容积卡尔曼一致滤波

针对卡尔曼一致滤波的应用受限于被估计系统需 满足线性条件的问题,通过容积卡尔曼滤波(CKF)和一致性策 略的动态结合,提出一种容积卡尔曼一致滤波(CKCF)算法。算法采用分布式融合机制, 传感器节点采集可通信相邻 节点的信息,并作为自身节点的量测信息应用于CKF,获取局部状态估计 值。在此基础上,利用一 致性策略实现对整个量测系统中传感器节点局部估计值的优化,进而通过增强传感器节点估 计值一致性实现目标 状态估计精度的提升。相对于标准卡尔曼一致滤波,本文算法将一致性策略推广到非线性系 统估计领域。理论分析 与仿真实验验证了算法的可行性与有效性。     

基于格拉布斯准则和改进粒子滤波算法的水下传感网目标跟踪

水下无线传感网络(UWSN)执行目标跟踪时,因为各个传感器节点测量值对目标状态估计的贡献不一样以及节点能量有限,所以探索一种好的节点融合权重方法和节点规划机制能够获得更好的跟踪性能。针对上述问题,该文提出一种基于Grubbs准则和互信息熵加权融合的分布式粒子滤波(PF)目标跟踪算法(GMIEW)。首先利用Grubbs准则对传感器节点所获得的信息进行分析检验,去除干扰信息和错误信息。其次,在粒子滤波的重要性权值计算的过程中,引入动态加权因子,采用传感器节点的测量值与目标状态之间的互信息熵,来反映传感器节点提供的目标信息量,从而获得各个节点相应的加权因子。最后,采用3维场景下的簇-树型网络拓扑结构,跟踪监测区域内的目标。实验结果显示,该算法可有效提高水下传感器网络测量数据对目标跟踪预测的准确度,降低跟踪误差。     

基于能量有效的水下分布式粒子滤波跟踪算法

为降低水下无线传感器网络目标跟踪算法能耗并提高定位精度,提出基于能量有效的分布式粒子滤波跟踪算法(EEPF算法)。EEPF算法通过能量有效的最优分布式动态成簇机制和启发式能量有效的调度算法来平衡水下节点间的能耗,延长网络生存期,并在预测、滤波、重采样阶段对粒子滤波算法进行改进,在保障期望目标跟踪精度的同时降低了运算能耗。仿真结果表明,EEPF算法是一种轻量级的能量有效的目标跟踪算法,该算法能耗低,网络存活时间长,且跟踪精度较传统粒子滤波算法有了较大提高。     

变拓扑非完全连通网络中的分布式机动目标跟踪算法研究

完全分布式的机动目标跟踪是传感器网络等应用中亟待解决的关键问题。本文针对变拓扑非完全连通网络,提出一种基于网络共识的多模型信息滤波器( Consensus based Multiple Model Information Filter, C-MMIF)。 C-MMIF基于标准IMM框架,保证了估计最优性;并通过构造目标运动模式概率和状态估计的信息滤波形式,使节点间运算相互独立。同时,每个独立节点仅需与其相邻节点通讯,利用平均网络共识分布式优化算法对自身信息状态进行更新,实现节点间对目标运动模式及状态的一致估计。最后在无人机与地面传感器网络协同对地机动目标跟踪场景下进行算法仿真验证,结果证明该方法可以在无融合处理中心且网络拓扑变化情况下,使各节点实现对机动目标的一致有效跟踪。     

一种基于异步传感器网络的空间目标分布式跟踪方法

为解决传感器网络在空间目标分布式跟踪过程中的异步采样及通信延迟问题,该文提出一种异步分布式信息滤波算法(ADIF)。首先,局部传感器与相邻节点之间以一定的拓扑结构传递带采样时标的局部状态信息和量测信息,然后将收到的异步信息按时间排序,使用ADIF算法进行计算,分别对目标状态进行估计。该方法实现简单,传感器间通信的次数少,支持网络拓扑的实时变化,适用于空间目标监测中的多目标跟踪问题。该文分别对空间单目标、多目标跟踪进行了仿真,结果表明算法可以有效解决异步传感器滤波问题,分布式滤波精度一致逼近于集中式结果。     

雷达机动目标跟踪的卡尔曼粒子滤波算法

为解决不敏粒子滤波算法对雷达机动目标跟踪实时性差和跟踪起始阶段收敛慢的问题,引入卡尔曼粒子滤波算法。通过坐标转换将实际的极坐标雷达观测数据转换为直角坐标数据,然后用线性最优的卡尔曼滤波器估计粒子状态先验概率密度,最后用非线性最优的粒子滤波器精确估计目标状态后验概率。仿真实验表明,与不敏粒子滤波相比,卡尔曼粒子滤波以牺牲较少精度(减少约6%)的代价,实现机动目标跟踪的实时性(约为前者的1/5),起始阶段收敛性更好。     

自调整分层卡尔曼粒子滤波的快速目标跟踪

分层卡尔曼粒子滤波成功应用于目标跟踪,但其只对目标位置进行了优化,忽略了其他仿射参数,导致跟踪中的粒子数目仍然很大。为了实现复杂环境下的快速目标跟踪,提出一种带有自调整策略的分层卡尔曼粒子滤波方法。该方法将目标划分为线性和非线性状态空间,并通过少量粒子的迭代过程在非线性状态空间逐步搜索最优状态。其详细过程如下:首先,利用卡尔曼滤波预测目标位置,结合目标运动信息计算潜在目标区域;然后在该区域内生成一组随机粒子,通过在线姿态估计对粒子状态进行调整,并将观测结果与目标模板进行比较,修正粒子摄动的方向以逼近目标。把该方法应用于大机动目标的视频序列中,并与现有的跟踪方法进行了对比。结果表明,所提方法能够以少量粒子实现准确、稳定的目标跟踪,大大降低了跟踪算法的运算量,提高了跟踪效果。     

卡尔曼粒子滤波中基于精确运动模型的局部区域估计

粒子滤波广泛应用于对精度和稳定性要求较高的目标跟踪,但其计算量大,并且计算复杂度随着状态量和粒子数目增长迅速增加。将目标跟踪转化为由粗到精的搜索过程,提出了一种基于精确运动模型的改进分层卡尔曼粒子滤波算法。该方法利用加速度的运动模型在真实目标位置的周围估计目标的散布范围,并在该范围内随机生成粒子,寻找精确的目标位置。文中引入加加速度模型主要是由于现有方法的状态量阶数不足,导致模型精确度较低,无法应对大机动目标的跟踪。因此,引入了高阶状态变量加加速度,并将其用于改进分层卡尔曼粒子滤波的运动模型。利用分层卡尔曼粒子滤波、粒子滤波以及提出的方法进行了跟踪试验,结果表明,基于精确运动模型的改进分层卡尔曼粒子滤波模型的跟踪方法能够提高线性运动的预测精度,实现复杂环境下精确稳定的跟踪。     

一种适用于分布式摄像机网络的SCIWCF算法

本文使用容积卡尔曼滤波器来处理分布式摄像机网络中的目标跟踪问题.平方根容积信息滤波（Square-Root Cubature Information Filter,SCIF）是容积卡尔曼滤波的一种扩展,其具有有效性和可靠性等方面优势,有利于对多源信息进行信息融合.然而当该算法应用于像摄像机网络这种大规模网络时,如果采用一般的集中式处理,中心节点可能会承受较大的计算压力.针对这个问题,本文首先将平方根容积信息滤波器进行了扩展,提出分布式平方根容积信息滤波器,使其能适应大规模网络.另外在摄像机网络中,由于摄像机装置在一个较大的区域内,由于摄像机观测区域有限,目标可能会出现在观察的盲区,这样就会存在某些摄像机的测量数据无效.针对这个问题,本文提出了平方根容积信息加权一致性滤波器（Square-Root Cubature Information Weighted Consensus Filter,SCIWCF）对状态信息和信息矩阵加权,减小这些无效信息在一致性算法的作用,从而提高整体的滤波性能.仿真实验结果表明,本文提出的算法能够在摄像机网络中对目标进行有效跟踪,在估计精度和滤波器稳定性等方面要优于传统的信息滤波.     

自适应非线性GM-PHD滤波及在无源跟踪中的应用

首先针对无源传感器目标跟踪中的非线性问题,将高斯-厄米特求积分规则运用于高斯混合概率假设密度滤波,提出一种求积分卡尔曼概率假设密度滤波。其次,针对未知时变过程噪声,将基于极大后验估计原理的噪声估计器运用到概率假设密度滤波中,同时依据目标状态一步预测与状态滤波结果之间的残差,提出一种对滤波发散情况判断和抑制的算法。最后通过无源传感器双站跟踪仿真表明:相较于已有的非线性高斯混合概率假设密度滤波,所提算法有更高的精度,并且在未知时变噪声环境中具有较好跟踪效果。     

基于动态分簇路由优化和分布式粒子滤波的传感器网络目标跟踪方法

针对无线传感器网络中节点通信能力及能量有限的情况,该文提出基于动态分簇路由优化和分布式粒子滤波的传感器网络目标跟踪方法。该方法以动态分簇的方式将监测区域内随机部署的传感器节点划分为若干个簇,并对簇内成员节点与簇首节点之间、簇首节点与基站之间的通信路由进行优化,确保网络能耗的均衡分布,在此基础上,被激活的簇内成员节点并行地执行分布式粒子滤波算法实现目标跟踪。仿真结果表明,该方法能有效地降低传感器网络中节点的总能耗,能在实现跟踪的同时保证目标跟踪的精度。     

面向目标跟踪的无线传感器网络动态分簇

在考虑节点剩余能量和节点探测信号强度的基础上,提出了一种面向目标跟踪的无线传感器网络动态分簇方法。由于在簇首选择时考虑了节点的剩余能量,因此可以有效地避免簇首节点能量不足导致的跟踪失败。方法既能保证各个节点均衡地担任簇首节点,又能避免簇首切换过于频繁,因此可以延长网络的使用寿命。计算机仿真结果表明,在满足簇首节点能量需求和簇内节点大部分能观测到目标的情况下,应该取尽量小的能量阈值和尽量大的距离阈值。     

Prediction‐based cluster management for target tracking in wireless sensor networks

The key impediments to a successful wireless sensor network (WSN) application are the energy and the longevity constraints of sensor nodes. Therefore, two signal processing oriented cluster management strategies, the proactive and the reactive cluster management, are proposed to efficiently deal with these constraints. The former strategy is designed for heterogeneous WSNs, where sensors are organized in a static clustering architecture. A non‐myopic cluster activation rule is realized to reduce the number of hand‐off operations between clusters, while maintaining desired estimation accuracy. The proactive strategy minimizes the hardware expenditure and the total energy consumption. On the other hand, the main concern of the reactive strategy is to maximize the network longevity of homogeneous WSNs. A Dijkstra‐like algorithm is proposed to dynamically form active cluster based on the relation between the predictive target distribution and the candidate sensors, considering both the energy efficiency and the data relevance. By evenly distributing the energy expenditure over the whole network, the objective of maximizing the network longevity is achieved. The simulations evaluate and compare the two proposed strategies in terms of tracking accuracy, energy consumption and execution time. Copyright © 2010 John Wiley & Sons, Ltd.     

Low-Energy Adaptive Unequal Clustering Protocol Using Fuzzy c-Means in Wireless Sensor Networks

Clustering technique in wireless sensor networks incorporate proper utilization of the limited energy resources of the deployed sensor nodes with the highest residual energy that can be used to gather data and send the information. However, the problem of unbalanced energy consumption exists in a particular cluster node in the network. Some more powerful nodes act as cluster head to control sensor network operation when the network is organized into heterogeneous clusters. It is important to assume that energy consumption of these cluster head nodes is balanced. Often the network is organized into clusters of equal size where cluster head nodes bear unequal loads. Instead in this paper, we proposed a new protocol low-energy adaptive unequal clustering protocol using Fuzzy c-means in wireless sensor networks (LAUCF), an unequal clustering size model for the organization of network based on Fuzzy c-means (FCM) clustering algorithm, which can lead to more uniform energy dissipation among the cluster head nodes, thus increasing network lifetime. A heuristic comparison between our proposed protocol LAUCF and other different energy-aware protocol including low energy adaptive clustering hierarchy (LEACH) has been carried out. Simulation result shows that our proposed heterogeneous clustering approach using FCM protocol is more effective in prolonging the network lifetime compared with LEACH and other protocol for long run.     

基于形态匹配聚类的近邻扩展目标跟踪算法

针对传统扩展目标跟踪（Extended Target Tracking, ETT）算法在处理近邻目标时面临的计算效率低下和跟踪不准确的问题,提出了一种形态匹配聚类量测集划分与高斯逆威沙特概率假设密度（Gaussian Inverse Wishart Probability Hypothesis Density, GIW-PHD）滤波器相结合的跟踪处理方法。该方法首先由GIW-PHD滤波器得到预测的目标状态,其次使用DBSCAN(Density-Based Spatial Clustering of Applications with Noise, DBSCAN)算法完成量测集的初步划分,在此基础上利用较高权重的预测分量实现对多个近邻目标混合量测簇的判断,进而使用椭圆形状约束（Elliptic Shape Constraint, ESC）的FCM(Fuzzy C-Means, FCM)算法（ESC-FCM）对混合簇进行二次划分得到更精确的划分结果,最后将划分结果合并后送入GIW-PHD滤波器完成目标状态的更新。仿真结果表明,本文所提量测集划分方法能够充分利用GIW-PHD滤波器预测步获取...     

Energy-efficient collaborative target tracking algorithm using cost-reference particle filtering in wireless acoustic sensor networks

Target tracking is one of the most important applications of wireless sensor networks. Optimized computation and energy dissipation are critical requirements to save the limited resource of sensor nodes. A new robust and energy-efficient collaborative target tracking framework is proposed in this article. After a target is detected, only one active cluster is responsible for the tracking task at each time step. The tracking algorithm is distributed by passing the sensing and computation operations from one cluster to another. An event-driven cluster reforming scheme is also proposed for balancing energy consumption among nodes. Observations from three cluster members are chosen and a new class of particle filter termed cost-reference particle filter (CRPF) is introduced to estimate the target motion at the cluster head. This CRPF method is quite robust for wireless sensor network tracking applications because it drops the strong assumptions of knowing the probability distributions of the system process and observation noises. In simulation experiments, the performance of the proposed collaborative target tracking algorithm is evaluated by the metrics of tracking precision and network energy consumption.     

Dual head static clustering algorithm for wireless sensor networks

Clustering of nodes is often used in wireless sensor networks to achieve data aggregation and reduce the number of nodes transmitting the data to the sink. This paper proposes a novel dual head static clustering algorithm (DHSCA) to equalise energy consumption by the sensor nodes and increase the wireless sensor network lifetime. Nodes are divided into static clusters based on their location to avoid the overhead of cluster re-formation in dynamic clustering. Two nodes in each cluster, selected on the basis of the their residual energy and their distance from the sink and other nodes in the cluster, are designated as cluster heads, one for data aggregation and the other for data transmission. This reduces energy consumption during intra-cluster and inter-cluster communication. A multi-hop technique avoiding the hot-spot problem is used to transmit the data to the sink. Experiments to observe the energy consumption patterns of the nodes and the fraction of packets successfully delivered using the DHSCA suggest improvements in energy consumption equalisation, which, in turn, enhances the lifetime of the network. The algorithm is shown to outperform all the other static clustering algorithms, while being comparable with the performance of the best dynamic algorithm.     

基于能量和密度的WSNs动态分区成簇路由算法

针对传统的层次型网络存在的分簇不合理和能耗不均衡等问题,提出了一种基于能量和密度的动态非均匀分区成簇路由算法。该算法先根据节点与基站之间的距离将网络合理地进行动态的区域划分,在区域内成簇,使靠近基站的簇规模小于距离基站较远的簇,减少靠近基站的簇首负担和能量消耗;通过综合考虑节点剩余能量和节点密度等因素来优化簇的非均匀划分和簇首的选择,簇首间采取基于数据聚合的多跳传输机制。仿真结果表明,与经典路由算法LEACH相比,该算法能有效均衡节点能耗,延长网络生命周期。     

基于强跟踪滤波的传感器目标跟踪算法

为了提高运动目标的跟踪精度,提出一种基于强跟踪滤波的传感器目标跟踪算法.首先通过传感器节点测量目标的状态值,并通过融合中心对信息进行融合,然后利用Cholesky分解技术变换成噪声独立的量化融合系统,并采用强跟踪滤波算法对目标状态进行估计,最后与其它目标跟踪算法进行对比实验.结果表明,本文算法不仅提高了目标跟踪的精度,而且具有更好的鲁棒性.     

Near optimal cluster-head selection for wireless sensor networks

Clustering in wireless sensor networks is an effective way to save energy and reuse band- width. To our best knowledge, most of the clustering protocols proposed in literature are of a dynamic type, where cluster heads are selected in each period, followed by cluster formation. In this paper, a new static type clustering method called Hausdorff clustering, which is based on the location of sensor nodes as well as communication efficiency and network connectivity, is proposed. The cluster head, however, is rotated within the cluster by a fuzzy logic algorithm that optimizes the network lifetime. Simulation results show that this approach can significantly increase the lifetime of the sensor network.