首页> 外文会议>International Conference on Information Fusion >Track Initiation in the Presence of Multipath Effect and Reflecting Point Uncertainties
【24h】

Track Initiation in the Presence of Multipath Effect and Reflecting Point Uncertainties

机译:在存在多径效应和反射点不确定性的情况下追踪启动

获取原文

摘要

Track initiation for a low observable target with multipath measurements can be very challenging due to reflecting point uncertainties (e.g., with Over-the-Horizon Radar (OTHR) systems and underwater target tracking). Existing algorithms cannot handle multipath measurements with reflecting point uncertainties. In OTHR tracking, reflecting point uncertainties amount to reflecting height uncertainties. Existing methods assume that this height is exactly known a priori. In practice, however, the reflecting height is time varying slowly. To initiate tracks, initial target parameters and the reflecting height are considered jointly since they depend on each other. Then an approach by embedding joint smoothing (JS) into an expectation-maximization (EM) algorithm is proposed, where the information contained in measurements and the prior information about this height are utilized. An approximate solution is presented, where a closed form is derived for track initiation based on the latest estimated reflecting height and then the current estimated reflecting height is obtained by a smoother, which has a recursive form by using the current estimate of initial target parameters. The performance of our JS-EM approach is further improved through iterative update. Simulation results show that the JS-EM approach outperforms the approach using a specific value of the reflecting height.
机译:由于反射点不确定性(例如,具有超地平线雷达(OTHR)系统和水下目标跟踪),具有多径测量的低可观察目标的轨道启动可能非常具有挑战性。现有算法无法处理多径测量,反映点不确定性。在OTHR跟踪中,反映点不确定性的数量以反映高度不确定性。现有方法假设此高度正恰恰是先验的。然而,在实践中,反射高度慢慢变化。为了启动曲目,初始目标参数和反射高度是共同认为的,因为它们彼此依赖。然后,提出了一种通过将关节平滑(JS)嵌入一个预期最大化(EM)算法的方法,其中利用测量中包含的信息和关于该高度的先前信息。提出了一种近似解决方案,其中基于最新的估计反射高度导出闭合形式,然后通过使用初始目标参数的当前估计来获得电流估计的反射高度。通过迭代更新进一步提高了我们的JS-EM方法的表现。仿真结果表明,JS-EM方法使用反射高度的特定值越高的方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号