基于多帧聚类的紧凑型地波雷达海上目标航迹起始方法

紧凑型地波雷达是专属经济区内海上船只目标监测预警的重要手段。由于其发射功率低、目标回波信噪比低,在对海上船只目标检测过程中较低的检测概率极易导致目标漏检,采用序贯类方法难以及时起始航迹。对此,本文通过分析杂波和目标的地理位置以及径向速度随时间变化的特点,提出了一种基于多帧聚类的紧凑型地波雷达海上目标航迹起始方法。该方法利用目标和杂波在连续多帧内运动特征的差异,在具有噪声的基于密度的空间聚类算法（Density-Based Spatial Clustering of Applications with Noise, DBSCAN）求解聚类ε邻域的过程中增加运动特征约束,将各船只目标在多帧内的点迹聚类为不同的簇,实现海上船只目标点迹与杂波点的区分,将簇内点迹按时间顺序顺次连接得到起始航迹。利用仿真与实测紧凑型地波雷达数据开展了航迹起始实验,结果表明,与逻辑法相比,本文方法得到的航迹起始时间平均提前了5.95 min,丢点率平均降低了12.68%,解决了海上弱目标航迹起始时间滞后的问题,适用于杂波区目标与雷达远端船只目标的航迹起始。     

紧凑型高频地波雷达目标探测研究进展

高频地波雷达是海上船只目标大范围连续探测的主要手段,与大型阵列雷达相比,紧凑型地波雷达具有占用场地小、功耗低、架设和维护方便等优点。本文首先概述了高频地波雷达目标探测技术的国内外研究状况,然后通过阐述紧凑型地波雷达目标探测技术的研究进展,分析紧凑型地波雷达目标探测中难点和急需解决的问题,给出紧凑型地波雷达目标探测的新方案。该方案利用检测跟踪一体化方法解决单方位紧凑型地波雷达弱信号检测困难的问题,利用多方位目标航迹融合确定精确目标航迹,能够降低目标航迹丢失概率,提高目标探测精度。最后总结了紧凑型地波雷达目标探测的发展趋势和研究重点。     

高频地波雷达海上目标航迹跟踪新思路

高频地波雷达是对海上运动目标进行监视监测的一种重要手段,为了提高地波雷达对海上特定目标独立跟踪探测时的性能,本文对高频地波雷达海上目标跟踪技术的研究现状进行了综述,分析总结了目前航迹跟踪方法存在的主要问题。结合海上目标跟踪的实际应用需求,借助目前流行的深度学习方法充分挖掘其他同步探测手段获取的目标信息,提出了基于知识辅助的特定目标跟踪方法,改善后续地波雷达对特定目标独立跟踪时的航迹质量,初步的航迹跟踪结果验证了提出方法的有效性。提出的地波雷达特定目标跟踪方法对目标跟踪方法的理论研究及地波雷达目标跟踪系统的业务化应用均具有重要意义及参考价值。     

基于FJPDA关联算法的地波超视距雷达目标跟踪

针对地波超视距雷达(SWOTHR)目标跟踪中JPDA算法关联复杂度高、计算量大的问题,将模糊联合概率数据关联(FJPDA)算法应用于地波超视距雷达目标跟踪中。在FJPDA跟踪方法中采用了目标点迹与航迹的模糊隶属度,解决目标点迹与航迹的关联概率问题,避免了传统JPDA算法中对有效矩阵的拆分导致的计算量组合爆炸的现象,可以降低目标跟踪算法的复杂度,提高运行效率。通过实测地波雷达数据的目标跟踪实验分析及与同步AIS信息的比对分析,结果表明:基于FJPDA算法的跟踪方法可以有效跟踪SWOTHR中的目标航迹,跟踪效果与JPDA算法相当,且运行时间明显减少,适合工程应用。     

基于弧段检测的高频地波雷达特定目标航迹跟踪方法研究

本文对现有的高频地波雷达目标跟踪方法进行了概述,提出了一种地波雷达目标长时连续跟踪的方法,基本思想是:充分挖掘航迹弧段特征,基于特征对船只运动建模,并结合杂波背景进行融合决策。进一步,为了达到长时间连续跟踪的需求,借鉴深度学习的思想,利用新获取的弧段数据对算法估计结果不断递归校正,使得随着获取数据的增加跟踪越准确。该方法适用于杂波环境且在航道附近存在众多干扰船只的情况下对机动目标航迹的实时稳定跟踪,为高频地波雷达在复杂干扰环境下特定目标持续跟踪提供理论基础和方法指导,为充分发挥地波雷达在海上监视监测中的作用提供技术支撑。     

高频地波雷达海上目标航迹校正方法

高频地波雷达对海上目标跟踪探测时,由于距离和方位角分辨率低且受到随机噪声的干扰,导致对目标位置探测不准确,形成的航迹会偏离目标的真实位置,影响跟踪的准确性。针对这个问题,结合海上特定目标跟踪的实际,本文从航迹的角度出发,基于对距离和方位角误差时间序列的统计建模,提出了一种基于同步船舶自动识别系统(Automatic Identification System,AIS)信息的高频地波雷达海上目标航迹校正方法。在高频地波雷达与AIS同步跟踪期间,采用统计回归的方法建立目标距离和方位角误差的校正模型;在高频地波雷达独立跟踪时,利用得到的模型分别对距离和方位角进行校正。利用实测高频地波雷达目标探测数据进行了航迹校正实验,校正后距离和方位角的均方根误差分别减少了约90%和75%,航迹跟踪效果得到了显著改善。     

基于自适应α—β滤波的HFSWR海上目标航迹跟踪

针对基于α-β滤波的高频地波雷达海上目标航迹跟踪方法跟踪精度较低、工程适用性较差的问题,提出了一种极坐标下α-β滤波自适应系数选取方法,该方法有效利用了高频地波雷达目标量测参数的特点,能够在跟踪过程中根据实际量测对滤波系数进行自适应调整。利用仿真及实测地波雷达数据以及同步AIS航迹对提出方法的航迹跟踪性能进行了实验验证,并与现有自适应系数选取方法的跟踪性能进行了对比,结果表明,本文方法能够得到平滑、持续的航迹输出,跟踪距离远,跟踪率及跟踪精度高。     

浅谈天发地收超视距雷达在南海监测中的应用

我国南海海域范围大,海洋动力环境恶劣,常规手段难以实现对南海的大范围连续观测。天发地收超视距雷达（天地波雷达）能够同时实现海洋动力要素监测和海上移动目标监视监测,具有天波雷达覆盖面积大、探测距离远的特点,又具有地波雷达对目标发现概率高的特点,是一种新的适用于我国南海的监视监测手段。文章分析了天地波雷达的特点,论述了天地波雷达在南海海洋动力监测和海上目标探测的应用前景。     

地波雷达目标检测跟踪联合处理粒子滤波方法

针对固定粒子数PF-TBD算法计算量大、复杂环境下地波雷达海上船只目标检测与跟踪性能不佳的问题,本文将粒子滤波方法应用于地波雷达船只目标检测与跟踪中,提出了基于自适应粒子滤波的地波雷达目标检测与跟踪联合处理方法。该方法结合地波雷达回波谱中目标展宽特性,充分利用了地波雷达回波谱中面目标的粒子权重信息来设置粒子自适应采样策略,提高了目标检测和跟踪联合处理的效果。通过地波雷达实测数据的目标跟踪结果及与同步AIS信息的比对分析,结果表明:提出的检测跟踪联合处理方法在对低信噪比、快速机动等复杂环境下的多目标跟踪时,可提高目标整体跟踪性能。     

基于交互式多模型的双站高频地波雷达机动目标跟踪算法

在实际应用中,由于高频地波雷达的测向精度较低,利用单站式雷达系统进行目标跟踪容易造成偏差。然而,在双站式雷达系统中,可以只利用位置和速度信息进行跟踪来避免了方位不精确带来的影响。同时,在双站式雷达体制下,仍然存在提高机动目标跟踪精度的需求。因此,本文提出一种基于交互式多模型(IMM)的跟踪算法,该算法通过建立不同的运动模型来反映目标不同的运动状况,并以此来估计目标的真实运动状态。仿真结果表明,对于双站高频地波雷达机动目标跟踪,IMM算法表现出比传统的单模算法更好的跟踪性能。     

Passive sonar detection and localization by matched velocityfiltering

This paper presents a new bearings-only method of detecting and tracking low signal-to-noise ratio (SNR) wideband targets on a constant course and velocity trajectory. A track-before-detect strategy based on matched velocity filtering is adopted using spatial images constructed from a sequence of power bearing map (PBM) estimates accumulated during a track. To lower the threshold SNR for detection, a discrete bank of matched velocity filters integrates the PBM images over a range of hypothesized trajectories, such an approach eliminates the need to estimate the number of targets since signal detection is determined by comparing the output of each matched filter (MF) to a decision threshold. The distribution of the MF output is derived based on a single point target in diffuse noise assumption. Receiver operating characteristic curves show a definite detection gain under low SNR conditions for matched velocity filtering (track-before-detect) over detection from a single PBM     

Multi-Bernoulli Filter for Tracking Multiple Targets Using Sensor Array

This paper presents a multi-Bernoulli filter for tracking the direction of arrival(DOAs) of time-varying number of targets using sensor array. Our method operates directly on the measurements of sensor array and does not require any detection. Firstly, more information is reserved and compared with the after-detection measurements using a finite set of detected points. It can significantly improve the tracking performance, especially in low signal-to-noise ratio. Secondly, it inherits the advantages of the multi-Bernoulli approximation which models each of the targets individually. This allows more accurate multi-target state estimation, especially when targets cross. The proposed filter does not need clustering step and simulation results showcase the improved performance of the proposed filter.     

Robust broadband beam-forming based on the feature of underwater target radiated noise

To the problem of the unknown underwater target detection, according to the feature that the underwater target radiated noise contains the stable line spectrum, a weighted method based on the main-to-side lobe ratio (MSLR) is proposed for broadband beam-forming. This weighted method can be implemented by using the following steps. Firstly, optimize the spatial spectrum of each frequency unit by the second-order cone programming (SOCP), and obtain the optimized spatial spectrum with lower side lobe. Secondly, construct weighting factors based on the MSLR of the optimized spatial spectrums to from weight factors. Lastly, cumulate the spatial spectrum of each frequency unit via the weight statistical method of this paper. This method can restrain the disturbance of background noise, enhance the output signal-to-noise ratio (SNR), and overcome the difficulty of traditional four-dimensional display. The theoretical analysis and simulation results both verify that this method can well enhance the spatial spectrum of line spectrum units, restrain the spatial spectrum of background noise units, and improve the performance of the broadband beam-forming.     

Temporal variability of tsunami arrival detection distance revealed by virtual tsunami observation experiments using numerical simulation and 1-month HF radar observation

The combination of a high-frequency ocean surface radar and a tsunami detection method should be assessed as the onshore-offshore distribution of tsunami detection probability, because the probability will vary in accordance with the signal-to-noise ratio (SNR) and the tsunami magnitude in addition to the radar system specifications. Here, we statistically examine the tsunami detection distance based on virtual tsunami observation experiments by using signals received by a high-frequency radar in February 2014 installed on the southern coast of Japan and numerically simulated velocities induced by a Nankai Trough earthquake. In the experiments, the Doppler frequencies associated with the simulated velocities were superimposed on the receiving signals of the radar, and the radial velocities were calculated from the synthesized signals by the fast Fourier transform. Tsunami arrival was then detected based on the temporal change in the cross-correlation of the velocities, before and after tsunami arrival, between two points 3 km apart along a radar beam. We found that the possibility of tsunami detection primarily depends on the kinetic energy ratio between tsunami current and background current velocities. The monthly average detection probability is over 90% when the energy ratio exceeds 5 (offshore distance: 9 km ≤ L ≤ 36 km) and reduces to 50% when the energy ratio is approximately 1 (L = 42 km) over the shelf slope. The ratio varied with the background current physics and SNR, which was mainly affected by ocean surface wave heights and ionospheric electron density.     

X-波段雷达近海海浪频谱反演的神经网络模型

X-波段雷达作为国内海浪观测的一种新工具,在海浪频谱获取和有效波高反演方面仍存在较多问题.本文利用非线性回归方法,将现场实测浮标数据频谱和雷达一维图像谱分别与标准频谱模型进行拟合,发现浮标频谱和一维图像谱具有标准频谱的特征,能够较准确地获取相应的谱参数.提出了建立由雷达一维图像谱参数反演海浪频谱参数的神经网络模型,同时在模型中加入影像序列信噪比,进而反演有效波高,并将反演结果与现场实测数据和传统算法(建立影像序列信噪比与有效波高之间的线性回归方程)进行了对比,结果表明,获取谱参数的误差和反演有效波高的平均误差在20％以内,而传统算法计算有效波高平均误差在20％以上.     

Study on dim target detection and discrimination from sea clutter

Dim target detection from sea clutter is one of the difficult topics in ocean remote sensing application. By aiming at the shortcoming of false alarms when using track before detect (TBD) based on dynamic programming, a new discrimination method called statistics of direction histogram (SDH) is proposed, which is based on different features of trajectories between the true target and false one. Moreover, a new series of discrimination schemes of SDH and Local Extreme Value method (LEV) are studied and applied to simulate the actually measured radar data. The results show that the given discrimination is effective to reduce false alarms during dim targets detection.     

Super-resolution processing for HF surface wave radar based onpre-whitened MUSIC

Owing to the decametric wavelength, the angular resolution of high-frequency surface wave radar (HFSWR) is usually coarse, especially when dimensions of antenna arrays are restricted such as in shipborne HFSWR applications. In this paper, the relative strength of atmospheric noise and sea clutter that will heavily degrade the capabilities of HFSWR in target detection and resolution are calculated, then a method for estimating the spatial covariance matrix of background noise is presented. By introducing a pre-whitening procedure in multiple signal classification (MUSIC), the resolution performance of MUSIC is enhanced in spatially colored noise environment. Results with data from the aircraft detecting experiment conducted by the Research Institute of Electronic Engineering of Harbin Institute of Technology in 1994 and simulated data of two targets show that pre-whitened MUSIC can provide a better resolution and accurate determination of target number. Furthermore, a post-processing method is proposed to eliminate the sidelobe of spatial spectrum arising from the estimation errors of a noise covariance matrix