低可观测目标无源协同定位技术研究

发布时间：2018-08-30 17:55

【摘要】：随着隐身技术和低空突防技术的飞速发展,雷达探测环境日益复杂,极大影响了有源雷达的目标检测跟踪性能。如何在强杂波背景下对低可观测目标进行检测跟踪,成为雷达预警领域亟待解决的关键问题。无源协同定位系统体积小,自身静默,抗干扰能力强,其空间分布特性能有效提高系统对低可观测目标的探测性能,具有重要的军事研究意义和应用价值。本文基于无源协同定位系统,对低可观测目标的检测跟踪问题展开了深入研究,主要创新点如下:首先,针对双基站无源协同定位系统单目标航迹起始及维持问题,提出一种模拟退火极大似然概率数据关联算法。该方法通过多帧测量积累构建对数似然函数,利用改进模拟退火算法优化求解,并采用滑窗批处理技术实现航迹维持。仿真结果验证了所提算法的有效性。其次,针对双基站无源协同定位系统结构单一,目标检测性能较差的问题,提出一种基于遗传算法极大似然概率多假设的多基站无源协同定位方法。该方法通过遗传算法优化求解,并对多基站信息进行融合,利用滑窗批处理实现航迹维持。与同类算法对比,所提方法显著提高了低可观测目标检测跟踪性能。最后,针对目标个数未知时的多目标检测跟踪问题,提出一种基于拟蒙特卡罗模拟退火极大似然概率多假设的多目标无源协同定位方法。该方法利用多假设原理确定目标个数,通过拟蒙特卡罗模拟退火技术实现航迹初始,采用滑窗批处理技术实现航迹维持。仿真结果显示所提方法能有效解决多目标检测跟踪问题。
[Abstract]:With the rapid development of stealth technology and low altitude penetration technology, radar detection environment is becoming more and more complex, which greatly affects the target detection and tracking performance of active radar. How to detect and track low observable targets in the background of strong clutter has become a key problem to be solved in the field of radar early warning. The passive cooperative positioning system is small in size, quiet in itself and strong in anti-jamming ability. Its spatial distribution can effectively improve the detection performance of the system to low observable targets. It has important military research significance and application value. Based on the passive cooperative positioning system, the detection and tracking of low observable targets is studied in this paper. The main innovations are as follows: firstly, the problem of track initiation and maintenance of single target in passive cooperative positioning system based on two base stations is discussed. A simulated annealing maximum likelihood probability data association algorithm is proposed. The logarithmic likelihood function is constructed by multi-frame measurement accumulation, and the improved simulated annealing algorithm is used to optimize the solution, and the sliding window batch processing technique is used to realize track maintenance. Simulation results verify the effectiveness of the proposed algorithm. Secondly, a multi-base station passive co-location method based on the maximum likelihood probability and multiple assumptions of genetic algorithm is proposed to solve the problem of single structure and poor target detection performance of the dual-base station passive co-location system. The method is optimized by genetic algorithm, and the multi-base station information is fused, and the sliding window batch processing is used to realize track maintenance. Compared with the similar algorithms, the proposed method improves the performance of low observable target detection and tracking significantly. Finally, a multi-target passive co-location method based on quasi-Monte Carlo simulated annealing maximum likelihood probability multi-hypothesis is proposed for multi-target detection and tracking when the number of targets is unknown. In this method, the number of targets is determined by the principle of multiple hypotheses, the initial track is realized by quasi-Monte Carlo simulated annealing, and the track maintenance is realized by sliding window batch processing. Simulation results show that the proposed method can effectively solve the problem of multi-target detection and tracking.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN95

【参考文献】