Adaptive Fusion Detection for Dual-Radar with Non-Identical Clutter Distributions
-
摘要: 经典多雷达融合检测常假设杂波幅度分布和目标信杂比一致，并在此基础上进行联合处理。然而，多部雷达的工作参数差异会导致杂波幅度分布和目标信杂比的差异，使现有检测器在实际应用时模型失配。针对这一问题，本文针对瑞利分布的S波段海杂波和韦布尔分布的X波段海杂波融合检测场景，提出了一种双雷达自适应融合检测方法。该方法根据各雷达局部检验统计量与信杂比估计结果，自适应确定融合权重与全局检验统计量。同时推导了虚警概率与检测概率关于判决门限的闭式表达式，蒙特卡洛仿真验证了闭式表达式的准确性。基于实测海杂波数据的实验表明，在任意信杂比组合的条件下，所提方法比传统的OR融合和AND融合方法具有更优的检测性能。Abstract:
Objective Small sea-surface targets, characterized by low radar cross-sections, generate echoes easily marked by intense sea clutter, resulting in extremely low signal-to-clutter ratios (SCR). Constrained by single operating frequency bands and fixed observation angles, single radar systems present notable performance bottlenecks, whereas multi-radar collaborative detection offers an effective solution to this limitation. Existing multi-radar fusion detection algorithms are derived under two ideal premises: identical clutter distributions and equal SCRs across radar nodes, assumptions that rarely hold in practice. In practice, discrepancies in radar parameters (frequency band, range resolution, and grazing angle) result in distinct statistical characteristics of sea clutter. Furthermore, target RCS fluctuates with observation azimuth and operating frequency, yielding inconsistent SCRs for the same target across different radars and inducing severe model mismatch in traditional fusion detectors. To address the coexistence of heterogeneous clutter distributions and unequal SCRs, this paper proposes a Neyman-Pearson (NP) criterion-based dual-radar adaptive fusion detector, termed NP-AFD, for Rayleigh and Weibull sea clutter backgrounds. Methods Amplitude distribution fitting is conducted on measured S-band and X-band sea clutter datasets using five mainstream models: Rayleigh, Lognormal, Weibull, Gamma, and K-distribution. Fitting accuracy is evaluated via the mean square error (MSE) of the probability density function (PDF) and complementary cumulative distribution function (CCDF), as listed in Table 1 . Based on the fitting results, local optimal test statistics are derived separately for Rayleigh and Weibull clutter backgrounds. As the two radars operate independently, the joint likelihood ratio equals the product of their individual likelihood ratios. The optimal fusion statistic is formulated as a weighted sum of the two local statistics, where adaptive weights are determined by real-time estimated SCRs, and radar channels with higher SCRs are assigned larger weights. Closed-form analytical expressions linking the false alarm probability and detection probability to the decision threshold are also derived.Results and Discussions Monte Carlo simulations over 105 independent trials verify the validity of all derived closed-form expressions. Under identical SCRs for both radars, the simulated detection curves of single radars and NP-AFD show strong agreement with theoretical curves ( Fig. 3 ). Compared with decision-level OR and AND fusion, NP-AFD consistently achieves the highest detection probability, with OR fusion ranking second and AND fusion delivering the worst. Performance gain analysis shows that NP-AFD maintains a positive gain over the better-performing single radar across all tested Weibull shape parameters. In contrast, OR fusion exhibits negative gain at low SCRs under small Weibull shape parameters, while AND fusion maintains negative gain across the full SCR range (Fig. 4 ). The simulated false alarm rate is stably controlled around the preset value of 10–3 (Fig. 5 ). Furthermore, two-dimensional joint evaluation is performed with independently varying SCRs of the two radars, and detection probability contours are plotted with the two radar SCRs as coordinates (Fig. 6 ). The results demonstrate that NP-AFD requires lower SCR combinations to achieve the same detection probability, compared with OR and AND fusion. Experiments on measured sea clutter data further verify that NP-AFD yields the highest detection probability among all compared methods (Fig. 7 ,Fig. 8 ), and its detection performance shows agreement with theoretical predictions (Fig. 9 ,Fig. 10 ).Conclusions This paper addresses the challenges of non-identical clutter distributions and varying SCRs in multi-radar collaborative detection. Based on the Neyman-Pearson criterion, a dual-radar adaptive fusion detection method, NP-AFD, is derived for Rayleigh and Weibull clutter backgrounds. The proposed method provides closed-form expressions for fusion weights, decision threshold, and detection probability. Theoretical derivations, simulations, and experiments on measured sea clutter data consistently demonstrate that NP-AFD outperforms OR fusion and AND fusion under arbitrary SCR combinations, delivering superior detection performance and robustness. -
Key words:
- sea clutter /
- multi-radar fusion detection /
- non-identical clutter distributions /
- NP-AFD
-
表 1 雷达回波幅度分布拟合均方误差
分布类型 雷达A 雷达B PDF -MSE CCDF -MSE PDF -MSE CCDF -MSE 瑞利分布 4.68×10–21 4.78×10–5 2.99×10–15 7.57×10–6 对数正态分布 7.66×10–20 7.85×10–4 3.33×10–13 9.14×10–4 韦布尔分布 7.80×10–21 6.09×10–5 2.27×10–15 5.03×10–6 伽马分布 1.18×10–20 5.15×10–5 7.71×10–14 1.11×10–4 K分布 7.38×10–21 7.00×10–5 6.25×10–15 1.80×10–5 
下载: 导出CSV
-
[1] YANG Yong and YANG Boyu. Overview of radar detection methods for low altitude targets in marine environments[J]. Journal of Systems Engineering and Electronics, 2024, 35(1): 1–13. doi: 10.23919/JSEE.2024.000026. [2] 许述文, 何绮, 茹宏涛. 基于无监督图互信息最大化的海面小目标异常检测[J]. 电子与信息学报, 2024, 46(7): 2712–2720. doi: 10.11999/JEIT230887.XU Shuwen, HE Qi, and RU Hongtao. Anomaly detection of small targets on sea surface based on deep graph infomax[J]. Journal of Electronics & Information Technology, 2024, 46(7): 2712–2720. doi: 10.11999/JEIT230887. [3] LIU Lichao, GUO Qiang, HUANG Shuai, et al. Detection of small targets in sea clutter using dual-polarization correlation features and one-class classifier[J]. IEEE Geoscience and Remote Sensing Letters, 2025, 22: 3503105. doi: 10.1109/LGRS.2025.3554337. [4] 石兆, 时晨光, 汪飞, 等. 针对多目标跟踪的组网雷达检测门限与功率分配联合优化算法[J]. 电子与信息学报, 2024, 46(5): 2065–2075. doi: 10.11999/JEIT231242.SHI Zhao, SHI Chenguang, WANG Fei, et al. Joint detection threshold and power allocation optimization strategy for multi-target tracking in radar networks[J]. Journal of Electronics & Information Technology, 2024, 46(5): 2065–2075. doi: 10.11999/JEIT231242. [5] CHAIR Z and VARSHNEY P K. Optimal data fusion in multiple sensor detection systems[J]. IEEE Transactions on Aerospace and Electronic Systems, 1986, AES-22(1): 98–101. doi: 10.1109/TAES.1986.310699. [6] GUPTA K, MERCHANT S N, and DESAI U B. A multistage approach to decision fusion using a distributed network of non-identical nodes[J]. Signal Processing, 2016, 129: 106–118. doi: 10.1016/j.sigpro.2016.06.003. [7] 胡勤振, 苏洪涛, 周生华, 等. 多基地雷达中双门限CFAR检测算法[J]. 电子与信息学报, 2016, 38(10): 2430–2436. doi: 10.11999/JEIT151163.HU Qinzhen, SU Hongtao, ZHOU Shenghua, et al. Double threshold CFAR detection for multisite radar[J]. Journal of Electronics & Information Technology, 2016, 38(10): 2430–2436. doi: 10.11999/JEIT151163. [8] 周生华, 姜昊志, 窦法兵, 等. 分布式雷达信号级融合检测的数据压缩与组网架构设计[J]. 现代雷达, 2024, 46(9): 30–36. doi: 10.16592/j.cnki.1004-7859.2024.09.005.ZHOU Shenghua, JIANG Haozhi, DOU Fabing, et al. Data compression and network design for signal fusion based distributed radar[J]. Modern Radar, 2024, 46(9): 30–36. doi: 10.16592/j.cnki.1004-7859.2024.09.005. [9] 龚树凤, 龙伟军, 贲德, 等. 组网雷达自适应模糊CFAR检测融合算法[J]. 系统工程与电子技术, 2022, 44(1): 100–107. doi: 10.12305/j.issn.1001-506X.2022.01.14.GONG Shufeng, LONG Weijun, BEN De, et al. Adaptive fuzzy CFAR detection fusion algorithm for netted radar[J]. Systems Engineering and Electronics, 2022, 44(1): 100–107. doi: 10.12305/j.issn.1001-506X.2022.01.14. [10] 王楠, 许蕴山, 夏海宝, 等. 多基地雷达自适应CFAR检测融合算法[J]. 信号处理, 2018, 34(7): 818–823. doi: 10.16798/j.issn.1003-0530.2018.07.008.WANG Nan, XU Yunshan, XIA Haibao, et al. An adaptive fusion algorithm of CFAR detection for multisite radar[J]. Journal of Signal Processing, 2018, 34(7): 818–823. doi: 10.16798/j.issn.1003-0530.2018.07.008. [11] FANTE R L. Multifrequency detection of a slowly fluctuating target[J]. IEEE Transactions on Aerospace and Electronic Systems, 1996, 32(1): 495–497. doi: 10.1109/7.481294. [12] FISHLER E, HAIMOVICH A, BLUM R S, et al. Spatial diversity in radars—models and detection performance[J]. IEEE Transactions on Signal Processing, 2006, 54(3): 823–838. doi: 10.1109/TSP.2005.862813. [13] ZHOU S H and LIU H W. Signal fusion-based target detection algorithm for spatial diversity radar[J]. IET Radar, Sonar & Navigation, 2011, 5(3): 204–214. doi: 10.1049/iet-rsn.2010.0100. [14] CHEN Peng, ZHENG Le, WANG Xiaodong, et al. Moving target detection using colocated MIMO radar on multiple distributed moving platforms[J]. IEEE Transactions on Signal Processing, 2017, 65(17): 4670–4683. doi: 10.1109/TSP.2017.2714999. [15] LI Ping, HUANG Bang, JIA Wenkai, et al. Adaptive detection of distributed target for FDA-MIMO radar in compound-Gaussian clutter[J]. IEEE Transactions on Aerospace and Electronic Systems, 2025, 61(5): 11382–11393. doi: 10.1109/TAES.2025.3567279. [16] 丁昊, 董云龙, 刘宁波, 等. 海杂波特性认知研究进展与展望[J]. 雷达学报, 2016, 5(5): 499–516. doi: 10.12000/JR16069.DING Hao, DONG Yunlong, LIU Ningbo, et al. Overview and prospects of research on sea clutter property cognition[J]. Journal of Radars, 2016, 5(5): 499–516. doi: 10.12000/JR16069. [17] 连静, 杨勇, 谢晓霞, 等. 大掠射角对海雷达导引头实测回波特性分析[J]. 系统工程与电子技术, 2024, 46(5): 1535–1543. doi: 10.12305/j.issn.1001-506X.2024.05.08.LIAN Jing, YANG Yong, XIE Xiaoxia, et al. Analysis of radar seeker sea-surface echoes at a high grazing angle[J]. Systems Engineering and Electronics, 2024, 46(5): 1535–1543. doi: 10.12305/j.issn.1001-506X.2024.05.08. [18] 周围, 朱勇, 杜玉晗, 等. Gamma分布在海面目标检测中的应用[J]. 雷达科学与技术, 2021, 19(3): 310–321. doi: 10.3969/j.issn.1672-2337.2021.03.012.ZHOU Wei, ZHU Yong, DU Yuhan, et al. Applications of Gamma distribution in sea clutter modelling and maritime target detection[J]. Radar Science and Technology, 2021, 19(3): 310–321. doi: 10.3969/j.issn.1672-2337.2021.03.012. [19] HENRIQUES-RODRIGUES L, CAEIRO F, and GOMES M I. Improvements in the estimation of the Weibull tail coefficient: A comparative study[J]. Mathematical Methods in the Applied Sciences, 2024, 47(11): 8255–8274. doi: 10.1002/mma.10013. [20] 张坤, 水鹏朗, 王光辉. 相参雷达K分布海杂波背景下非相干积累恒虚警检测方法[J]. 电子与信息学报, 2020, 42(7): 1627–1635. doi: 10.11999/JEIT190441.ZHANG Kun, SHUI Penglang, and WANG Guanghui. Non-coherent integration constant false alarm rate detectors against K-distributed sea clutter for coherent radar systems[J]. Journal of Electronics & Information Technology, 2020, 42(7): 1627–1635. doi: 10.11999/JEIT190441. [21] WANG Xinyang, LI Yang, and ZHANG Ning. A robust variability index CFAR detector for Weibull background[J]. IEEE Transactions on Aerospace and Electronic Systems, 2023, 59(2): 2053–2064. doi: 10.1109/TAES.2022.3206256. [22] ZHANG Baiqiang, ZHOU Jie, XIE Junhao, et al. Weighted likelihood CFAR detection for Weibull background[J]. Digital Signal Processing, 2021, 115: 103079. doi: 10.1016/j.dsp.2021.103079. [23] AUBRY A, CAROTENUTO V, DE MAIO A, et al. Testing stationarity and statistical independence of multistatic/polarimetric sea-clutter with application to NetRAD data[J]. IEEE Transactions on Geoscience and Remote Sensing, 2024, 62: 5103415. doi: 10.1109/TGRS.2024.3362872. -
下载:
图(10) / 表(1)
计量
- 文章访问数: 31
- HTML全文浏览量: 6
- PDF下载量: 1
- 被引次数: 0
