论文部分内容阅读
本文研究密集多回波环境下的机动多目标数据关联问题.通过对联合概率数据关联(JPDA)方法性能特征的分析,将其归结为一类约束组合优化问题,进而应用随机神经网络Bolltzmann机的组合优化求解策略,结合改进的模拟增益退火方法,提出了一种新颖有效的机动多目标快速随机神经数据关联组合优化算法(FSNJPDA),克服了传统JPDA存在出现的计算组合爆炸现象.仿真结果表明,该方法不仅收敛速度快,而且计算量小,关联效果好,回波愈密集,其优越性能愈为突出.
This paper studies the maneuvering multi-target data association problem in dense multi-echo environment. By analyzing the performance characteristics of the joint probability data association (JPDA) method, it is reduced to a class of constrained combinatorial optimization problems. Then, the stochastic neural network Bolltzmann machine is used to solve the combinatorial optimization strategy. Combined with the improved simulation gain annealing method, A novel and efficient algorithm for the combination of multiple targets and fast stochastic neural network (FSNJPDA) is proposed, which overcomes the computational combinatorial explosion phenomenon of traditional JPDA. Simulation results show that the proposed method not only has the advantages of fast convergence speed, small amount of calculation, good correlation effect and dense echo, the superiority of the method is more prominent.