面向三维有向感知模型的WMSN全目标覆盖控制算法研究
发布时间：2018-07-25 20:49
【摘要】：无线多媒体传感器网络,是一种新型的传感器网络,由具有音频、视频、图像等多媒体信息感知功能的传感器节点组成的网络。在该网络中,由于大量传感器节点随机分布在监测区域内,可能导致多个传感器节点同时覆盖一个目标或目标有遗漏等问题,这样会造成传感器资源的浪费和网络性能的下降,为了实现有效的监测,监控全部的目标。即用最少的传感器覆盖全部的目标,这就是全目标覆盖问题。本课题为解决全目标覆盖问题,从以下三方面进行研究,具体如下:由于全目标问题属于NP-hard问题,而群智能优化算法是解决该问题最为有效的方法。引力搜索算法是该领域现在最具前沿、最具典型性、性能最优的进化算法之一,与其他群智能优化算法相比,引力搜索算法在解决无线传感器网络全目标覆盖问题上具有一定优势,但是在实际无线多媒体传感器网络中,所用的传感器节点数量都非常多,这对核心优化算法的性能提出更大要求。为了提高基础引力搜索算法的优化性能,在原有进化策略的基础上引入了差分变异策略;为了平衡算法的全局探索和局部搜索能力,引入权重函数公式;为最大限度的防止算法陷入局部最优解,本课题把质量值的大小进行降序排列,将个体分为优秀、中间、劣质三类,在不同的迭代时期,采用不同类别的个体进行引力的计算,综合以上分析提出一种基于权重函数分段的引力搜索算法。通过仿真实验,证明本课题所提算法与其它改进算法相比,收敛精度大部分可以达到理论最优值,收敛速度明显加快。为贴近无线多媒体传感器网络的实际监测场景,提高覆盖控制算法的实际应用效果,建立三维有向感知模型的全目标覆盖数学模型。通过分析三维有向感知模型的拓扑结构,明确各点坐标,经过数学推导,确定目标覆盖条件,即目标同时满足在传感器的感知范围之内和区域视角中,判断出传感器与目标的覆盖关系,建立全目标覆盖数学模型。为解决全目标覆盖问题,提出了全目标覆盖控制算法。该算法通过调整各传感器的仰俯角和偏向角,使每个传感器之间相互配合覆盖全部目标。而调整传感器的仰俯角和偏向角属于一个优化问题,从而采用本课题提出的基于权重函数分段的引力搜索算法作为解决该优化问题的核心算法,实验结果表明,该算法能够用较少的传感器覆盖全部的目标。
[Abstract]:Wireless multimedia sensor network (WSN) is a new type of sensor network, which is composed of sensor nodes with the functions of audio, video, image and other multimedia information perception. In this network, because a large number of sensor nodes are randomly distributed in the monitoring area, it may lead to multiple sensor nodes covering one target or missing targets at the same time, which will lead to the waste of sensor resources and the degradation of network performance. To achieve effective monitoring, monitor all targets. That is, using the least sensor to cover all the targets, this is the whole target coverage problem. In order to solve the whole target coverage problem, this paper studies the following three aspects: because the whole objective problem belongs to the NP-hard problem, the swarm intelligence optimization algorithm is the most effective method to solve the problem. Gravitational search algorithm is one of the most advanced, typical and optimal evolutionary algorithms in this field, compared with other swarm intelligence optimization algorithms. Gravity search algorithm has some advantages in solving the problem of full target coverage in wireless sensor networks, but in practical wireless multimedia sensor networks, the number of sensor nodes is very large. This puts forward more requirements for the performance of the core optimization algorithm. In order to improve the optimization performance of the basic gravitational search algorithm, the differential mutation strategy is introduced on the basis of the original evolutionary strategy, and the weight function formula is introduced to balance the global exploration and local search ability of the algorithm. In order to prevent the algorithm from falling into the local optimal solution, in this paper, the size of the mass value is arranged in descending order, and the individuals are divided into three categories: excellent, intermediate and inferior. In different iteration periods, different classes of individuals are used to calculate the gravity. Based on the above analysis, a gravity search algorithm based on weight function segmentation is proposed. The simulation results show that compared with other improved algorithms, the convergence accuracy of the proposed algorithm can reach the theoretical optimal value, and the convergence speed is obviously accelerated. In order to get close to the actual monitoring scene of wireless multimedia sensor network and improve the practical application effect of coverage control algorithm, a full target coverage mathematical model of 3D directed perception model is established. By analyzing the topological structure of the 3D directed perception model and defining the coordinates of each point, the target coverage conditions are determined by mathematical derivation, that is, the target is satisfied within the sensor's perceptual range and the regional perspective. The relation between sensor and target coverage is determined, and the mathematical model of full target coverage is established. In order to solve the problem of full target coverage, a full target coverage control algorithm is proposed. By adjusting the pitch angle and deflection angle of each sensor, the algorithm can cover all the targets by matching each sensor with each other. Adjusting the pitch angle and deflection angle of the sensor belongs to an optimization problem, so the gravity search algorithm based on the weight function segmentation is used as the core algorithm to solve the optimization problem. The experimental results show that, The algorithm can cover all targets with fewer sensors.
【学位授予单位】：东北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN919.8
[Abstract]:Wireless multimedia sensor network (WSN) is a new type of sensor network, which is composed of sensor nodes with the functions of audio, video, image and other multimedia information perception. In this network, because a large number of sensor nodes are randomly distributed in the monitoring area, it may lead to multiple sensor nodes covering one target or missing targets at the same time, which will lead to the waste of sensor resources and the degradation of network performance. To achieve effective monitoring, monitor all targets. That is, using the least sensor to cover all the targets, this is the whole target coverage problem. In order to solve the whole target coverage problem, this paper studies the following three aspects: because the whole objective problem belongs to the NP-hard problem, the swarm intelligence optimization algorithm is the most effective method to solve the problem. Gravitational search algorithm is one of the most advanced, typical and optimal evolutionary algorithms in this field, compared with other swarm intelligence optimization algorithms. Gravity search algorithm has some advantages in solving the problem of full target coverage in wireless sensor networks, but in practical wireless multimedia sensor networks, the number of sensor nodes is very large. This puts forward more requirements for the performance of the core optimization algorithm. In order to improve the optimization performance of the basic gravitational search algorithm, the differential mutation strategy is introduced on the basis of the original evolutionary strategy, and the weight function formula is introduced to balance the global exploration and local search ability of the algorithm. In order to prevent the algorithm from falling into the local optimal solution, in this paper, the size of the mass value is arranged in descending order, and the individuals are divided into three categories: excellent, intermediate and inferior. In different iteration periods, different classes of individuals are used to calculate the gravity. Based on the above analysis, a gravity search algorithm based on weight function segmentation is proposed. The simulation results show that compared with other improved algorithms, the convergence accuracy of the proposed algorithm can reach the theoretical optimal value, and the convergence speed is obviously accelerated. In order to get close to the actual monitoring scene of wireless multimedia sensor network and improve the practical application effect of coverage control algorithm, a full target coverage mathematical model of 3D directed perception model is established. By analyzing the topological structure of the 3D directed perception model and defining the coordinates of each point, the target coverage conditions are determined by mathematical derivation, that is, the target is satisfied within the sensor's perceptual range and the regional perspective. The relation between sensor and target coverage is determined, and the mathematical model of full target coverage is established. In order to solve the problem of full target coverage, a full target coverage control algorithm is proposed. By adjusting the pitch angle and deflection angle of each sensor, the algorithm can cover all the targets by matching each sensor with each other. Adjusting the pitch angle and deflection angle of the sensor belongs to an optimization problem, so the gravity search algorithm based on the weight function segmentation is used as the core algorithm to solve the optimization problem. The experimental results show that, The algorithm can cover all targets with fewer sensors.
【学位授予单位】：东北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP212.9;TN919.8
【相似文献】
相关期刊论文 前10条
1 葛声,马殿富,怀进鹏;基于角色的群体感知模型[J];软件学报;2001年06期
2 闫临霞,曾建潮;基于任务的群体感知模型[J];太原重型机械学院学报;2004年02期
3 闫临霞,曾建潮;基于任务的群体感知模型的形式化描述[J];计算机工程与应用;2005年30期
4 李建国;汤庸;黄世平;黄帆;;基于任务依赖关系的群体感知模型[J];东南大学学报(自然科学版);2011年02期
5 张晓健;廖开际;;移动协作学习的上下文感知模型研究[J];计算机应用与软件;2010年12期
6 葛声,怀进鹏;基于角色的群体感知模型实现机制[J];计算机工程与应用;2001年23期
7 杨武勇,史美林,姜进磊,杨胜文;基于角色的层次型同步协作感知模型[J];清华大学学报(自然科学版);2005年04期
8 衣晓;薛兴亮;;基于概率感知模型的N段有向覆盖与K重全向覆盖研究[J];现代电子技术;2013年16期
9 贾伟,张佑生,周庆松;一种利用感知模型增加容量的新颖盲水印算法[J];系统仿真学报;2004年07期
10 孙飞显;王建红;李涛;;基于抗体浓度的网络安全态势感知模型[J];郑州大学学报(理学版);2010年01期
相关会议论文 前3条
1 黄帆;汤庸;道炜;冀高峰;吴桂宾;王鹏;;上下文技术在群体感知中的研究与应用[A];第二十四届中国数据库学术会议论文集（研究报告篇）[C];2007年
2 顾春;盛胜我;;厅堂脉冲响应与人耳感知模型的研究[A];2005年声频工程学术交流会论文集[C];2005年
3 朱伟春;;认知网络分布式业务感知模型研究[A];江苏省电子学会2010年学术年会论文集[C];2010年
相关博士学位论文 前1条
1 陈媛Z，
本文编号：2145045
本文链接：https://www.wllwen.com/shoufeilunwen/xixikjs/2145045.html
