Standalone noise and anomaly detection in wireless sensor networks: A novel time-series and adaptive Bayesian-network-based approach

Wireless sensor networks (WSNs) consist of small sensors with limited computational and communication capabilities. Reading data in WSN is not always reliable due to open environmental factors such as noise, weakly received signal strength, and intrusion attacks. The process of detecting highly noisy data is called anomaly or outlier detection. The challenging aspect of noise detection in WSN is related to the limited computational and communication capabilities of sensors. The purpose of this research is to design a local time-series-based data noise and anomaly detection approach for WSN. The proposed local outlier detection algorithm (LODA) is a decentralized noise detection algorithm that runs on each sensor node individually with three important features: reduction mechanism that eliminates the noneffective features, determination of the memory size of data histogram to accomplish the effective available memory, and classification for predicting noisy data. An adaptive Bayesian network is used as the classification algorithm for prediction and identification of outliers in each sensor node locally. Results of our approach are compared with four well-known algorithms using benchmark real-life datasets, which demonstrate that LODA can achieve higher (up to 89%) accuracy in the prediction of outliers in real sensory data.     

无线传感网离群点检测技术研究综述

无线传感器网络(wireless sensor network,WSN)受电池能量、计算能力、通信能力和内存空间及传感数据多维特征的限制,传统的离群点检测技术不能直接应用于WSN,因此出现了一系列针对WSN的离群点检测技术.对已有的WSN离群点检测技术进行了概述,根据各离群点检测技术的特征进行了分类和分析,并结合现有技术的缺陷和需求,展望了WSN离群点检测技术的未来研究方向和目标.     

A destination-based approach for cut detection in wireless sensor networks

Wireless sensor networks often suffer from disrupted connectivity caused by its numerous aspects such as limited battery power of a node and unattended operation vulnerable to hostile tampering. The disruption of connectivity, often referred to as network cut, leads to ill-informed routing decisions, data loss and waste of energy. A number of protocols have been proposed to efficiently detect network cuts; they focus solely on a cut that disconnects nodes from the base station. However, a cut detection scheme is truly useful when a cut is defined with respect to multiple destinations (i.e. target nodes), rather than a single base station. Thus, we extend the existing notion of cut detection, and propose an algorithm that enables sensor nodes to autonomously monitor the connectivity to multiple target nodes. We introduce a novel reactive cut detection solution, the point-to-point cut detection, where given any pair of source and destination, a source is able to locally determine whether the destination is reachable or not. Furthermore, we propose a lightweight proactive cut detection algorithm specifically designed for a network scenario with a small set of target destinations. We prove the effectiveness of the proposed algorithms through extensive simulations; specifically, in our network configurations, proposed cut detection algorithms achieve more than an order of magnitude improvement in energy consumption, when coupled with an underlying routing protocol.     

无线传感器网络中基于空间关联性的聚类异常检测算法

针对无线传感器网络节点定位场景中的环境变化、障碍物及传感器节点失效等异常问题, 提出了一种基于空间关联性的聚类异常检测算法（ODCASC算法）。该算法通过聚类技术完成节点的异常数据判断, 同时还利用邻居节点间感知数据的空间关联性过滤噪声数据, 提取异常事件信息, 辅助系统决策。室内实测结果显示, ODCASC算法弥补了时空关联检测算法的不足, 可以有效地检测并区分网络中的环境噪声及异常事件。     

基于时间序列数据的无线传感器网络的异常检测方法

伴随着无线传感器网络的发展和广泛应用,针对传感器网络在恶劣环境下传感器之间采样值差异大以及随着无线传感器网络故障节点增多导致事件检测不准确的问题,提出一种基于传感器网络时间序列数据的检测方法.利用传感器采集的K个正常数据的中位数建立枢轴量,构造置信区间,提出一种计算数据区间差异度的方法以判断发生异常的来源.实验结果表明,本文方法对传感器网络的异常数据检测率保持在98%以上,误报率保持在0.5%以下,具有一定的实用性.     

分布式无线传感器网络容错事件边界检测

事件监测是无线传感器网络的一种重要应用。针对该应用中软故障节点提供的错误数据会降低监测的准确性的问题,提出了一种分布式的容错事件边界检测算法。节点只需与邻节点交换一次传感数据,通过简单地计算识别故障;正常的事件节点利用统计比较的方法判断其是否处于事件边界,边界宽度可根据网络用户的要求调节。该算法执行时所需的通信量小,计算复杂度低,时延小,对大规模网络具有很好的可扩展性。仿真结果表明即使节点故障率很高,应用该算法仍可以获得很好的检测效果。     

管网检测的无线传感网络信息采集同步

无线传感网络的时钟校准决定了各节点信息采集的同步性。在供水管网泄漏检测中,各节点信息采集的时间同步误差将引入漏点的定位误差。在泛洪时间同步协议的基础上,提出拨号时间同步方案,在同步过程中采用定向拨号和线性回归相结合的方法,周期补偿终端节点间因本地时钟源频差而产生的时钟偏移,实现网络时间校准。对单跳网络,每2分钟进行1次校正,测试表明节点间的同步误差每秒在10微秒以内,累积误差不超过1毫秒,这样引起的管网漏点定位误差在1米以内,可以满足供水管网泄漏检测的要求。     

无线传感器网络中基于分簇的节点定位异常检测*

在引入WSN分簇结构基础上,提出一种分布式的节点定位异常检测方法,利用聚类拓扑减少通信量,同时降低以往集中式检测存在的单点风险。该方法不需要任何已知的部署知识或额外的硬件,每个簇的簇头节点只需根据该簇节点报告的位置和邻居表信息进行过滤计算,更新权值,即可确定和撤销定位异常的节点。通过理论分析和仿真模拟验证了这种基于分簇的节点定位异常检测方法的正确性和有效性。     

基于无线传感器网络的车辆检测识别算法研究

针对无线传感器网络(WSNs)的特点,利用车辆发出的声音信号,提出并研究了一种改进的信号检测算法,能够有效地从被噪声严重污染的声音信号中提取出车辆信号。使用小波包变换提取16维信号特征,支持向量机进行目标分类,得到单节点识别结果。提出了基于能量的全局决策融合算法,对多个节点做出的决策进行融合,得到网络的最终识别结果。为了评估算法,使用了来自DARPA SensIT实验中的真实数据,其中包含了履带车和重型卡车的大量声音信号。实验结果表明:该算法用于WSNs中的车辆识别方面是有效的。     

随机功率控制在分簇结构无线传感器网络中的应用

分析了分簇结构无线传感器网络中随机功率控制的成功通信概率,并提出由网络对成功通信概率的要求来决定簇内通信时节点发送功率的最大值和最小值。仿真数据表明,在节点密度较高的分簇结构无线传感器网络中,随机功率控制与固定功率控制相比,有更高的能量效率。     

构建军用无线传感器网络预警与探测系统设想

未来信息化战场上,以夺取制信息权为目的的信息对抗行动将贯穿于作战全过程,谁在信息的获取、传输、处理上占据优势,谁就能抢占战争的主动权.无线传感器网络作为一种新兴的技术发展方向,实现从信息获取到信息传输、处理以及目标系统控制的一体化智能管理,是信息技术发展的更高阶段,将在不远的将来掀起一场军事技术革命和未来战争的变革.研...     

CROSS: A probabilistic constrained random sensor selection scheme in wireless sensor networks

Due to the application-specific nature of wireless sensor networks, the sensitivity to coverage and data reporting latency varies depending on the type of applications. In light of this, algorithms and protocols should be application-aware to achieve the optimum use of highly limited resources in sensors and hence to increase the overall network performance. This paper proposes a probabilistic constrained random sensor selection (CROSS) scheme for application-aware sensing coverage with a goal to maximize the network lifetime. The CROSS scheme randomly selects in each round (approximately) k data-reporting sensors which are sufficient for a user/application-specified desired sensing coverage (DSC) maintaining a minimum distance between any pair of the selected k sensors. We exploit the Poisson sampling technique to force the minimum distance. Consequently, the CROSS improves the spatial regularity of randomly selected k sensors and hence the fidelity of satisfying the DSC in each round, and the connectivity among the selected sensors increase. To this end, we also introduce an algorithm to compute the desired minimum distance to be forced between any pair of sensors. Finally, we present the probabilistic analytical model to measure the impact of the Poisson sampling technique on selecting k sensors, along with the optimality of the desired minimum distance computed by the proposed algorithm.     

MRNN:一种新的基于改进型递归神经网络的WSN动态建模方法:应用于故障检测

提出了一种适用于无线传感器网络WSN的故障检测方法,该方法运用改进的递归神经网络MRNN为WSN的节点、节点的动态特性以及节点间的关系建立相关模型,对WSN节点进行识别和故障检测。MRNN的输入选择建模节点的先前输出值及其邻居节点的当前及先前输出值,模型基于一种新的改进的反向传播型神经网络,该神经网络的输入以及传感器网络的拓扑结构基于通用的非线性传感器模型。仿真实验将MRNN方法与卡尔曼滤波法进行了全面的比较。实验表明,MRNN在置信因子较小的情况下与卡尔曼滤波方法相比有较高的故障检测精度。     

基于分簇的无线传感器网络克隆攻击检测方案

针对无线传感器网络面临的节点克隆攻击,提出了一种基于分簇的检测方案。该方案给网络中的消息都标记上其源节点的ID,在成簇阶段依据节点ID和RSSI值由簇头判断各个簇内是否存在克隆节点,待全部节点的ID汇聚至基站后,利用分簇算法中各个簇是不相交的子集这一特点,在基站处检测是否存在同一个ID属于多个簇的异常情况。仿真实验表明,该方案不但有较高的检测率,还能降低网络的存储和通信开销。     

WSN中一种基于单类1/4球体支持向量机的异常检测方案

针对数据非平稳分布而导致的异常检测差错率较高的问题,提出了一种可用单类1/4球体支持向量机模型参数表示的异常率参数动态调整自适应算法。该算法在线迭代运行,依次生成基于训练集的最优模型,最后提出一种可以迭代得到训练集异常率的参数确定算法,可发现数据集中的异常。仿真实验结果表明,与异常率固定的静态模型相比,该算法可以利用训练集生成最优模型,进而实现分类集差错率最小。     

无线传感器网络轻量级数据加密机制设计

针对无线传感器网络处理能力、存储空间、能量等有限的特点,设计了轻量级数据加密机制。该机制对RC6算法进行了改进,添加了"对称层"运算,使改进后的RC6算法在运算工作量变化不大的情况下,硬件实现更加容易,硬件资源消耗更小。为进一步提高密文的安全性与数据加密强度,使用双密钥对明文进行两级加密,并引入了随机密钥管理机制,使网络节点每次加密时都能使用不同的密钥,提高了密钥的安全性。数据加密机制还使用了节点ID认证、带有身份标识的密钥池认证等多种安全认证机制来阻止非法节点的接入。实验基于低功耗Cortex-M3内核的控制芯片搭建无线传感器网络节点硬件平台,设计了通信协议,并在硬件平台上移植与实现了该机制。实验结果表明,该加密机制能够很好地在低功耗平台上运行。     

基于信任的无线传感器网络时隙分配博弈分析

为了帮助无线传感器网络作出既有利于自身收益又能抑制恶意节点的决策,提出了一种信任激励的时隙分配博弈模型。根据收益矩阵对恶意节点和簇头之间的非零和博弈关系进行了认知和分解,指出了破坏行为可以抑制的原因是纳什均衡可以作为惩罚阻止节点偏离收益更高的策略组合。在此基础上建立了博弈模型,然后证明了无限重复博弈中的纳什回归策略成为子博弈完美均衡的充分必要条件。采用单轮纳什均衡惩罚合作性策略的偏离者,从而使恶意节点与簇头的无限重复博弈能够产生合作效应。仿真实验结果表明,根据该模型作出的决策可以增加网络收益并抑制恶意节点的破坏行为。     

FDR: fault detection and recovery scheme for wireless sensor networks using virtual grid

Due to autonomous operation and constrained resources, nodes of a wireless sensor network are susceptible to failures. Due to multiple node failures, network topology may change or partition into many disconnected segments causing data/query path breakage. Early detection and recovery of faults is desirable in most scenarios. Also, due to limited battery life of a node, the solution must consume minimum possible energy for prolonged network operation. Hence in this paper, we propose a Fault Detection and Recovery scheme which is an energy efficient fault detection and recovery strategy that achieves minimum data loss by efficiently replacing faulty node on the present route or by finding full or partial alternate paths. Topology is managed by constructing a virtual grid over the entire network which helps in managing dynamically changing network topology easily and makes failure detection and recovery effective. It also helps to create energy efficient path between a source and a sink by finding shortest possible path. Further, nodes’ cooperation is exploited to create certain zones on the data/query path which provides alternate nodes or possible alternate paths if required on some node failures. Thus, scheme achieves fault tolerance and at the same time achieves energy efficiency by always selecting shortest path for data delivery between source and sink. Analytical and simulation study reveals the significant improvement in terms of fault detection and energy conservation over existing similar schemes.     

无线多媒体传感器网络能量均衡的多跳图像传输机制

针对由普通节点和图像节点组成的异构无线多媒体传感器网络,提出了基于图像节点邻域协作压缩的多跳图像传输机制MHIT。该机制在发送图像前,首先根据传输距离和路由跳数判断是否需要压缩图像,若压缩图像后再传输消耗更多能量,则直接发送图像;否则,图像节点将图像压缩任务分发给邻域内的普通节点协作完成,均衡了网络能量消耗,极大地缓解了图像节点的能耗压力。实验结果表明,MHIT有效解决了无线多媒体传感器网络因图像压缩而引发的能量空洞问题,明显延长了网络生存期,特别适合于大规模无线多媒体传感器网络远距离图像传输。