改进的特征匹配雷达视频运动目标跟踪算法

针对密集杂波和复杂多目标情况下,特征匹配跟踪算法定位跟踪精度较低的问题,提出一种改进的特征匹配雷达视频运动目标跟踪算法。首先,从单帧雷达图像检测结果中提取目标的面积、位置、不变矩信息,在运动目标得到判定后,利用卡尔曼滤波的预测位置以及上一帧特征信息进行目标匹配;然后,根据匹配位置和预测位置给出目标估计位置,并进行特征信息更新;最后,输出目标的特征信息、位置信息和运动参数。利用某型导航雷达上采集的实测数据验证了算法的有效性。     

雷达参数对SAR图像中Kelvin尾迹的影响

舰船尾迹的SAR成像受到船艇自身、海洋环境和SAR系统等多种因素的影响,其中起主要作用的是雷达参数的影响。文章重点对影响舰船Kelvin尾迹SAR成像的SAR系统参数进行了详细讨论和分析,其中包括极化方式、视向角、频率以及入射角对Kelvin尾迹的影响因素。给出了各种参数对尾迹成像的影响规律,通过此规律有助于下一步得出舰船尾迹的最佳观测条件。     

基于中值滤波和参考点的风洞压敏涂料试验图像处理技术

提出了一种结合图像中值滤波和试件参考点的风洞压敏涂料（PSP）试验图像数据处理方法。与通常用校准箱获取校准曲线进而得到PSP试验试件表面压力或压力系数（Cp）的方法不同，该方法通过空心中值滤波技术恢复PSP光强比图像中测压孔处的光强比数据，结合若干参考点处常规测压技术实测所得Cp值，获得Cp值和PSP图像光强比数据的关系曲线，进而得到整个试件表面的Cp值数据，该方法获得的数据结合了风洞试验诸多因素对试验结果的影响。利用MATLAB对PSP技术试验图像数据进行了处理，并将PSP方法获得的Cp值与常规测压方法获得的Cp值进行了比较，得到的结果显示，笔者采用的方法获得的Cp值与常规测压方法获得的Cp值吻合较好。该方法为PSP图像数据的处理提供了一种新的方法，具有较大的发展潜力和应用价值。     

基于Gauss平均引数的RNAV进近程序航迹引导参数计算

为准确计算基于WGS-84坐标的RNAV进近程序航路点坐标和航迹引导参数,提供精密RNAV导航数据,降低RNAV进近系统误差,提出了实测跑道入口WGS-84坐标,然后沿跑道中线外推各航路点WGS-84坐标的计算方法。采用Gauss平均引数法对各航段的航线角及航段距离进行了反算,以验证外推航路点坐标的计算精度。通过西南地区某机场＂T＂型RNAV进近程序的设计及飞行验证,该方法可推广到其他机场的RNAV进近程序设计。     

Evaluation of global modeling of M(3000)F2 and hmF2 based on alternative empirical orthogonal function expansions

Global modeling of M(3000)F2 and hmF2 based on three alternative EOF (empirical orthogonal function) expansion methods is described briefly. Data used for the model construction is the monthly median hourly values of M(3000)F2 from the ionosonde/digisonde stations distributed around the world for the period of 1975–1985 and the hmF2 data of the same period converted from the measured M(3000)F2 based on the strong anti-correlation existing between them. Independent data of a low (1965) and a high (1970) solar activity year are used to validate the three alternative models based on different EOF expansion methods. Comparisons between the modeled results and observed data for both the low (1965) and high (1970) solar activity years showed good agreement for both M(3000)F2 and hmF2 parameters. Statistical analysis on the differences between model values and observed data showed that all the three alternative models (Model A, B and C) based on the different EOF expansion methods have better agreement with the observed data than the models currently used in IRI. All three alternative EOF based models have almost the same accuracy. Discussion on the preference of the three alternative EOF based models is given.     

Estimation of equatorial electrojet from total electron content at geomagnetic equator using Kalman filter

The temporal variation of the equatorial electrojet is estimated utilising a suitably designed Kalman filter and using the established empirical relations between the anomaly component of equatorial TEC and the modified electrojet. TEC data obtained from dual frequency GPS receivers are used for the purpose. Estimation requires the a-priori knowledge of the peak electrojet value of the day and hence can be made in post temporal scenario only. Initial results obtained during a low solar activity time in an equinoctial month shows acceptable accuracy of the proposed algorithm. Limited analysis is done by segregating the results into temporal sessions of pre-attainment and post-attainment of the electrojet peak.     

一种用于GPS时差比对数据处理的卡尔曼滤波器设计

介绍了一种用于GPS时差比对数据处理的自适应卡尔曼滤波器,在Matlab环境下编写了离散时间形式的卡尔曼滤波算法程序,利用该算法对GPS信号与铯原子频标的时差比对数据进行了分析处理,最后介绍了该算法的数字芯片FPGA实现。     

Validation of B2bot in the NeQuick model during high solar activity at Hainan station

NeQuick ionospheric electron density model, which has been developed to version 2, produces the full electron density profile in the ionosphere. Each part of the profile is modeled using Epstein layer formalism. Simple empirical relations are used to compute the thicknesses of each layer. In order to validate the B2_bot parameter in the NeQuick model during high solar activity, we use the data at Hainan, China (109.1°E, 19.5°N; Geomagnetic coordinates: 178.95°E, 8.1°N), measured with DPS-4, and study the diurnal and seasonal variations of B2_bot, ΔB2 (B2_best − B2_NeQuick 2) and the seasonal median values of B2_best/B2_NeQuick 2 at that region. The results show that, (1) The differences between B2_best and B2_NeQuick 2 have diurnal and seasonal variations. (2) The diurnal variations of B2_NeQuick 2 are smaller than those of B2_best. (3) Generally, except for early morning the experimental values are properly reproduced. (4) Generally, during morning the NeQuick model has an underestimation. The magnitude of underestimation varies with LT and season.     

空间陀螺仪显式标定与隐式标定方法

针对航天器陀螺仪误差系数的标定问题,提出了显式与隐式两种标定方法。建立了显式标定与隐式标定方法的状态方程与测量方程,利用扩展Kalman滤波获得了陀螺仪误差系数的高精度估计,同时比较了显式标定与隐式标定的不同,仿真结果表明,在当前参数设置下,显式标定与隐式标定精度基本相同,两种方法得到的误差系数估计相对误差在10%以内,而隐式标定方法实现较为容易,但对航天器姿态测量精度较为敏感。     

An intercomparison of meteor radar measurements at the South Pole using two different processing systems

A new meteor radar system was installed at the Amundsen–Scott station South Pole in 2001 to further the understanding of the dynamics of the Antarctic region. The antenna array consists of four yagis pointed along the 0°, 90°, 180°, and 270° meridians and five folded crossed dipoles arranged in a cross configuration and operating as an interferometer to provide position measurements for the detected radio meteors. The four yagis are time division multiplexed and used for both transmitting and receiving while the five folded crossed dipoles are only used for reception. The current arrangement of data acquisition (DAQ) systems at the South Pole allows the collection of meteors in a configuration similar to the previous meteor radar system that operated at the South Pole in the mid 1990s while also using an interferometer to accurately determine the meteor positions in the sky, which enables the determination of the vertical structure of the observed waves. This has been accomplished through the use of two DAQ and post-processing systems: COBRA (Colorado Obninsk radar) connected to the yagis and MEDAC (meteor echo detection and collection) connected to the folded crossed dipoles. With two separate DAQ systems operating in parallel we have the ability to directly compare the results and understand the inherent variability in the derived scientific results based on different system architectures and processing assumptions. The impact of operating a system without an interferometer on the amplitudes and phases of the observed wave components is considered. We find that the lack of altitude resolution of the COBRA DAQ system leads to an underestimation of the amplitude of the s = 1 component of the semidiurnal tide of ∼20% during the summer months.