卫星在役考核相关问题研究

各类卫星在现代战争中发挥着重要和关键作用。我国传统的卫星试验和鉴定具有很强的科研试验性质,承制单位和用户重点关注的是性能检验,较少考虑作战效能和部队适用性等指标。为推动卫星装备建设和应用,装备试验鉴定部门初步组织开展了相应的在役考核工作。文章对比了美军装备作战试验与鉴定、我国作战试验与在役考核的概念;提出卫星在役考核的六大特点;分析卫星在役考核的组织方式、组织机构和组织保障的不足,提出相应发展建议;研究卫星在役考核指标体系设计、分析评估方法、问题反馈与整改等方面的不足,并提出相应措施建议。     

大力推进新时代中国特色科技兴军

习主席站在时代发展与战略全局的高度,着眼军队建设全局,深刻阐述了科技兴军在强军征程中的重要作用,明确了科技兴军的举措抓手,丰富和发展了新时代党的强军思想理论体系。必须毫不动摇地坚持走中国特色科技兴军之路,为实现党在新时代的强军目标提供强有力的支撑。     

基于窗口预测匹配的序列图像点目标轨迹检测算法

序列图像中运动点目标轨迹的实时检测算法,是目标识别、实时跟踪处理系统中的关键算法之一。在分析不同层次处理方法的基础上,给出了一种基于预测、窗口匹配的点轨迹预测—匹配检测算法,并根据仿真实验的效果,指出此算法可有效地对低信噪比序列图像运动点目标轨迹进行实时检测,具有多目标轨迹检测能力。     

BIT系统的三态马尔可夫模型分析

BIT技术是改善系统测试性和诊断能力的重要途径,但是较高的虚警率一直是阻碍BIT广泛应用的一个重要原因。研究过程中将系统状态划分为正常、间歇、故障三种状态,建立了马尔可夫模型,分析、对比了三态模型与两态模型的关系。理论分析结果表明:基于三态模型的BIT,在提高故障检测率的同时,还可以较大程度地降低虚警率。     

Dynamic inventory and pricing models for competing retailers

We address infinite‐horizon models for oligopolies with competing retailers under demand uncertainty. We characterize the equilibrium behavior which arises under simple wholesale pricing schemes. More specifically, we consider a periodic review, infinite‐horizon model for a two‐echelon system with a single supplier servicing a network of competing retailers. In every period, each retailer faces a random demand volume, the distribution of which depends on his own retail price as well as those charged by possibly all competing retailers. We also derive various comparative statics results regarding the impact several exogenous system parameters (e.g., cost or distributional parameters) have on the equilibrium decisions of the retailers as well as their expected profits. We show that certain monotonicity properties, engrained in folklore as well as in known inventory models for centralized systems, may break down in decentralized chains under retailer competition. Our results can be used to optimize the aggregate profits in the supply chain (i.e., those of the supplier and all retailers) by implementing a specific wholesale pricing scheme. © 2003 Wiley Periodicals, Inc. Naval Research Logistics, 2004.     

Dual sourcing: Responsive hedging against correlated supply and demand uncertainty

This article analyzes dual sourcing decisions under stochastically dependent supply and demand uncertainty. A manufacturer faces the trade‐off between investing in unreliable but high‐margin offshore supply and in reliable but low‐margin local supply, where the latter allows for production that is responsively contingent on the actual demand and offshore supply conditions. Cost thresholds for both types of supply determine the optimal resource allocation: single offshore sourcing, single responsive sourcing, or dual sourcing. Relying on the concept of concordance orders, we study the effects of correlation between supply and demand uncertainty. Adding offshore supply to the sourcing portfolio becomes more favorable under positive correlation, since offshore supply is likely to satisfy demand when needed. Selecting responsive capacity under correlated supply and demand uncertainty is not as straightforward, yet we establish the managerially relevant conditions under which responsive capacity either gains or loses in importance. Our key results are extended to the broad class of endogenous supply uncertainty developed by Dada et al. [Manufact Serv Operat Mange 9 (2007), 9–32].© 2012 Wiley Periodicals, Inc. Naval Research Logistics, 2012     

含噪稀疏信号重构的l0范数期望值最小化方法

压缩感知理论是对信号压缩的同时进行感知的新理论,而如何通过有限的测量值重构稀疏信号是压缩感知理论中的核心问题。针对测量值受噪声污染的含噪稀疏重构问题,提出了近似l0范数期望值最小化方法。该算法基本思想是将含噪稀疏重构问题转化为近似l0范数期望值最小化问题,并利用噪声的统计特征将随机最优化问题化简为常规的最优化问题,然后采用最速下降法求解。数值仿真表明,本文提出的方法具有更好的重构精度,且计算量较小。     

基于能量度量的星载AIS信号自适应码元同步抽取算法

基带信号码元同步抽取是目前星载AIS(Automatic Identification System)信号处理中的一个难点问题,针对过零检测和迟门早门等现有算法存在的不足,提出了基于能量度量的基带信号自适应码元同步抽取算法。将基带信号根据采样率和码速率的比值分割成相同的段的组合,之后分别计算每个段内相应样点的能量度量,最后选择使能量度量最大时的样点集作为最优同步抽取样点,该样点集具有最大信噪比。仿真结果验证了该算法误码率优于过零检测和迟门早门算法。     

垃圾评论自动过滤方法

针对互联网上存在的大量垃圾评论,提出一种基于电阻网络的垃圾评论检测方法,该方法用电阻距离来度量评论之间的上下文语义相似性,把整个评论数据表示成一个电阻网络,把垃圾评论当作该网络上的语义离群点来处理,根据网络节点对电阻网络平均电能消耗的影响,建立电离群因子来度量数据的离群程度,以此来识别垃圾评论。实验证明了该方法的有效性,在多个数据集上取得了较好的效果。     

Screening and selection procedures with control variates and correlation induction techniques

We consider the problem of identifying the simulated system with the best expected performance measure when the number of alternatives is finite and small (often < 500). Recently, more research efforts in the simulation community have been directed to develop ranking and selection (R&S) procedures capable of exploiting variance reduction techniques (especially the control variates). In this article, we propose new R&S procedures that can jointly use control variates and correlation induction techniques (including antithetic variates and Latin hypercube sampling). Empirical results and a realistic illustration show that the proposed procedures outperform the conventional procedures using sample means or control variates alone. © 2012 Wiley Periodicals, Inc. Naval Research Logistics, 2012