陆空通话歧义浅析

航空无线电陆空通话（以下简称通话）以英语作为国际通用语言，陆空双方通过话音通讯这种基本途径联接，信息交流主要依靠听和说两种交际形式。歧义现象渗透于语言交际的各个层面，如：词汇、语音、语法和语境，无论英语是否是通话者的母语，通话者都无法完全超脱偶合的语言歧义的干扰或者制约。在航空史上曾有不少重大事故的起因不在于技术或者能力缺陷，而仅仅在于通话语言的问题，有的空难甚至就源于一字之差。本文拟对陆空通话中的语言歧义现象进行归类，分析其特点、成因和危害，探讨避免和排除歧义的方法。所谓语言歧义，就是一个词、…     

基于统计语言模型的译文选择方法

机器翻译是把一种自然语言转变成另一种自然语言的过程。其中，源语言词汇的译文选择直接决定机器翻译的质量，是人们关注的难题。本方法以信息论和计算语言学为理论基础，充分利用了日语分析生成的句子结构信息和语言模型提供的统计信息，来解决机器翻译中歧义词的译文选择问题。将该方法用于日汉机器翻译系统的译文选择实验，具有良好的译文选择性能。它的特点在于利用了日语分析的句法信息，使译文选词的质量得到明显改善。     

新型磁悬浮振动陀螺的设计与分析研究

振动陀螺是一种利用哥氏效应检测角速度的传感器,其谐振子的结构精度和阻尼均匀性限制了陀螺性能的提升。为减小谐振子结构与支撑锚点的影响,提出了一种全新概念的磁悬浮振动陀螺。该陀螺利用电磁悬浮的方法将谐振子悬空,从而简化了谐振子为无支撑锚点的集中质量块,降低了其结构精度要求,消除了机械结构阻尼,最终达到提升陀螺性能的目的。基于经典振动陀螺模型,理论分析了磁悬浮振动陀螺的基本工作原理,并说明了谐振子误差对陀螺性能的影响规律,设计了新型磁悬浮振动陀螺的结构,并对该结构的磁感应强度进行了仿真分析。仿真结果证明,悬浮质量块振动稳定,具有较好的磁场均匀性。最后对陀螺样机进行了测试,其固有频率为20Hz,标度因子约为1.6mV/[(°)/s],测试结果验证了所提磁悬浮振动结构的陀螺效应。     

通信系统试飞中语音清晰度判定研究

描述了飞机通信系统试飞中语音清晰度的概念以及"SINPO"等级划分。介绍了国内目前在试飞中语音清晰度判定的方法及其不足之处,通过对国外"同韵词表格法"的分析和研究,在此基础上提出了适合我国通信系统试飞的"近音词表格法"。该方法用定量计算的结果来判定语音清晰度的等级进而准确判定通信系统的功能符合性,可为从事飞机通信系统试飞领域的人员提供一种新的思路和方法。     

英语歧义和模糊现象阐释

语言歧义是一种存在于任何语言中的普遍现象,歧义的出现会造成交际双方的理解障碍,借助语境,歧义可自行消除;而模糊现象是语言符号内在的语义不清楚和意义不明确造成的,是语言的内在特征,模糊现象不可避免。歧义和模糊是语言不确定性的主要表现形式,是两个容易混淆和不易区分的概念。将对英语中的歧义和模糊现象进行阐释,分析两种现象产生的根源,区分差异,以及歧义和模糊现象在语言应用中达到的语用效果。     

局部自主遥操作中的抓取构型识别研究

针对局部自主遥操作过程中识别目标准确率低的问题,提出了一种基于改进快速区域卷积神经网络的抓取构型识别方法,通过对其区域生成网络中锚点尺度、前景特征区域、候选框的线性回归和分类网络分别进行改进,以提高抓取构型识别的准确率。首先将抓取构型参数化,然后在目标区域中利用锚点法对抓取构型参数进行识别,结合视觉传感器采集到的深度信息确定目标高度,并通过线性回归方法对抓取区域进行修正。通过搭建机器人试验平台,利用Cornell Grasp Dataset进行训练与测试进行验证。试验结果表明,提出的方法在简单网络识别准确率为96.4%,并成功实现机器人对目标的抓取。     

基于语义相似度的句法歧义结构消解

以汉语句法结构分析中最普遍存在的句法结构歧义现象,即“动词短语 名词短语 的 名词短语”(VP NP1 的/u NP2)为研究对象,提出了一种基于知网知识库的歧义消解策略。据此提出利用改进的知网相似度计算的歧义消解策略。通过对从人民日报的语料中抽取1000个“VP NP1 的/u NP2”的实例进行测试,实验结果正确率达到83.4%,表明这是一种有效的识别“VP NP1 的/u NP2”歧义结构的方法。     

基于SNA的航空发动机空中停车文献分析

由于发动机故障引起的空中停车存在高危险性和不确定性,必须做好防范和处置。在 CNKI 数据库中以“空中停车”关键词检索相关研究文献,应用 SATI3.2 软件对检索文献数据进行数据预处理,获取高频关 键词并构建共词矩阵;应用 Ucinet6.0 软件对关键词进行社会网络分析和聚类分析。结果表明:航空发动机空中停车研究领域热点集中于管理、维修、人员及企业运营等方面;民用航空与军用航空在该领域的研究重点存在差异;非常规类飞机的空中停车受到关注,无人机空中停车研究将成为该领域的热点问题;军用飞机空中停车研究成果相比民用飞机存在一定差距,随着军用飞机适航工作的开展和推进,军用飞机空中停车研究将在该领域占据重要地位。     

面向结构化场景的激光雷达点云高精度配准与定位方法

在基于先验地图的激光雷达室内导航方案中,通常采用点云配准的方法进行无人设备位姿初始化.在结构化场景下,传统配准算法特征鲁棒性较差,导致点云配准的误差较大且易陷入局部最优.针对该问题,提出了一种基于多平面空间模型的点云快速配准方法.首先该方法利用特征直方图的思想对空间点云进行快速粗聚类,根据平面一致性将粗聚类后的点集进行...     

一种高效的基于启发式规则和词典相结合的双语词对齐方法

双语词对齐是指在互译的双语句对中找到词汇级的对译关系,是自然语言处理领域一个非常有用而又比较困难的研究课题。本文在对当前主流的词对齐方法进行实验分析的基础上,综合考虑了各种因素,提出了基于启发式统计规则和词典相结合的方法,该方法充分利用现有资源,同时考虑到了后续的应用问题。实验表明,该方法在训练语料规模较小的情况下,取得了较好的对齐结果。     

Volume Properties for the Wideband Ambiguity Function

It is shown that the volume under the wideband ambiguity function is close to the square of the signal energy. The ambiguity volume is asymptotically conserved as the signal approaches the narrowband case. The narrowband ambiguity volume is a lower bound for the volume of the wideband ambiguity function.     

Resolution Performance of Pulse Trains with Large Time-Bandwidth Products

The coherent pulse train has good clutter suppression performance because the energy in its matched-filter response is essentially concentrated within sharp ambiguous spikes. However, this is so only when the Doppler distortions are neglected, so that the Doppler effect is taken as a simple translation of the carrier frequency. This paper analyzes the consequences of Doppler distortions on the resolution performance of pulse trains. It is found that Doppler distortions widen the Doppler ambiguities of the pulse train response, with the widening factor proportional to the order of the Doppler ambiguity. This reduces the interval between Doppler ambiguities, and hence the Doppler width of a clutter space that can be accommodated without severe clutter interference. For an operation in a Doppler-ambiguous mode, it also degrades nominal Doppler resolution performance. A detailed analysis of the effects is presented, and numerical results on the widening of the Doppler ambiguities are obtained.     

On the Ambiguity Function of Random Binary-Phase-Coded Waveforms

The ambiguity function of truly random binary-phase-coded waveforms, as an approximation to those waveforms commonly employed in binary-modulated pseudonoise systems/encoded radar systems, is investigated. In a statistical sense, the ambiguity function is analytically derived in which the normally used deterministic cross-correlation process is replaced by its ensemble average. Various Doppler filter responses are presented and discussed. The results are compared with those obtained by transmitting an aperiodic maximum length pseudorandom sequence. It is shown that the ambiguity function of the latter case is closely represented by the ensemble-average response of the truly random binary signal.     

Ambiguity Resolution in Interferometry

A comprehensive theory of interferometry from a system viewpoint with particular emphasis on the ambiguity resolution problem is developed. The derived error equations include contributions from all system uncertainties, i.e., phase measurement, frequency, and element phase center position errors in three dimensions. The direction-of-arrival errors are inversely proportional to the interferometer baseline and it is customary to make the baseline large enough to meet the accuracy requirements. A system with a baseline greater than a half-wavelength results in the well known direction-of-arrival ambiguity problem with the addition of a third element to each baseline being a common method for resolving the ambiguity. It is shown that contrary to previous thinking there are many equally optimal positions for adding the third element to resolve the ambiguity. In addition, it is shown how the measurement made to resolve the ambiguity can also be applied to increase the accuracy of the angle-of-arrival measurement. A central result is the derivation of expressions specifying the probability of correct resolution of ambiguities as a function of system parameters and system errors. Moreover the concept of an acceptance criterion designed to reduce processing of erroneous measurements is developed. Narrowing the criterion reduces the percentage of data accepted for processing, but increases the probability of correct ambiguity resolution. This is analogous to the relationship between the probability of detection and the probability of false alarm in radar theory.     

Delay-Doppler Ambiguity Function for Wideband Signals

The conventional ambiguity function is extended to include the Doppler distortions of the modulation function. The distinctive features of the extension are the use of the complex notation for wideband signals, and inclusion of the Doppler effect on the signal amplitude. The result is an ambiguity function from which Woodward's form can be found by inspection. It is shown that the well-known volume constraint also applies, in unchanged form, to the generalized ambiguity function. For the volume to be constant, it is not required that the distortions of the modulation function be neglected. Rather, the volume constancy is related to the sinusoidal fluctuations of a modulated carrier-type signal and thus is strictly a matter of the percentage bandwidth of the signal.     

A Logarithmic Frequency Allocation Algorithm for Wideband Discrete Frequency Pulse Trains

It is shown that signal waveforms utilizing discrete frequency modulation (DFM) which are generated using a narrowband or frequency shift algorithm have ambiguity sidelobe distortion which is caused by the approximation of time compression by frequency shift. A logarithmic frequency allocation algorithm is presented which couches the signal design problem in terms of band and step ratios, rather than in terms of bandwidth and frequency steps, and is consistent with the wideband formulation of the ambiguity function. The algorithm makes use of the same basic code generating sequence used for narrowband frequency allocation, but the resulting signal will have invariant ambiguity sidelobe positions for any receiver realization in the delay-time compression plane.     

部分模糊度固定方法在GNSS/INS紧组合中的应用研究

针对动态环境下GNSS/INS导航定位模糊度固定难的问题,提出了一种基于部分模糊度固定的GNSS/INS紧组合导航定位算法,通过选择模糊度的最优子集进行固定,并利用实测车载实验进行了算法验证.实验结果表明:在开阔环境下,基于部分模糊度固定的GNSS/INS紧组合与基于全模糊度固定的GNSS/INS紧组合算法的定位精度相...     

Parametric Methods for Strip-Map SAR Doppler Ambiguity Resolution

Physical modeling of the Doppler centroid (DC) can be used to predict synthetic aperture radar (SAR) Doppler ambiguity when antenna attitude is controlled or measured precisely enough. It is shown that the same model proves useful even in the cases of higher attitude uncertainty, if it is combined with suitable adaptive techniques. In this paper, Doppler ambiguity resolution is formulated as a hypothesis testing problem over a domain of integer values that are directly related to the attitude uncertainty. A test statistic is derived from the entire SAR scene using data adaptive processing. A broad class of such adaptive algorithms is analyzed in a unified way, starting from the range-azimuth coupling in the frequency domain and multilook techniques. The analysis includes two well-known and two new multilook methods for Doppler ambiguity resolution. A suitable test statistic is proposed for each of these methods and its dependency on the scene spatial correlation is discussed. Experimental results confirm the robustness of the combined scheme.     

Doppler ambiguity resolution using multiple PRF

An algorithm for velocity ambiguity resolution in coherent pulsed Doppler radar using multiple pulse repetition frequencies (PRF) is presented. It relies on the choice of particular values for the PRFs. The folded frequency of the target signal is obtained by averaging the folded frequency estimates for each PRF, and a quasi maximum likelihood criterion is maximized for ambiguity order estimation. The fast implementation of this nonambiguous estimation procedure is based on the fast Fourier transform (FFT), The proposed waveform allows full exploitation of any (even) number of PRFs, which appears to be important for estimation improvement. The effects of the waveform parameters and the folded frequency estimation variance on the performance of the ambiguity order estimation procedure are evaluated theoretically and through computer simulations. Mean square error (MSE) curves are given to assess the Doppler frequency estimation accuracy. Finally, the new method is compared with a classical technique and the implementation of the algorithm in a clutter environment is addressed.     

Optimum Signals for Radar

The research reported herein deals with the general problem of the selection of radar waveforms. The investigation is specifically concerned with the synthesis of radar signals which are optimum in the sense that they are characterized by ambiguity surfaces minimized over certain predetermined regions of the ambiguity plane. The weighted ambiguity surface is utilized as the weighted error criterion. This error criterion is mathematically tractable and pertinent to radar system performance but is not unduly restrictive as some orientation parameters are left unspecified for subsequent cost or penalty function analysis. The signal optimization is approached by variational techniques augmented by equality and inequality constraints, for example, limiting the amount of bandwidth or frequency modulation to be less than some system requirement. Several examples are presented demonstrating the optimization techniques and providing a minimum error for the stated problem. It is shown that for any given type of amplitude modulation of the radar signal, the variance or dispersion of the ambiguity surface is not decreased for any type of phase modulation added. The optimum signal for an elliptical weighting function is derived for several cases. The minimum error is shown to depend upon the constraints and the unspecified orientation parameters and, for one case, on the second moment of the signal.