本文選題：非均勻雜波 + 雜波邊緣��； 參考：《南京理工大學》2015年碩士論文

【摘要】：雷達信號的恒虛警率(Constant False Alarm Rate, CFAR)處理技術是現(xiàn)代雷達信號處理的重要研究內(nèi)容之一,在雷達目標自動檢測中占有不可或缺的重要地位。CFAR處理技術的理論研究雖然已取得豐碩成果但現(xiàn)有的經(jīng)典恒虛警檢測算法一般難以在不同環(huán)境中得到了性能的平衡,而大多恒虛警檢測器都面對在非均勻雜波環(huán)境下的雜波邊緣和多干擾目標環(huán)境中檢測性能下降的問題。本文在分析了幾種典型雜波模型的基礎上,針對非均勻雜波背景下的雷達恒虛警檢測技術進行了研究。論文的主要工作如下：1)在討論了瑞利分布、對數(shù)正態(tài)分布及韋布爾分布等經(jīng)典雜波分布模型的基礎上,以韋布爾分布雜波為例,采用零記憶非線性變換(ZMNL)方法進行了非相干及相關雜波模擬仿真,為后續(xù)研究工作奠定基礎。2)分析了傳統(tǒng)的均值(ML, Mean Level)類和有序統(tǒng)計量(OS, Order Statistics)類的CFAR檢測器,仿真比較了其在均勻雜波背景、多干擾目標雜波背景及存在雜波邊緣背景下的性能,為后續(xù)CFAR檢測器的設計討論提供了性能評估依據(jù)。3)針對存在雜波邊緣的非均勻雜波背景下,仿真分析了MLC-CFAR、Ⅵ-CFAR等CFAR檢測算法,針對Ⅵ-CFAR在雜波邊緣漏警概率增大和在多干擾目標環(huán)境檢測性能下降的缺點,提出了一種多策略CFAR檢測器(ACSVI-CFAR)。仿真分析表明ACSVI-CFAR檢測器既改善了Ⅵ-CFAR在均勻背景的恒虛警損失及雜波邊緣環(huán)境下的虛警概率控制能力,同時具有S-CFAR檢測器在多干擾目標背景下的優(yōu)異性能4)針對未知數(shù)目的多干擾目標雜波背景環(huán)境,在研究了排序數(shù)據(jù)方差(ODV)刪除算法的基礎上,仿真分析了基于排序數(shù)據(jù)方差的自適應刪除最大似然恒虛警(ACMLH-CFAR)檢測器,針對ACMLH-CFAR檢測器在雜波邊緣背景檢測性能的不足,采用Ⅵ-CFAR算法的均值比對其進行改進,仿真結果表明,改進后的ACMLH-CFAR在雜波邊緣性能得到了改善。
[Abstract]:Constant False Alarm rate (CFAR) processing technology of radar signal is one of the important research contents of modern radar signal processing. Although the theoretical research of CFAR processing technology has made a lot of achievements, the existing classical CFAR detection algorithms are generally difficult to achieve a performance balance in different environments. Most CFAR detectors face the problem of poor detection performance in heterogeneous clutter environment and multi-jamming target environment. In this paper, based on the analysis of several typical clutter models, the CFAR detection technology in heterogeneous clutter background is studied. The main work of this paper is as follows: (1) on the basis of discussing the classical clutter distribution models such as Rayleigh distribution, logarithmic normal distribution and Weibull distribution, we take Weibull distribution as an example. The non-coherent and correlated clutter simulation is carried out by using the zero-memory nonlinear transform (ZMNL) method, which lays a foundation for further research. (2) the traditional CFAR detectors of MLL, Mean level class and ordered statistics class are analyzed. Its performance in homogeneous clutter background, multi-jamming target clutter background and clutter edge background is compared by simulation, which provides a basis for performance evaluation for the design and discussion of CFAR detector. 3) for heterogeneous clutter background with clutter edge. The CFAR detection algorithms such as MLC-CFAR and 鈪，

本文編號：1811224