水声目标机动检测与跟踪

当今海战场水声目标往往采用机动方式航行,打破了目标匀速直线运动假定的传统。通过对水声目标的机动模式分析,可将目标的运动轨迹近似为折线运动进行研究。提出了水声目标的机动检测与跟踪方法,简化机动检测的复杂度,最后讨论了对机动目标长时跟踪的策略。     

Making long-range planning work: the case of the US Army's 30-year strategic modernization plan

The extensive timespan of evolving assumptions about future adversaries, US military engagements, and technology inherent in the US Army's 30-year modernization strategy can overwhelm the management capacity of planners, and misdirect acquisition investments. Some military scholars have argued that long-range planning is futile due to the complexities of the global security environment. So how can the US Army manage the evolving assumptions inherent in its 30-year modernization strategy to ensure it remains a superior global force? This study will answer the above question by arguing that the US Army's 30-year modernization strategy, while emulative of a similar modernization approach in the threat-based planning environment of the Cold War, is viable if supported by a method and a tool that manage investments and planning assumptions.     

装甲车辆驾驶动作模糊模式识别研究

针对装甲车辆驾驶技能等级评判系统中,如何将驾驶训练过程中采集的驾驶动作数据还原成实际驾驶操作动作的瓶颈问题,采用模糊模式识别的方法,实现了对无限长待识序列数据的识别,并通过对实车采集数据的分析,验证了识别的有效性和准确性。     

一种长周期伪码信号的跳跃式扩展重叠捕获方法

随着直扩技术的发展,为提高应用系统的信号质量,出现了长周期伪码信号。然而,正是由于长周期伪码的出现及广泛应用,引发了系统接收的同步搜索误差大、速度慢等关键问题。为此,针对长周期伪码特性,考虑现有方法的局限和不足,提出了一种跳跃式扩展重叠捕获方法,通过扩展分段的跳跃式重叠技术来扩大搜索范围及提高处理速度。仿真结果表明,新方法在捕获性能及处理速度、检测概率等方面都明显优于现有的方法。此研究可以为扩频信号的高效同步及系统应用提供技术基础。     

基于多种群遗传算法的远程作战体系加油路径优化

为解决加油机在远程作战体系中的高效使用与经济性问题,推导了远程作战体系中加油机和歼击机的油耗方程模型,建立以耗油量最少为目标的空中加油路径优化模型。该模型基于一定的作战任务构想,利用多种群遗传算法进行了求解,从求解结果来看,该算法可以快速搜索到能够在所有满足作战任务要求的任务路径中寻找到最小消耗的位置坐标,以获得最大的作战效能,从而达到优化远程作战体系中的作战和训练中加油机使用问题。     

I-TCP:一个高速长距离网络的AIMD算法

针对互联网中高速长距离网络上的拥塞控制问题,基于同步AIMD源在Drop-Tail网络环境下竞争带宽的网络模型,提出了一个新的AIMD拥塞控制算法,I-TCP。通过合理地设计高速和低速模式以及模式切换规则,使得该算法适合于配置到高速长距离网络,同时在传统低速网络中也具有很好的性能。仿真试验表明,I-TCP中的AIMD算法在高速长距离网络中表现出了很好的效率、公平性和网络响应性。     

柔性长鳍仿生装置波动控制技术

以“尼罗河魔鬼”柔性长鳍的波动运动为背景,主要研究了柔性长鳍仿生装置的波动控制技术。首先研究了柔性长鳍推进波产生原理及其技术实现方案;其次,针对仿生装置的实际系统建立了鳍面运动学模型,从中提取了能够实现柔性长鳍多种波动模式的波动控制参数,并对柔性长鳍几种特定波动模式的实现方法进行了仿真研究;进一步,根据仿生装置的物理约束条件提出了控制参数的合法性判据;最后,研究了柔性长鳍仿生装置的波动控制策略及其控制流程。研究结果表明柔性长鳍仿生装置通过合理设置控制参数能够模拟仿生对象的多种波动运动模式,并实现推进波的多参数控制,能够满足柔性长鳍多种波动参数下的非定常流体动力学测试的需要。     

一种面向分布式寄存器文件的VLIW调度新策略

新一代面向密集计算的高性能处理器普遍采用分布式寄存器文件来支撑ALU阵列,并通过VLIW开发指令级并行.面向分布式寄存器文件的编译成为新兴的研究热点,在斯坦福大学的kernelC编译器ISCD中最早提出了面向分布式寄存器的VLIW调度问题,在该领域处于领先水平,但是没有解决重负载下的分布式寄存器分配问题,使应用编程受到极大限制.在其基础上提出了一种新的VLIW调度策略,实验结果表明能够很好地解决重负载下的分布式寄存器分配问题.     

Risk‐sensitive sizing of responsive facilities

We develop a risk‐sensitive strategic facility sizing model that makes use of readily obtainable data and addresses both capacity and responsiveness considerations. We focus on facilities whose original size cannot be adjusted over time and limits the total production equipment they can hold, which is added sequentially during a finite planning horizon. The model is parsimonious by design for compatibility with the nature of available data during early planning stages. We model demand via a univariate random variable with arbitrary forecast profiles for equipment expansion, and assume the supporting equipment additions are continuous and decided ex‐post. Under constant absolute risk aversion, operating profits are the closed‐form solution to a nontrivial linear program, thus characterizing the sizing decision via a single first‐order condition. This solution has several desired features, including the optimal facility size being eventually decreasing in forecast uncertainty and decreasing in risk aversion, as well as being generally robust to demand forecast uncertainty and cost errors. We provide structural results and show that ignoring risk considerations can lead to poor facility sizing decisions that deteriorate with increased forecast uncertainty. Existing models ignore risk considerations and assume the facility size can be adjusted over time, effectively shortening the planning horizon. Our main contribution is in addressing the problem that arises when that assumption is relaxed and, as a result, risk sensitivity and the challenges introduced by longer planning horizons and higher uncertainty must be considered. Finally, we derive accurate spreadsheet‐implementable approximations to the optimal solution, which make this model a practical capacity planning tool.© 2008 Wiley Periodicals, Inc. Naval Research Logistics, 2008     

组网电子对抗中应对远程威胁的对抗优选决策

在组网电子对抗作战背景下,简单分析作战中远程威胁目标,讨论了威胁决策因子及其优属度获取办法,在此基础上,提出了经过对抗方案总体设计、最优方案选择的决策方法,建立了应对远程威胁目标的决策模型。最后,通过合理假定作战场景进行实例分析,验证了该方法是科学合理的。