微纳卫星姿态确定与控制半实物仿真系统设计

航天器姿态控制系统需要特殊的运行环境,在地面很难考核,这给系统可靠性带来一定的风险。针对微纳卫星的特点,设计并研制了一套面向微纳卫星的姿态确定与控制半实物仿真系统。该系统通过数字化模型模拟卫星姿态轨道运动、敏感器模型产生敏感器测量数据、执行器模型生成控制力矩、敏感器模拟器实现通信协议,最终实现姿态控制系统的全系统仿真。这套系统可以接入卫星控制系统回路,实现对姿控系统软件、硬件的考核,同时验证算法的性能。基于该系统,对天拓三号卫星姿控系统进行地面半实物仿真,并对比在轨试验数据,结果表明系统设计合理,仿真结果可信。     

基于DDS舰载火控系统通信

针对舰载火控系统通信需求,提出基于DDS的舰载火控体系结构,实现了系统松耦合,同时满足了实时性要求。通过发布订阅服务的设计,基于默认QoS策略,分别使用内置数据类型和自定义数据类型实现舰载火控系统的发布订阅功能。通过可靠性QoS策略的配置分析其对系统实时性和可靠性的影响。测试结果表明,DDS在舰载火控系统中有广阔的应用前景。     

Indian and Chinese espionage

India and China both have powerful spy networks; completely different in their approaches to espionage; both effective against their perceived enemies. China focuses first on internal threats, on Taiwan and Hong Kong, and then the US and Japan. India’s defense policy focuses on Pakistan and internal terrorist threats, and then on China. In reality, however, when it comes to spying on each other, both China and India suffer from incompetence and apathy – which endangers both their own security and regional stability. This article looks at how they spy on each other, and asks why and how they need to improve. The narrative also touches upon some of the individuals who are waging the spy war, from India’s wily spymaster Ajit Doval down to junior Chinese agents such as Wang Qing and Pema Tsering. The two countries are not friends. They have the largest territorial dispute in the world on their hands, covering an area the size of North Korea, and they have large armies facing each other along 4000 kilometers of frontier. But they also lay claim to the world’s two oldest and richest civilizations, with a rich history of exchange, and now with a combined population of 2.6 billion people and more than a quarter of the world’s economic output. If they cooperated, they could solve many of the world’s problems; but if they lurch into conflict, the potential consequences are terrifying to contemplate. Unfortunately, despite their geographical closeness, they do not know much about each other. They have few cultural interchanges, little diplomacy, few trade missions. They do not watch each other’s films, read each other’s books or listen to each other’s music. Chinese tourists would rather fly to New Zealand for their holidays than cross the border to India, and Indian students would rather study in Europe than China. China and India are neighbors that barely talk to each other. Most significantly, they do not spy on each competently. For countries that do not interact socially, defensive understanding is important for security – but China prefers the glamor of facing up to its Pacific and other maritime rivals such as the US and Japan. India, for its part, does talk a great deal about the China threat, but its resources and expertise are wrapped up in controlling its security threat from Pakistan and the Islamic world. When China and India do try to spy on each other, it is often without the benefit of a long-term focus or understanding. India has some very skilled operatives within the Research and Analysis Wing, but few that specialize in China. China has an enormous pool of resources spread across several government departments, including the Ministry of Public Security, and also has extensive facilities and manpower in the Joint Staff Department of the Central Military Commission (the JSD) and the new Strategic Support Force (the SSF). However, China’s intelligence services generally behave as if India is not worth spying on. Given that the two countries do not have the cultural or political machinery in place to understand each other, espionage and intelligence gathering is vital to ensure that miscalculations do not take place. This has been apparent over the last few years in stand-offs in the Himalaya, as well as top-level suspicions on each side about a variety of subjects including terrorism, covert operations in Sri Lanka and Burma, and the two countries’ nuclear weapons programs. Both countries do occasionally make efforts in espionage against each other, especially during sensitive periods such as the mountain stand-offs of 2014 and 2013 and during policy developments in nuclear warfare. In this article the author looks at actual spying incidents between the two countries, their methodologies, their staff, their technical capabilities, and how the act of spying, which is usually viewed as intrinsically adversarial, can be a force for good. The article relies on interviews with actual participants in intelligence from both countries as well as extensive use of contemporary online sources, and secondary analysis by both military and academic experts from China, India and NATO countries.     

压水堆运行中ΔT超温度/超功率保护分析

从核电站压水堆动力装置在运行中ΔT超温度 /超功率保护功能出发 ,介绍了压水堆ΔT超温度 /超功率保护系统 ,分析了确定极限限制线以及保护区的方法 ,并对此作出了评价 .     

Multiple common due dates

Common due date problems have been extensively discussed in the scheduling literature. Initially, these problems discussed finding a common due date for a set of jobs on a single machine. These single machine problems were later extended to finding the common due date on a set of parallel machines. This paper further extends the single machine problem to finding multiple common due dates on a single machine. For a basic and important class of penalty functions, we show that this problem is comparable to the parallel machine problem. © 2001 John Wiley & Sons, Inc. Naval Research Logistics 48: 293–298, 2001     

The selection allocation problem

The Selection Allocation Problem (SAP) is a single period decision problem which involves selecting profit‐maximizing (or cost‐minimizing) activities from various distinct groups, and determining the volume of those activities. The activities in each group are selected subject to the availability of that group's resource, which is provided by either pooling or blending raw inputs from several potential sources. Imbedded in the decision process is the additional task of determining how much raw input is to be allocated to each group to form the resource for that group. Instances of this problem can be found in many different areas, such as in tool selection for flexible manufacturing systems, facility location, and funding for social services. Our goal in this paper is to identify and exploit special structures in the (SAP) and use those structures to develop an efficient solution procedure. © 1999 John Wiley & Sons, Inc. Naval Research Logistics 46: 707–725, 1999     

基于虚拟仪器的便携式轴角测试系统

针对轴角测试在武器控制领域的重要性,以及满足现代工业对便携式数据采集与测量的要求,提出了一种采用虚拟仪器技术的便携式轴角信号测试方法.由便携式计算机和轴角信号采集器组成轴角测试硬件平台,通过EPP实现两者间的高速数据通讯,由软件完成对数据的采集、显示、存储以及分析处理.可实现四通道数据同时测量,单通道采集率可达400kbyte/s.该系统具有体积小,重量轻,功耗低、现场移动使用方便等特点,特别适用于科研、生产及维修领域的现场测试.     

面向大规模集群的并行I/O用户层配置优化策略

影响应用I/O性能的关键因素主要有三个层次:包括应用的I/O接口实现、体系结构和文件系统组件的性能以及应用的I/O参数配置。从应用I/O配置优化的视角,分析了大规模集群并行I/O的配置调优空间,在此基础上,给出了一套大规模集群并行I/O性能特征测试分析方法。基于该方法,在某国产超级计算集群上开展了一系列I/O测试分析来刻画系统的I/O性能特征,进而指导并行应用程序的I/O配置优化。基于优化后的配置参数,在两类典型的并行I/O场景中,针对某类生产应用程序,8192进程下的重启动数据写操作时间下降了15%,4096核的程序作业加载时间从10 min缩短到了5 s。     

面向大规模集群并行I/O的用户层配置优化策略研究

近年来,大规模并行场景下的I/O性能越来越受到应用科学家的关注。影响应用I/O性能的关键因素主要有三个层次：包括应用的I/O接口实现、体系结构和文件系统组件的性能以及应用的I/O参数配置。本文从应用I/O配置优化的视角,分析了大规模集群并行I/O的配置调优空间,在此基础上,给出了一套大规模集群并行I/O性能特征测试分析方法,基于该方法,在某国产超级计算集群上开展了一系列I/O测试分析来刻画系统的I/O性能特征,进而指导并行应用程序的I/O配置优化。基于优化后的配置参数,在两类典型的并行I/O场景中,针对某类生产应用程序,8192进程下的重启动数据写操作时间下降了15%,4096核的程序作业加载时间从10分钟缩短到了5s。本文提出的I/O配置调优空间及优化方法,可以推广应用到其它同类系统平台,对于大规模集群上的用户层并行I/O配置调优具有借鉴意义。     

GSVM:一种支持Gather/Scatter的向量存储器

宽单指令多数据流(Single Instruction Multiple Data, SIMD)架构数字信号处理器一般都能高效支持地址连续或等距跨步等规则应用的向量访存,但对于科学与工程计算中广泛存在的不规则应用的数据访存则带宽利用率往往较低,从而大幅降低了其整体运算能效。为了提高不规则应用的向量访存性能,基于某SIMD数字信号处理器的体系结构,设计了一种支持Gather/Scatter访存的向量存储器GSVM。通过设计与SIMD宽度相匹配的向量地址计算单元和合适深度的冲突缓冲器阵列,实现了Gather/Scatter指令向量地址计算、仲裁与缓存的全流水访存操作。实验结果表明,相比以前不支持Gather/Scatter访存的存储器,GSVM在增加22%的硬件代价基础上,基于稀疏矩阵向量乘的测试程序集获得了2～8的性能加速比。