基于自更新置信分类网络的雷达点迹识别算法

多雷达协同组网进行目标探测识别时,受复杂战场环境影响,获取的数据富含大量杂波和不确定信息,传统雷达点迹识别算法在处理此类数据时具有一定局限。为此,文章提出了基于自适应置信分类网络的雷达点迹识别算法。构建置信分类网络,获取各轮迭代下雷达点迹隶属目标、杂波和不确定的置信度。然后基于点迹的空间分布特性构造决策证据并进行修正融合。融合结果促使点迹类别更新,更新点迹则再次驱动置信分类网络训练学习。如此迭代优化,直至所有雷达点迹类别标签不再更新。实测雷达点迹验证实验显示,与传统典型雷达点迹识别算法相比,新算法可有效提升点迹识别正确率。此外对训练样本依赖较小,便于推广应用。

A Radar Plots Recognition Algorithm Based on Adaptive Confidence Classification Network

When multiple radars collaborate for target detection and recognition, the obtained data is rich in clutter and uncertain information due to the complex battlefield environment. Traditional radar plots recognition algorithms have certain limitations in processing such data. Therefore, a radar plots recognition algorithm based on adaptive confidence classification network (RPR-ACCE) has been proposed in this paper. Firstly, construct a confidence classification network to obtain the belief of each radar plots belonging to target, clutter, and uncertainty that under each iteration. Then, based on the spatial distribution characteristics of these plots, decision evidences are constructed and corrected for fusion. The fusion result updates the class label of the plots, and the updated plots also drive the training of the confidence classification network again. This iterative optimization is carried out until the class labels of all radar plots are no longer updated. Experiments based on measured radar plots show that compared with traditional typical radar plot recognition algorithms, the new algorithm can effectively improve the accuracy of plot recognition. In addition, the dependence on training samples is relatively small, making it easy to promote and apply to other scenarios.