无线传感器网络动态分簇路由BWAS的算法研究

为加快无线传感器网络(WSN)路径搜索速度,减少了路径寻优能量消耗,提出了基于最优-最差蚂蚁系统(BWAS)算法的无线传感器网络动态分簇路由算法。该算法是基于WSN动态分簇能量管理模式,在簇头节点间运用BWAS算法搜寻从簇头节点到汇聚节点的多跳最优路径,以多跳接力方式将数据发送至汇聚节点。BWAS算法在路径搜寻过程中评价出最优-最差蚂蚁,引入奖惩机制,加强搜寻过程的指导性。结合动态分簇能量管理,避免网络连续过度使用某个节点,均衡了网络节点能量消耗。通过与基于蚁群算法(ACS)路由算法仿真比较,本算法减缓了网络节点的能量消耗,延长了网络寿命,在相同时间里具有较少的死亡节点,具有较强的鲁棒性。     

一种基于UAV的无人海岛监控网络数据收集策略

针对传统的最小跳路由无线传感器网络（WSN）在数据汇聚上较高的能量开销问题,提出了一种基于无人机(UAV)数据收集的动态分簇算法,其主要思想是利用节点剩余能量来确定那些节点可以当选簇首,同时利用节点坐标位置和设定地分簇半径来划分簇的大小。该算法的优势是能最大程度地均衡每个传感器节点的能量,使整体的节点剩余的能量维持在同一水平。为了提高数据收集的效率,采用蚁群算法规划了无人机数据收集的最短路径。仿真结果表明,与相同的分簇算法下传统的最小跳路由无线传感器网络相比,所提出的基于无人机的无线传感器网络（UAV-WSN）在能量利用率和生命周期方面分别提升了15%和25%,并且以上两种网络的能量利用率高达70%。     

引入移动数据收集器的WSN节能策略

在无线传感器网络中引入移动数据收集器可以减少因数据远距离传输带来的能量消耗,在层次型无线传感器网络中,移动数据收集器只需要访问所有的簇头节点就可以收集到全网数据。为了防止簇头节点在移动数据收集器到来之前因存储空间不足发生溢出,提出了最大化缓存机制(Time Highest-Overflow-Based Mechanism,THM)算法,将每个簇内所有节点的存储空间合并成大空间的虚拟存储器。实验结果表明,THM算法能够有效地延长网络的生存时间。     

低能耗和低时延的无线传感器网络数据融合算法

针对无线传感器网络的节点能量有限,且在进行信息传输时存在数据冲突、传输延时等问题,提出并设计了基于最大生存周期的无线传感器网络数据融合算法。该算法将整个网络中的节点分成多个簇,并根据节点的传输范围,将每个簇中的节点均匀分布,每个节点根据自己的本地信息和剩余能量选择通信方式向簇头节点传输数据,从而形成传输数据的最短路径;并根据集中式TDMA(时分多址)调度模型,运用基于微粒群的Pareto优化方法,使得网络在完成规定的信息传输时每个节点耗费的平均时隙和平均能耗最优。仿真结果表明,上述算法不但可以最大化网络的生存时间,还可以有效的降低数据融合时间,减少网络延时。     

基于仿人体血管路径WSN故障容错路由算法研究

为提高无线传感器网络(WSN)故障容错性和传输稳定性,实现网络负载均衡,提出了一种仿人体血管路径的WSN故障容错路由算法.通过研究人体血管路径特性,将其引入到WSN故障容错路由设计中,在对网络节点分区域进行等级标定的基础上实行能耗均衡的静态分簇;运用改进的蚁群算法生成节点路径并计算各路径信息素值,以确定传输路径选择概率并建立仿血管拓扑结构路由.理论与仿真结果表明,此算法具有良好的性能.     

基于分簇的无线多媒体传感器网络数据聚合方案研究

该文提出了一种基于分簇的无线多媒体传感器网络(WMSNs)数据聚合方案(Cluster-based Data Aggregation Algorithm, CDAA)。利用新的分簇方法和数据聚合策略,CDAA可以有效延长网络生命期。根据多媒体节点数据采集的方向性和节点剩余能耗,该文提出新的无线多媒体传感器网络的分簇方法,并基于该分簇方法进行网内多媒体数据聚合。仿真结果表明,该方法能够有效减少冗余数据的传送,与LEACH, PEGASIS等传统WSNs路由协议和针对WMSNs的AntSensNet协议相比,在能耗均衡和节能方面表现出更好的性能。     

基于LEACH的WSN路由算法研究

为了提高无线传感器网络的可扩展性在其路由协议中通常会采用分簇技术。由于无线传感器网络由能量有限的节点组成,网络中节点的能量多为电池供电,因此高效节能以延长网络生命周期是无线传感器网络必须要充分考虑的问题。针对LEACH(低功耗自适应分簇)路由算法在簇首选择时存在的问题,提出一种改进建议。对改进算法利用仿真工具NS2进行仿真,并对节点存活率和网络能耗两个方面进行比较与分析,仿真结果表明,改进算法很好地延长了网络的寿命。     

无线传感器网络路由协议与改进

详细介绍无线传感器网络(WSN)的两种代表性协议:信息协商传感器(SPIN)协议和低能量自适应分簇路由(LEACH)协议的概念、原理和优缺点.提出路由协议中需要进一步解决的问题.改进的WSN路由算法应尽可能降低节点能耗.以延长网络生存时间.     

无线传感器网络恶意节点溯源追踪方法研究

 传感器节点可能被攻击者俘获用来发送大量虚假数据,从而耗尽整个网络的资源.本文提出一种实用的溯源追踪解决方案:基于概率包标记算法,每个节点按照一定概率标记其转发的包,标记信息填写于包头中的确定域,通过收集到足够多的数据包,汇聚节点能够重建一条到源节点的路径.本文证明了此方案能够应对所有类型的攻击,并针对基本标记方法的不足提出了两种改进标记方法.实验结果表明该溯源追踪解决方案是高效以及实用的.     

基于扇形分簇的无线传感器网络路由算法

无线传感网络中低功耗自适应聚类分簇(LEACH)路由算法等概率选取簇首节点,容易导致整个网络节点能量损耗出现极端化,减少网络生存时间。为此,提出一种针对簇首节点选取和分簇的改进LEACH算法。该算法把整个网络区域分为四个扇形区域,在每个区域内独立进行分簇路由;然后基站根据节点剩余能量和与基站的距离进行簇首节点选择,节点根据簇首节点和基站接收信号强度选择路由方式,以均衡网络能量消耗。仿真结果表明,改进LEACH算法的网络寿命是原有LEACH算法的150%,数据吞吐量提升了3倍。     

IP追踪中的自适应包标记

拒绝服务(DoS)攻击是目前最难处理的网络难题之一.最近,研究人员针对DoS攻击提出了多种方案,这些方案都各有优缺点.其中,由Savage等人提出的概率包标记方案受到了广泛的重视,也有不少的变种出现.在这一类的标记方案中,路由器以固定的概率选择是否标记一个数据包,这导致受害需要较多的数据包进行攻击路径的重构.本文提出一种自适应的标记策略,经实验验证受害者用较少的数据包即可重构攻击路径,这不仅为受害者及早地响应攻击争取了更多的时间,还限制了攻击者的伪造能力.     

区分服务中一种TCP友好的公平数据包标记算法

在DiffServ网络中,存在一些对带宽利用的不公平性。为了解决这些不公平性,该文提出了一种基于TSW的标记算法,称之为TCP友好的公平数据包标记算法,它能以一种近似按比例的方式在各个汇聚流之间公平地分配网络中的剩余带宽,同时还能在汇聚流内部对TCP流和UDP流进行公平的带宽分配。通过仿真试验对算法进行了验证,并与其它几种标记算法进行比较,结果证实本算法比其它几种算法具有更好的公平性。     

基于流量序列包标记的IP追踪方法

文章分析了分布式拒绝服务(DDos)攻击的特点和现有的追踪方法,认为对攻击的追踪和辨伪应在网络边界入口和被攻击端进行.根据分布式精确协同较难的特点,提出了一种基于流量序列的包标记法(Flow Sequence Packet Marking,FSPM),用于追踪攻击子网的地址.讨论分析了该方法的算法、有效性、复杂度以及进一步的研究方向.     

基于有序标记的IP包追踪方案

包标记方案是一种针对DoS攻击提出的数据包追踪方案,由于其具有响应时间快、占用资源少的特点,近年来受到了研究者的广泛关注.但由于包标记方案标记过程的随机性,使得受害者进行路径重构时所需收到的数据包数目大大超过了进行重构所必需收到的最小数据包数目,从而导致重构误报率的提高和响应时间的增长.本文提出了一种基于有序标记的IP包追踪方案,该方案通过存储每个目标IP地址的标记状态,对包标记的分片进行有序发送,使得在DoS发生时,受害者重构路径所需收到的标记包的数目大大降低,从而提高了对DoS攻击的响应时间和追踪准确度.该算法的提出进一步提高了包标记方案在实际应用中的可行性.     

基于历史梯度平均方差缩减的协同参数更新方法

随机梯度下降算法(SGD)随机使用一个样本估计梯度,造成较大的方差,使机器学习模型收敛减慢且训练不稳定。该文提出一种基于方差缩减的分布式SGD,命名为DisSAGD。该方法采用历史梯度平均方差缩减来更新机器学习模型中的参数,不需要完全梯度计算或额外存储,而是通过使用异步通信协议来共享跨节点的参数。为了解决全局参数分发存在的“更新滞后”问题,该文采用具有加速因子的学习速率和自适应采样策略：一方面当参数偏离最优值时,增大加速因子,加快收敛速度;另一方面,当一个工作节点比其他工作节点快时,为下一次迭代采样更多样本,使工作节点有更多时间来计算局部梯度。实验表明：DisSAGD显著减少了循环迭代的等待时间,加速了算法的收敛,其收敛速度比对照方法更快,在分布式集群中可以获得近似线性的加速。     

光分组网节点结构研究的现状与演化趋势

介绍了光分组网和光分组交换技术,从技术角度分析了光分组交换方式的优点,比较并评价了具有代表性的几类光分组交换节点结构,展望了光分组交换的前景及光网络节点结构的演化趋势。     

MDPF: Minimum Distance Packet Forwarding for Search Applications in Mobile Ad Hoc Networks

This paper introduces a message forwarding algorithm for search applications within mobile ad hoc networks that is based on the concept of selecting the nearest node from a set of designated nodes. The algorithm, which is called Minimum Distance Packet Forwarding (MDPF), uses routing information to select the node with the minimum distance. The goal of the proposed algorithm is to minimize the average number of hops taken to reach the node that holds the desired data. Numerical analysis and experimental evaluations using the network simulation software ns2 were performed to derive the lower and upper bounds of the confidence interval for the mean hop count between the source node of the data request, on one hand, and the node that holds the desired data and the last node in the set of search nodes, on the other hand. In the experimental evaluation, the performance of MDPF was compared to that of Random Packet Forwarding (RPF) and Minimal Spanning Tree Forwarding (MSTF). The results agreed with the numerical analysis results and demonstrated that MDPF offers significant hop count savings and smaller delays when compared to RPF and MSTF.     

Energy-optimized cluster head selection based on enhanced remora optimization algorithm in underwater wireless sensor network

With the technological advancements, wireless sensor network (WSN) has played an impeccable role in monitoring the underwater applications. Underwater WSN (UWSN) is supported by WSN but subjected to data dissemination in an acoustic medium. Due to challenging conditions in underwater scenario, the limited battery resources of these sensor nodes stem to a crucial research problem that needs to address the energy-efficient routing in UWSN. In this research work, we intend to propose an energy-optimized cluster head (CH) selection based on enhanced remora optimization algorithm (ECERO) in UWSN. Since CH devours the maximum energy among the nodes, we perform selection of CH based on EROA while considering energy, Euclidean distance from sink, node density, network's average energy, acoustic path loss model and lastly, the adaptive quantity of CHs in the network. Further, to reduce the load on CH node, we introduce the concept of sleep scheduling among the closely located cluster nodes. The proposed work improves the performance of recently proposed EOCSR algorithm by great magnitude which claims to mitigate hot-spot problem, but EOCSR still suffers from the same due to relaying a large magnitude of data.     

LSA based multi-instance learning algorithm for image retrieval

Focusing on the problem of natural image retrieval, based on latent semantic analysis (LSA) and support vector machine (SVM), a novel multi-instance learning (MIL) algorithm is proposed, where a bag corresponds to an image and an instance corresponds to the low-level visual features of a segmented region. Firstly, in order to transform every bag into a single sample, a collection of “visual-word” is generated by k-means clustering method to construct a projection space, then a nonlinear mapping is defined using these “visual-word” to embed each bag as a point in the projection space, thereby obtaining every bag's projection feature. Secondly, the matrix consisted of all the projection features of training bags is regarded as a term-document matrix, and LSA method is used to obtain the latent semantic feature of each bag. As a result, the MIL problem is converted into a standard single instance learning (SIL) problem that can be solved directly by SVM method. Experimental results on the COREL data sets show that the proposed method, named LSASVM-MIL, is robust, and its performance is superior to other key existing MIL algorithms.     

ROTEE: Remora Optimization and Tunicate swarm algorithm-based Energy-Efficient cluster-based routing for EH-enabled heterogeneous WSNs

Since the development of Wireless Sensor Networks (WSNs), the limited battery of the sensor nodes has been an unavoidable concern. Hence, to keep the WSNs operational for a longer possible duration, the recharging of node's battery through harvesting the ambient energy from surroundings (for an example, solar energy) has been proposed. In this work, we focus not only on utilizing the energy harvesting (EH)-enabled sensor nodes for routing purposes but also introduce a novel hybrid optimization ROATSA that uses Remora Optimization Algorithm (ROA) and Tunicate Swarm Algorithm (TSA) for energy-efficient cluster-based routing. The proposed work is termed as ROA and TSA-based Energy-Efficient Cluster-based Routing for EH-enabled WSN (ROTEE). Hybrid ROATSA is chosen due to enhanced convergence and exploitation capabilities. To reduce the financial burden on the network, we use only four EH-enabled nodes and locate them at each periphery of the network, equidistant to each other and the other nodes are 3-level energy heterogeneous sensor nodes. The selection of cluster head (CH) is optimized through ROATSA by considering profile index of each node by evaluating them at energy, distance, load balancing, node density, the delay involved, and network's average energy. The proposed work ROTEE shows supreme performance against the recently proposed clustering techniques.