| 摘 要: 针对三维无源时差定位(TDOA)非线性方程组求解中的凸优化难题,提出了一种基于混合鲸鱼灰狼优化算法(HWWOA)的解决方案。首先,通过Chan式算法优化种群初始化,增强了种群多样性。其次,改进适应度函数,消除观测站之间的差异,降低定位误差。算法策略中,结合灰狼算法和莱维飞行,提升了全局搜索能力,并通过贪婪选择策略保留最优解。仿真实验结果显示:HWWOA在近场场景下的定位正确率基本保持一致,在临界点场景下的定位准确率高达99.5%,较其他算法至少提升了15%,500次迭代内适应度值下降更快;定位误差曲线更接近克拉美罗下限(CRLB),算法鲁棒性显著提升。该算法结构简单,参数少,易于实现,具有较好的实用价值和应用前景。 |
| 关键词: 混合鲸鱼灰狼优化算法 无源时差定位 适应度函数 莱维飞行 |
|
中图分类号: TP391
文献标识码: A
|
| 基金项目: 陕西省重点研发计划项目(2020GY-091);陕西科技大学博士科研基金(2020BJ-49) |
|
| Research on 3D Passive TDOA Localization Technology Based on Hybrid Whale Grey Wolf Optimization Algorithm |
|
TAN Xin1, ZHAO Dongyan1, ZHU Hongxi2, WU Bai1, ZHANG Ying1
|
(1.School of Electronic Information and Artificial Intelligence, Shaanxi University of Science & Technology, Xi’an, 710021, China;
2.BZT Electronics, Xi’an, 710199, China)
tanxin@sust.edu.cn; 221611048@sust.edu.cn; 271674351@qq.com; conwobo@outlook.com; 231612117@sust.edu.cn
|
| Abstract: To address the convex optimization challenge in solving nonlinear equations for 3D passive Time Difference of Arrival (TDOA) localization, a solution based on the Hybrid Whale Grey Wolf Optimization Algorithm (HWWOA) is proposed. Firstly, population initialization is optimized via Chan’s algorithm to enhance diversity.Secondly, the fitness function is refined to eliminate systematic errors between observation stations and reduce localization inaccuracies. Within the algorithmic framework, the integration of the Grey Wolf Optimizer (GWO) and Lévy flight bolsters global search capability, while a greedy selection strategy preserves elite solutions. Simulation results show that: HWWOA maintains basically consistent localization accuracy in nea-r field scenarios, achieves a
localization accuracy of up to 99.5% in critica-l point scenarios, improves by at least 15% compared with other algorithms, and has a faster decline in fitness values within 500 iterations. The localization error curve is closer to the Cramé-r Rao Lower Bound (CRLB), and the algorithm's robustness is significantly improved. Its simple architecture, minimal parameters, and ease of implementation underscore significant practical value and promising application prospects. |
| Keywords: Hybrid Whale Grey Wolf Optimization Algorithm (HWWOA) Time Difference of Arrival (TDOA) fitness function Lévy flight |
