认知无线网络中基于遗传算法的异构信道频谱感知策略

在具有异构信道的认知无线网络环境中,各异构信道具有不同带宽、不同感知时间、不同使用概率、不同稳定度。在给定的时间槽内,感知的信道数目不同,以及感知的信道不同,频谱感知获得的带宽也不同,并且频谱感知耗费的时间也不同,传输的时间也不同。因此,针对如何选择待感知的信道,提高次用户获得的带宽,以及用于传输的时间,提出基于遗传算法的异构信道频谱感知策略GAHS(Genetic Algorithm based on Heterogeneous channels spectrum Sensing strategy)。仿真结果表明GAHS能提高节点获得的带宽和用于传输的时间。     

网络自适应传输控制策略研究

针对流媒体传输中的拥塞问题，对网络自适应传输控制进行了研究，结合RTP协议的传输特性，提出了一种基于带宽的网络自适应传输策略并阐述了这种策略的控制机理。该策略能够根据网络资源状况动态调节传输速率、匹配网络带宽，实现了网络自适应传输。     

民航桥载设备运行监测网络吞吐量优化设计

民航桥载设备(包括桥载电源和桥载空调)运行监测网络辅助采用物联感知网IOT(Internet of Things)对影响航空安全的桥载设备工作参数和状态进行采集和传输。对于重要数据的低时延传输要求,采用保证时隙GTS(Guaranteed Time Slot)机制保障传输实时性。但在利用GTS传输用户数据过程中存在不同程度的GTS时长浪费、网络带宽利用率低和吞吐量小等问题。针对这些问题,采取对时隙进行细分且利用背包问题的求解方法对GTS申请进行合理响应,使得GTS带宽得到更好地利用,同时提高GTS时段内的吞吐量。仿真结果表明,采取时隙细分且按背包问题求解GTS的分配策略,在保证网络原有的竞争访问时长、低时延和低功耗性能的情况下,提高GTS网络带宽利用率和网络吞吐量。     

基于QoE感知的中继卫星带宽资源分配方法研究

为改善数据中继卫星的带宽分配对负载传输效率的影响，提出了一种基于用户体验质量QoE感知的中继卫星带宽资源分配方法。考虑到空间信息网络高动态链路情况下传输的可靠性，建立基于多个静止同步轨道卫星协同接入模型，研究了低轨道用户卫星进行多路数据上传的优势特点。通过设置链路效能函数，其将用户QoE满意度量化为链路反馈和链路成本，并用库恩塔克方法求解，实现最大化系统效能和带宽公平分配。研究结果表明，与Load-Balancing和Best-SNR-First方法相比，该方法不仅保证了用户QoE满意度，还提升了网络的接入性能。     

面向富互联网应用的数据传输策略研究

针对富互联网应用系统需要传输较大数据量而导致速度慢这一问题,研究了一种基于上下文感知的数据传输策略,通过感知用户的浏览记录和当前行为,系统可以自动预测用户下一步对数据的需求,并充分利用用户交互过程中的网络空闲时间传输数据,以提高系统的响应性.实验结果表明,该数据传输策略大大缩短了用户的等待时间,特别是在比较复杂的场景中,效果更为显著.     

基于萤火虫算法的移动边缘计算网络带宽资源优化策略

由于移动边缘计算网络在边缘位置部署,在多用户并发的情况下带宽资源优化策略容易出现高计算负荷,降低带宽资源优化的效果。为了解决这一问题,提出基于萤火虫算法的移动边缘计算网络带宽资源优化策略。在服务器之间数据连续传输的情况下,确定网络内用户分布情况,计算网络运行需要消耗的能量,利用萤火虫算法建立以带宽资源为中心的数学模型,以移动边缘计算网络的各项参数作为依据,对数学模型求解,得到最优解后,以用户最大收益为目标部署优化策略。实验结果表明:提出的基于萤火虫算法的带宽资源优化策略计算延迟小,网络带宽资源优化效能高,整体计算性能得到了明显提升。     

BDNC:基于网络编码的车载网商业消息传输协议

通常采用网络编码技术提高VANETs中商业应用的带宽利用率。然而,网络编码会引入额外的消息传输时延,这就存在带宽利用率和消息传输时延的权衡问题。为此,针对车载商业应用中的消息传输,提出了面向带宽和基于网络编码时延的消息传输协议(BDNC)。当两个源节点同时向同一区域广播消息时,转发节点(Relay)就利用网络编码技术降低重播次数和带宽消耗。当转发节点接收到消息时,它有两个选择:等待编码机会,节省带宽;或直接转发消息,降低时延。BDNC采用了两个不同的时延控制策略:缓冲区域控制(BSC)策略和时间控制(TCS)策略,以降低传输消息每一跳的时延和提高带宽利用率。仿真结果表明,两个策略较好地控制时延,且带宽利用率提高了38%。     

无线传感网中三维覆盖控制和节点调度的研究

无线传感器网络中采用二元感知模型方法计算休眠冗余节点具有局限性和不精确性.本文考虑到实际应用环境对节点感知能力的影响以及传输过程中的路径损耗问题,对监测区域采用三维网格建模,并在概率感知模型的基础上提出一种概率感知衰减模型,同时对当前二维覆盖控制算法存在的各种局限性,提出了一种基于概率的分阶段冗余唤醒策略.仿真结果表明,该算法均衡了网络节点能量,实现了网络能耗的有效性,延长了网络的生存期.     

面向InfiniBand数据中心的区别化传输层带宽划分机制

数据中心已经成为提供资源和服务的重要平台,在虚拟化基础上部署的多种应用对网络性能提出了更加严格和多样化的需求,从而促使改变以传统TCP为中心的传输层结构.文中从数据中心环境的特性出发,归纳得到改进数据中心网络传输层设计的3个方向.在深入剖析InfiniBand(IB)互联传输优化相关技术的基础上,文中提出了一种更好的支持网络可感知、应用可区分、策略可管理的数据中心网络新型传输控制模型.文中对IB传输控制核心的IB拥塞控制算法及其参数进行了深入的分析,证明了在流级通过区别化的拥塞控制参数配置能够实现成比例的带宽划分.最后,在真实系统中的实验验证了通过参数的动态配置能够保持租户级的公平带宽共享,更好支持云数据中心的带宽隔离需求.     

异构边缘网络中内容扩散策略及优化

通过内容扩散系统在边缘网络中部署热门内容的副本可以有效缓解骨干网络的负载压力。因为网络边缘设备的能力和其所属用户的兴趣可能存在差异,内容扩散策略需要充分考虑边缘网络的异构性。本文首先提出一种分层的边缘网络拓扑和去中心化的内容扩散策略实现热门内容副本的扩散。基于设备的能力、负载和兴趣特征,本文进一步优化内容扩散策略,包括：基于兴趣的扩散目标调整策略、启发式的缓存节点选择策略、覆盖率感知的传输内容选择和传输负载感知的接收节点选择策略。实验结果表明优化后的扩散策略在边缘网络中拥有更稳定的高服务成功率,提高了约25.5%的本地获取服务的概率,减少了约7.6%的完成扩散所需时间。     

面向上下文感知计算的贝叶斯网络结构自学习算法的研究

通过对上下文感知计算中上下文特点的详细分析,提出一种面向上下文感知计算的通用贝叶斯网络结构自学习方法。该方法能在足够实例数据的支撑下自动对上下文感知计算中上下文之间的关系进行学习,进而形成贝叶斯网络结构,用于从低层上下文向高层上下文的演化。通过对上下文感知计算中上下文的层次化特点的有效利用,该方法对贝叶斯网络自学习方法进行了有效优化。研究分析表明,该方法能显著降低贝叶斯网络学习过程中的时间复杂度。     

Context-aware inference in ubiquitous residential environments

Residential ubiquitous computing environments have focused on interconnecting environmental controls using a home network and the management of sensor data rather than a context-aware inference. Building upon previous research, this paper proposes a novel sensor-based context-aware system with a focus on three inference processes: rule, inference and pattern driven. Using pattern data derived from five families for a working week and establishing rules and inferences, the proposed context-aware system is demonstrated using electrical lighting as an example. The processes to develop an intelligent house are described so as to ameliorate personalized services in response to inhabitants’ needs.     

Recognizing multi-modal sensor signals using evolutionary learning of dynamic Bayesian networks

Multi-modal context-aware systems can provide user-adaptive services, but it requires complicated recognition models with larger resources. The limitations to build optimal models and infer the context efficiently make it difficult to develop practical context-aware systems. We developed a multi-modal context-aware system with various wearable sensors including accelerometers, gyroscopes, physiological sensors, and data gloves. The system used probabilistic models to handle the uncertain and noisy time-series sensor data. In order to construct the efficient probabilistic models, this paper uses an evolutionary algorithm to model structure and EM algorithm to determine parameters. The trained models are selectively inferred based on a semantic network which describes the semantic relations of the contexts and sensors. Experiments with the real data collected show the usefulness of the proposed method.     

支持感知上下文网络操作平台中间件的研究

普适计算的出现将大大改变人类与各种设备交互的方式。支持感知上下文计算的网络操作平台是普适计算环境下重要的应用支撑平台,为基于传感器网络的应用提供操作服务。操作平台以中间件为核心,采用XMLWebServices技术。文章以满足包括智能家居普适上下文应用需求的网络操作平台为基础,介绍该网络操作平台的机制、层次结构和数据模型的研究,并实现了基于该平台模型的XMLWebServices中间件应用方案。     

Analysis of trajectories in mobile networks based on data about the network proximity

This paper is devoted to location-based mobile services. The movement (trajectory) data extraction from logs related to network proximity is considered. Usually, this type of pattern extraction (search) relates to trajectory databases containing geoposition information. We consider a model of context-aware computing (a context-aware browser) based on network proximity. A mobile phone is considered as a proximity sensor. The geoposition information is replaced with the network proximity. Any existing or specially created network node can be regarded as a sensor of presence that provides access to dynamically determined network content. The disclosure of the content depends on the set of rules describing the conditions of network’s proximity. An algorithm is given for calculating the trajectories in mobile networks based on information about the network’s proximity.     

Conceptual data modeling for realizing context-aware services

Recently, context-aware applications, which provide customized services using personalized contextual information, have gained much attention. However, core IT technologies for realizing context-aware services are not sufficient to support the actual development of context-aware applications. Especially, it is very hard to find studies on context-aware data modeling that can provide different responses even for the same information according to users’ situation. In practice, most context-aware services are managing contextual information in the phase of application development in a rather ad-hoc way. Unfortunately, this approach has an obvious limitation of expansibility when the amount of rapidly increasing contextual information is considered. In this paper, we first define the concept of contextual dependency among data objects. Second, we provide a context-aware data modeling methodology for modeling contextual information into the Entity-Relationship Diagram (ERD, for short) based on contextual dependency. Finally, we design a simple scenario for context-aware services, construct an ERD from the scenario based on the discovered design rules, and implement a prototype of the service.     

基于深度学习的情境感知推荐系统研究进展

传统的推荐系统存在数据高度稀疏、冷启动及用户偏好建模难等问题,而把情境信息融入推荐系统中能有效缓解此类问题.深度学习技术已经成为人工智能领域研究热点,把深度学习应用在情境感知推荐系统当中,为推荐领域的研究带来新的机遇与挑战.本文从情境感知推荐系统相关概念出发,综合整理国内外研究相关文献,介绍深度学习技术融入情境感知推荐系统相关应用模型,提出了基于深度学习的情境感知推荐系统研究的不足以及对未来的展望.     

An easy-to-use 3D visualization system for planning context-aware applications in smart buildings

With the proliferation of wireless sensor network technologies, the context-aware applications of smart environments have become more and more popular. However, time-consuming and labor-intensive works hamper the development of smart building applications. Therefore, application developers need the proper software platforms to efficiently design the applications. Much research currently proposed the design approaches based on low-level concerns and the real-world deployments based on high-level programming abstraction are rare. Accordingly, the pressing issue is how to offer the easy-to-use tools with acceptable performance to rapidly and easily design the application scenarios for non-technical application users. Therefore, this study focuses on the application-layer simulation to propose a Visualization System of Context-aware Application Scenario Planning (VS-CaSP) for assisting non-technical developers and end-users in rapidly and easily designing the application scenario of smart buildings and in performing the acceptable and predictable simulation and evaluation. VS-CaSp applies rule-based and 3D visualization techniques to offer a 3D authoring environment integrated with real Zigbee sensor devices, where designers are able to rapidly construct immersive 3D buildings and easily plan the context-aware application scenario via GUI tool based on proposed three-tier rule hierarchy, to visually simulate and verify the planned scenario via virtual and real sensor devices, and to repeatedly modify the control strategies to enhance the deployment effectiveness. The experimental results show that VS-CaSP is easy to use for the support of quickly designing smart building applications, but not professional enough for developments of 3D modeling, animation, and rule expression. Accordingly, it is workable and expected to prove beneficial to non-technical users.     

Intelligent video surveillance system: 3-tier context-aware surveillance system with metadata

This paper presents an intelligent video surveillance system with the metadata rule for the exchange of analyzed information. We define the metadata rule for the exchange of analyzed information between intelligent video surveillance systems that automatically analyzes video data acquired from cameras. The metadata rule is to effectively index very large video surveillance databases and to unify searches and management between distributed or heterogeneous surveillance systems more efficiently. The system consists of low-level context-aware, high-level context-aware and intelligent services to generate metadata for the surveillance systems. Various contexts are acquired from physical sensors in monitoring areas for the low-level context-aware system. The situation is recognized in the high-level context-aware system by analyzing the context data collected in the low-level system. The system provides intelligent services to track moving objects in Fields Of View (FOVs) and to recognize human activities. Furthermore, the system supports real-time moving objects tracking with Panning, Tilting and Zooming (PTZ) cameras in overlapping and non-overlapping FOVs.     

基于上下文感知的RBAC动态访问控制研究

针对传统RBAC（Role Based access contr01）模型不能适应当前网络和分布式计算应用环境的分布、异构和动态性的特点,为了适应应用环境的动态要求,将上下文感知和上下文约束加入PBAC模型,提出了上下文感知的动态访问控制模型CDAC（Context—aware Dynamic Access Control Model）,并给出了该模型的一个应用实例。该模型不仅继承了PBAC的优点,还利用上下文感知和上下文约束,实现了角色和权限的实时的动态管理。