The increased trend toward multisensor target tracking system is driving an interest for distributed tracks fusion algorithm. In such a system, local tracks formed by each sensor may arrive in the fusion center with temporal out-of-sequence because of different sensor data processing time and random communication delay, moreover, the priori information on the origins of tracks to be fused is usually unknown due to clutter disturbance and presence of multitargets. This paper considers the problem of fusing out-of-sequence tracks (OOSTs) with track origin uncertainty in a distributed fusion setup, and proposes a novel all neighbor fusion-integrated forward prediction fusion and decorrelation (ANF-IFPFD) method. The proposed ANF-IFPFD enumerates and probabilistically evaluates all feasible track-to-track association events, and fuses the central tracks with extracted information purely contributed by the local OOSTs through an information decorrelation process. Furthermore, it enables to operate the false track discrimination (FTD) in the fusion center by using the fused probability of target existence as a track quality measure. Additionally, two implementation configurations of the proposed ANF-IFPFD method are also designed to provide a trade-off between the tracking performance and the system communication bandwidth consumption. Monte Carlo simulations are carried out to validate the superiority of the proposed ANF-IFPFD over the enhanced state-of-art methods, in terms of both tracking performance and also computation-storage requirement. Compared to the configuration without any information feedback, the partial information feedback configuration of the ANF-IFPFD method is verified to deliver intensive improvements for both local tracking and tracks fusing.

朝着多传感器目标跟踪系统发展的趋势正在引起人们对分布式跟踪融合算法的兴趣。在这样的系统中，由于不同的传感器数据处理时间和随机的通信延迟，由每个传感器形成的局部轨迹可能会以时间上的不顺序到达融合中心，此外，关于要融合的轨迹的起源的先验信息是通常由于杂波干扰和多目标的存在而未知。本文考虑了在分布式融合设置中融合乱序轨道（OOSTs）和轨道原点不确定性的问题，并提出了一种新颖的全邻域融合前向预测融合与去相关（ANF-IFPFD）方法。拟议的ANF-IFPFD列举并概率评估了所有可行的音轨间关联事件，并通过信息去相关过程将中央轨道与本地OOST纯粹贡献的提取信息融合在一起。此外，通过使用目标存在的融合概率作为轨道质量度量，它可以在融合中心中操作错误的轨道判别（FTD）。此外，还设计了所提出的ANF-IFPFD方法的两种实现配置，以在跟踪性能和系统通信带宽消耗之间进行权衡。在跟踪性能和计算存储需求方面，进行了蒙特卡洛模拟以验证所提出的ANF-IFPFD在增强型现有技术上的优越性。与没有任何信息反馈的配置相比，