基于多特征子空间与核学习的行人再识别

行人再识别指的是在无重叠视域多摄像机监控系统中, 匹配不同摄像机视域中的行人目标.针对当前基于距离测度学习的行人再识别算法中存在着特征提取复杂、训练过程复杂和识别效果差的问题, 我们提出一种基于多特征子空间与核学习的行人再识别算法.该算法首先在不同特征子空间中基于核学习的方法得到不同特征子空间中的测度矩阵以及相应的相似度函数, 然后通过比较不同特征子空间中的相似度之和来对行人进行识别.实验结果表明, 本文提出的算法具有较高的识别率, 其中在VIPeR数据集上, RANK1达到了40.7%, 且对光照变化、行人姿态变化、视角变化和遮挡都具有很好的鲁棒性.     

基于B+树的发布/订阅并行匹配算法

现有的消息匹配算法不能很好地满足大规模分布式发布/订阅系统中海量数据的有效分发。为此,提出一种高效的发布/订阅并行匹配算法。采用B+树的存储结构,订阅消息按属性-值进行空间划分,建立多级索引订阅属性树,消息匹配采用改进的B+树查询方法,结合计数的思想实现快速消息匹配。实验结果表明,该算法在大规模分布式发布/订阅系统中,消息匹配时间短,订阅匹配树生成迅速,匹配效果更好。     

不同仰角条件下的临近空间静止信道建模与仿真

在临近空间通信系统中,降雨、大气、闪烁等都会引起空地链路信道质量的恶化。通过分析Ka波段临近空间通信信道的电波传播特性,建立了Ka波段临近空间静止通信信道模型。分析了空地链路中仰角变化时,降雨对信号衰落的影响,估算了不同仰角下的相关参数值,并对不同仰角下Ka波段临近空间静止信道性能进行仿真,验证了模型的正确性,为临近空间链路分析及后续研究工作提供了平台。     

一种不确定数据流子空间聚类算法

针对不确定数据流上的聚类问题提出一种不确定数据流子空间聚类算法UDSSC.该算法使用滑动窗口机制接收新到达的数据,剔除陈旧的数据;还引入子空间簇生成策略和新型离群点机制;系统建立了三个缓冲区分别存储新到来的元组、要进行聚类的元组和离群点元组,以此获得高质量的聚类结果.实验表明,UDSSC算法与同类型算法相比,具有更好的聚类效果、更低的时间复杂度和更强的扩展性.     

基于距离度量的实体识别算法

传统的实体识别中,往往是利用字符串相似性函数来计算元组对在每个属性值上的相似度从而来判断其总的相似性(例如,元组对的相似性等于每个属性值上的相似度的加权求和)。然而这一类相似性测度不能够反映属性值内部不同的词在元组对相似性计算中的不同重要性。由于不能区分哪些词对元组对匹配更重要,就导致仍然存在某些匹配的元组相似性不高,而不匹配的元组相似性高的情况,故很难将匹配元组对和不匹配元组对有效区分开。为了解决这个问题,提出了以词为特征的距离度量函数,设计了基于词特征的距离度量学习算法,和基于距离度量的实体识别算法。扩展性实验对所提出的算法的有效性进行了验证。     

基于正交分解的递推子空间闭环辨识方法

为实现闭环系统在线辨识,提出递推正交分解闭环子空间辨识方法(RORT)。首先,根据闭环系统状态空间模型和数据间投影关系,构建确定-随机模型,并利用GIVENS变换实现投影向量的递推QR分解;然后,引入带遗忘因子的辨识算法,构建广义能观测矩阵的递推更新形式,以减少子空间辨识算法中QR分解和SVD分解的计算量;最后,针对某型号陀螺仪闭环系统进行实验。实验结果表明, RORT法的辨识拟合度高于91%,能够对陀螺仪闭环系统模型参数进行在线监测。     

并行子空间优化在无人机总体设计中的应用

飞机设计是一个多学科的复杂的系统工程,各个学科通常相互影响、相互耦合.这使得飞机设计过程日趋复杂,设计周期越来越长,开发成本越来越高,而并行子空间优化(CSSO)是解决这些问题的一种有效方法.文中对基于神经网络响应面的并行子空间优化算法及其在无人机总体方案设计优化中的应用进行了研究.并行子空间优化算法将多学科耦合的无人机设计优化问题分解为不同的子空间问题,在不同的子空间中建立各自的神经网络响应面,通过响应面完成各子空间之间的数据交换与协调,以此来逼近设计空间最优解.应用结果表明,CSSO算法能有效地应用于无人机总体方案优化设计.     

分布式元组空间协同模型的设计与描述

元组空间是一种结构化的分布式共享存储编程方式,使用一个共享的元组空间进行生成式通信。该文针对集中式元组空间的性能瓶颈、单点失效问题及可扩展性差等不足,提出一种混合式的分布式元组空间的体系架构,给出基于π-演算的分布计算演算模型,并对模型的语法和操作语义进行讨论,通过钩互模拟给出了系统的性质。     

基于人工鱼群微细分解的先验未知像素点修复算法

由于未知像素点先验信息缺失,因此模块匹配和边缘结构信息未知,全息修复困难。传统方法采用子空间特征信息多维搜索方法未能实现对图像纹理的微细结构信息的模板匹配,效果不好。引入人工鱼群算法,提出一种基于人工鱼群微细分解和亮度补偿的先验未知像素点全息修复算法,即采用子空间特征信息多维搜索方法进行先验未知像素点置信度的更新,以保持被修复的图像破损区域的连续性。构建人工鱼群算法的图像微细分解模型,结合边缘特征点亮度补偿策略,来实现对先验未知像素点的图像信息修复改进。实验结果表明,改进的图像修复算法具有良好的视觉效果,修复时间和计算开销较少,提高了稳定性和收敛性,图像修复后的信噪比误差较小,保持在6%以内,因此该算法的性能优越。     

有效的子空间支配查询算法——Ranking-k

针对Top-k dominating查询算法需要较高的时空消耗来构建属性组合索引,并且在相同属性值较多情况下的查询结果准确率低等问题,提出一种通过B⁺-trees和概率分布模型相结合的子空间支配查询算法--Ranking-k算法.首先,采用B⁺-trees为待查找数据各属性构建有序列表;然后,采取轮询调度算法读取skyline准则涉及到的有序列表,生成候选元组并获得k组终结元组;其次,根据生成的候选元组和终结元组,采用概率分布模型计算终结元组支配分数.迭代上述过程优化查询结果,直到满足条件为止.实验结果表明:Ranking-k与基本扫描算法(BSA)相比,查询效率提高了94.43%;与差分算法(DA)相比,查询效率提高了7.63%;与早剪枝Top-k支配(TDEP)算法、BSA和DA相比,查询结果更接近理论值.     

Privacy-Preserving Tuple Matching in Distributed Databases

We address the problems of privacy-preserving duplicate tuple matching (PPDTM) and privacy-preserving threshold attributes matching (PPTAM) in the scenario of a horizontally partitioned database among N parties, where each party holds a private share of the database's tuples and all tuples have the same set of attributes. In PPDTM, each party determines whether its tuples have any duplicate on other parties' private databases. In PPTAM, each party determines whether all attribute values of each tuple appear at least a threshold number of times in the attribute unions. We propose protocols for the two problems using additive homomorphic cryptosystem based on the subgroup membership assumption, e.g., Paillier's and ElGamal's schemes. By analysis on the total numbers of modular exponentiations, modular multiplications and communication bits, with a reduced computation cost which dominates the total cost, by trading off communication cost, our PPDTM protocol for the semihonest model is superior to the solution derivable from existing techniques in total cost. Our PPTAM protocol is superior in both computation and communication costs. The efficiency improvements are achieved mainly by using random numbers instead of random polynomials as existing techniques for perturbation, without causing successful attacks by polynomial interpolations. We also give detailed constructions on the required zero-knowledge proofs and extend our two protocols to the malicious model, which were previously unknown.     

A Fault-tolerant model for tuple space coordination in distributed environments

In distributed systems, tuple space is one of the coordination models that significantly maximizes system performance against failure due to its space and time decoupling features. With the growing popularity of distributed computing and increasing complexity in the network, host and link failure occurs frequently, resulting in poor system performance. This article proposes a fault-tolerant model named Tuple Space Replication (TSR) for tuple space coordination in distributed environments. The model introduces a multi-agent system that consists of multiple hosts. Each host in a multi-agent system comprises an agent space with a tuple space for coordination. In this model, we introduce three novel fault-tolerant algorithms for tuple space primitives to provide coordination among hosts with tolerance to multiple links and hosts failure. The first algorithm is given for out() operation to insert tuples in the tuple space. The second algorithm is presented for rdp() operation to read any tuple from the tuple space. The third algorithm is given for inp() operation to delete or withdraw tuples from the tuple space. These algorithms use less number of messages to ensure consistency in the system. The message complexity of the proposed algorithms is analyzed and found O(n) for out(), O(1) for rdp(), and O(n) for inp() operations which is comparable and better than existing works, where n is the number of hosts. The testbed experiment reveals that the proposed TSR model gives performance improvement up to 88%, 70.94%, and 63.80% for out(), rdp(), and inp() operations compared to existing models such as FT-SHE, LBTS, DEPSPACE, and E-DEPSPACE.     

Linda meets Unix

The limitations of the shared-memory and distributed-memory models for explicit parallel programming are discussed and a new model, the Linda parallel communication paradigm which was designed specifically for parallel programming, is examined. Processes communicate in Linda by way of a shared data space called tuple space which acts something like an associative memory, since tuples are identified by matching on a key rather than using a specific address. This model is adapted for use as the basis of a new class of operating systems and a specific instance, QIX, is presented. Like Linda, this operating system model can support both the shared-memory and the distributed-memory styles of programming. Thus, it provides the benefits of both, while avoiding hardware dependencies. QIX also incorporates a novel scheme for name resolution that is easier to use than other methods and provides significant benefits in the operating system and it directly supports communication between programs written in different languages     

Providing data confidentiality against malicious hosts in Shared Data Spaces

This paper focuses on the protection of the confidentiality of the data space content when Shared Data Spaces are deployed in open, possibly hostile, environments. In previous approaches, the data space content was protected against access from unauthorised application components by means of access control mechanisms. The basic assumption is that the hosts (and their administrators) where the data space is deployed have to be trusted. When such an assumption does not hold, then encryption schemes can be used to protect the data space content from malicious hosts. However, such schemes do not support searching on encrypted data. As a consequence, performing retrieval operations is very expensive in terms of resource consumption. Moreover, in these schemes applications have to share secret keys requiring a very complex key management. In this paper, we present a novel encryption scheme that allows tuple matching on completely encrypted tuples. Since the data space does not need to decrypt tuples to perform the search, tuple confidentiality can be guaranteed even when the data space is deployed on malicious hosts (or an adversary gains access to the host). Our scheme does not require authorised components to share keys for inserting and retrieving tuples. Each authorised component can encrypt, decrypt, and search encrypted tuples without knowing other components’ keys. This is beneficial inasmuch as it simplifies the task of key management. An implementation of an encrypted data space based on this scheme is described and some preliminary performance results are given.     

On the collective sort problem for distributed tuple spaces

In systems coordinated with a distributed set of tuple spaces, it is crucial to assist agents in retrieving the tuples they are interested in. This can be achieved by sorting techniques that group similar tuples together in the same tuple space, so that the position of a tuple can be inferred by similarity. Accordingly, we formulate the collective sort problem for distributed tuple spaces, where a set of agents is in charge of moving tuples up to a complete sort has been reached, namely, each of the N tuple spaces aggregate tuples belonging to one of the N kinds available. After pointing out the requirements for effectively tackling this problem, we propose a self-organizing solution resembling brood sorting performed by ants. This is based on simple agents that perform partial observations and accordingly take decisions on tuple movement. Convergence is addressed by a fully adaptive method for simulated annealing, based on noise tuples inserted and removed by agents on a need basis so as to avoid sub-optimal sorting. Emergence of sorting properties and scalability are evaluated through stochastic simulations.     

Programming mobile context-aware applications with TOTAM

In tuple space approaches to context-aware mobile systems, the notion of context is defined by the presence or absence of certain tuples in the tuple space. Existing approaches define such presence either by collocation of devices holding the tuples or by replication of tuples across all devices. We show that both approaches can lead to an erroneous perception of context. Collocation ties the perception of context to network connectivity which does not always yield the expected result. Tuple replication can cause that a certain context is perceived even if the device has left the context a long time ago. We propose a tuple space approach in which tuples themselves carry a predicate that determines whether they are in the right context or not. We present a practical API for our approach and show its use by means of the implementation of various mobile applications. Benchmarks show that our approach can lead to a significant increase in performance compared to other approaches.     

SecSpaces: a Data-driven Coordination Model for Environments Open to Untrusted Agent

In this paper we initiate an investigation about security problems which occur when exploiting a Linda-like data driven coordination model in an open environment. In this scenario, there is no guarantee that all the agents accessing the shared tuple space are trusted. Starting from the analysis of the few proposals already available in the literature, we present a novel coordination model which provides mechanisms to manage tuple access control. The first mechanism supports logical partitions of the shared repository: in this way we can restrict the access to tuples inside a partition, simply by limiting the access to the partition itself. The second mechanism consists of adding to the tuples some extra information which exploit asymmetric cryptography in order, e.g., to authenticate the producer of a tuple or to identify its reader/consumer. Finally, we support the possibility to define access control policies based on the kind of operations an agent performs on a tuple, thus discriminating between (destructive) input and (non-destructive) read operations.     

Dynamic routing of data stream tuples among parallel query plan running on multi-core processors

In this paper, a method for fast processing of data stream tuples in parallel execution of continuous queries over a multiprocessing environment is proposed. A copy of the query plan is assigned to each of processing units in the multiprocessing environment. Dynamic and continuous routing of input data stream tuples among the graph constructed by these copies (called the Query Mega Graph) for each input tuple determines that, after getting processed by each processing unit (e.g., processor), to which next processor it should be forwarded. Selection of the proper next processor is performed such that the destination processor imposes the minimum tuple latency to the corresponding tuple, among all of the alternative processors. The tuple latency is derived from processing, buffering and communication time delay which varies in different practical parallel systems.     

Incremental Evaluation of Sliding-Window Queries over Data Streams

Two research efforts have been conducted to realize sliding-window queries in data stream management systems, namely, query revaluation and incremental evaluation. In the query reevaluation method, two consecutive windows are processed independently of each other. On the other hand, in the incremental evaluation method, the query answer for a window is obtained incrementally from the answer of the preceding window. In this paper, we focus on the incremental evaluation method. Two approaches have been adopted for the incremental evaluation of sliding-window queries, namely, the input-triggered approach and the negative tuples approach. In the input-triggered approach, only the newly inserted tuples flow in the query pipeline and tuple expiration is based on the timestamps of the newly inserted tuples. On the other hand, in the negative tuples approach, tuple expiration is separated from tuple insertion where a tuple flows in the pipeline for every inserted or expired tuple. The negative tuples approach avoids the unpredictable output delays that result from the input-triggered approach. However, negative tuples double the number of tuples through the query pipeline, thus reducing the pipeline bandwidth. Based on a detailed study of the incremental evaluation pipeline, we classify the incremental query operators into two classes according to whether an operator can avoid the processing of negative tuples or not. Based on this classification, we present several optimization techniques over the negative tuples approach that aim to reduce the overhead of processing negative tuples while avoiding the output delay of the query answer. A detailed experimental study, based on a prototype system implementation, shows the performance gains over the input-triggered approach of the negative tuples approach when accompanied with the proposed optimizations     

Data Privacy in Tuple Space Based Mobile Agent Systems

More recently, distributed variants of tuple spaces have been proposed to exploit the Linda model for programming distributed applications over wide area networks, possibly exploiting code mobility. However, the flexibility of the shared tuple space model opens possible security holes; it basically provides no access protection to the shared data. In this paper we investigate some possible scenarios where mobile agents can benefit from our cryptographic tuple space based framework, CryptoKlava, and sketch how to possibly implement such agents in order to keep the privacy of items collected by the mobile agent during its itinerary. The functionalities of the framework are general enough to be applied to other Java frameworks using multiple distributed tuples spaces possibly dealing with code mobility.