-
通常理想目标表面材料的光散射特性复杂,理论上由粗糙面散射理论可获知实验数值,但实际上很难提供计算所需的粗糙面元微观特性如相关长度、统计分布等。因此,通常使用双向反射分布函数BRDF来描述目标面元的光散射特征。BRDF所描述的光散射特征分布和粗糙面元的微观特性、材料介电常数、入射光波长等因素紧密联系。通常粗糙面元会被看作是由大量微面元构成,且所有微面元的反射都遵循菲涅耳效应[10-11]。
如图 2所示,对于小面元dS,入射光源方向为(θi, φi),探测器的观测方向为(θr, φr)。其中θ是天顶角,φ是方位角,z是代表粗糙面元的法线。BRDF定义为入射到粗糙表面的光辐射照度与光辐射亮度相联系,在图中表示为dLr(θi,φi,θr,φr)与沿(θi,φi)方向入射到被照射面元的亮度dEi(θi, φi)之比,即:
$ {f_{\rm{r}}}\left( {{\theta _{\rm{i}}}, {\varphi _{\rm{i}}}, {\theta _{\rm{r}}}, {\varphi _{\rm{r}}}} \right) = \frac{{{\rm{d}}{L_r}\left( {{\theta _{\rm{i}}}, {\varphi _{\rm{i}}}, {\theta _{\rm{r}}}, {\varphi _{\rm{r}}}} \right)}}{{{\rm{d}}E({\theta _{\rm{i}}}, {\varphi _{\rm{i}}})}}({\rm{s}}{{\rm{r}}^{ - 1}}) $
(1) 式中, fr(θi, φi, θr, φr)可以简化为fr(θi, θr, φ),φ=φr-φi为小面元坐标系内入射角与反射角的差值。
-
根据粗糙面元的BRDF样片材料,可以获知样片表面的光散射特征。但由于直接获取样片BRDF很困难,实验次数有限,不能多次重复实验获取数据。一般情况下是采用所得到的部分BRDF实验值拟合五参量统计模型,五参量模型基于粗糙面元的目标光散射计算获得,最终给出了BRDF的五参量统计模型[12]:
$ \begin{array}{l} {f_{\rm{r}}}({\theta _{\rm{i}}}, {\theta _{\rm{r}}}, {\varphi _{\rm{r}}}) = \frac{{{k_{\rm{b}}}{k_{\rm{r}}}^2{\rm{cos}}\alpha }}{{1 + \left( {{k_{\rm{r}}}^2 - 1} \right){\rm{cos}}\alpha }}{\rm{exp}}\left[ {b \times } \right.\\ \;\;\;\;\;\;\;\;\;\;\left. {{{\left( {1 - {\rm{cos}}\gamma } \right)}^a}} \right]\frac{{G({\theta _{\rm{i}}}, {\theta _{\rm{r}}}, {\varphi _{\rm{r}}})}}{{{\rm{cos}}{\theta _{\rm{r}}}}} + {k_{\rm d}} \end{array} $
(2) 式中, kb,kd,kr,a,b为待定参量。kb为镜反射系数,kd为漫反射系数,kr为与微面元表面法线分布函数相关的参量,a和b是与表面菲涅耳函数相关的系数。根据能量守恒定律,kd+kb≤1,θi, θr, φr分别为入射角、散射角和相对方位角,γ为微观平面上本地坐标系的入射角。(2)式前半部分表示的是双向反射分布函数的相干散射,其中粗糙表面上偏振二向分布函数用kr2×cosα/[1+(kr2-1)cosα]标记,α为微面元平面法线与粗糙面元平面法线的夹角[13-14],菲涅耳反射系数用exp[b(1-cosγ)]来表示,遮蔽系数用G(θi, θr, φr)来表示。
遗传算法的收敛速度极快,它是一种全局优化概率算法,可以很快找到最佳解。而且各态历经性使其能够有效进行全局优化搜索,减少计算时间,保证算法有效性基础上得到参量值。因此可以通过遗传算法来计算五参量模型的参量值[15]。选取样片1和2的实测BRDF数据来进行优化建模,波长和五参量如表 1所示。
Table 1. Five parameters data table of sample piece
wavelength/μm kb kd kr a b error/% 0.905 1.436 0.109 26.36 0.411 -12.85 3.07 0.86 3.201 0.047 0.681 0.595 -98.04 4.71 选取不同入射波长的BRDF实测数据来进行优化建模并与实测数据相比较[16],入射波长分别为0.86μm和0.905μm。如图 3所示,实线为实测数据,虚线为五参量模型计算结果。可知对于不同材料的BRDF,目标回波的形状基本相似,幅值不同。而且随着入射角的增大,峰值功率逐渐增大。由于脉冲激光引信的探索过程是高速变化的,其入射角的波动范围值较广。因此一般检验时应选择在较大入射角的情况下进行[17-18],以便回波信号可以在能量较少时也能够成功实现检测。模拟曲线能很好地符合实验数据,证明了建模结果的正确性。
-
针对引信工作中的实际环境,如在云烟环境中,结合脉冲激光发射系统信号处理单元对目标回波的判别,提出了基于回波包络上升速率的区域联合判别方式,并通过实际的实验测试得出数据。测试实验中用表面反射率接近实体目标的白板代替,小型目标表面反射率约为0.33,大中型目标表面反射率约为0.9,具体以不同尺寸白板来区分各型号的探测目标,云烟环境用烟雾机产生的烟雾模拟。由于研究的引信探测视场呈环形,则每一路形成的探测视场角为90°。因此,设置3个测试角度,分别为中心点0°、左侧45°和右侧45°。
如图 5所示,通过背景板可以看到,激光光束照射的视场角以对称形式左右可分别到达45°,测试得到的光束光斑分布均匀。系统设计的小型目标作用距离为3m,中型目标作用距离为5m,大型目标作用距离为7m。在没有干扰下, 不同场景目标测试记录数据如表 2和表 3所示。
Table 2. Objective test data record sheet under ideal environment
targettype range number of measurement echo signal amplitude/mV left 45° center 0° right 45° small object 3m No.1 240 460 250 No.2 260 400 270 No.3 220 280 310 medium target 5m No.1 210 300 130 No.2 150 300 110 No.3 460 600 280 large target 7m No.1 130 200 160 No.2 110 110 110 No.3 280 350 250 Table 3. Objective test data record sheet under cloud and smoke environment
target type range number of measurement echo signal amplitude/mV left 45° center 0° right 45° small object 3m No.1 250 360 210 No.2 260 280 250 No.3 210 220 210 medium target 5m No.1 310 260 230 No.2 150 210 180 No.3 160 230 240 large target 7m No.1 200 190 180 No.2 130 130 130 No.3 130 140 130 将模拟目标白板在不同作用距离点上下移动,找到反射点,通过示波器显示反射回波,读取回波幅值。利用数据记录仪记录回波包络图如图 6a所示。
从记录结果可以看到, 在中心位置即发射激光直射到目标时, 发射回波的幅值最大,位于探测视场两侧的光束能量略有下降。由于作用距离变大,反射回波幅值会降低,这符合激光传输能量的变化趋势。图中矩形框内包络的上升速率呈递增趋势,到达峰值后稳定持续一段时间,包络走势满足对目标识别的判定准则。
使用烟雾机在装置探测视场内间隔喷出烟雾来模拟实际环境,测试结果如表 3所示。
在实际环境中,系统正常工作,比较表 2数据,回波幅值大小有所降低,偏差约为65.3%。这是因为激光脉冲抵抗干扰而损失能量,但在判定准则的应用下依然实现对目标的判定。利用数据记录仪记录此测试环境下的回波包络如图 6b所示。
与图 6a相比,系统探测到的回波包络图出现明显的差异,在探测光路中脉冲激光由于实际环境如云烟的影响,激光反射率较低,因此, 系统接收到此部分反射回波后作出的反应会出现小幅的的包络响应。当激光穿过云烟到达目标表面时,反射回波到达接收系统后会出现幅度较大的反应目标的包络图形。从包络图中可以明显看出, 目标形成的包络在前沿上升速率较快,且达到峰值后稳定持续一段时间。对比图中方框1和方框2可知,干扰形成的包络在达到某一峰值后会迅速下降,不会形成持续稳定得包络现象,因此, 系统在探测过程中依据判定准则较好地实现了抗干扰。
基于激光引信的回波仿真及抗干扰研究
Echo simulation and anti-jamming research based on laser fuze
-
摘要: 为了提取目标回波信号的特征信息、提高回波信号仿真精确度,采用地面验证实验时将回波数据记录下来对其数字仿真的方法,将二次曲面单元作为几何建模的基础单元来建立目标几何模型,基于目标表面的材料及双向反射分布函数实测数据,结合遗传优化算法对不同波长的样品从不同角度来进行五参量建模。采用回波包络上升速率的区域联合判别方式,对不同环境下目标的点回波幅值进行了实验验证。结果表明,电压在云烟环境与理想环境下的偏差约为65.3%,包络走势满足对目标识别的判定准则,且在云烟环境中实现了抗干扰,能够修正改善目标在云烟环境下回波信号的仿真模型,提升仿真精确度。Abstract: In order to extract the characteristic information of the target echo signal and improve the simulation accuracy of the echo signal, a digital simulation method was adopted to record the echo data in the ground verification test. Quadric surface element was used as the basic element of geometric modeling to establish the target geometric model. Based on the material of the target surface and the measured data of bidirectional reflectnace distrbution function (BRDF), five-parameter modeling was carried out for samples of different wavelength from different angles by combining genetic optimization algorithm. The amplitude of point return wave of target under different environment was verified by the method of region joint discrimination of echo envelope rising rate. The results show that the deviation between the voltage in the cloud environment and the ideal environment is about 65.3%, the enveloped trend satisfies the criterion of target recognition, and the anti-interference is realized in the cloud environment, which can modify and improve the simulation model of target echo signal in the cloud and smoke environment and improve the simulation accuracy.
-
Key words:
- laser technique /
- the signal simulation /
- BRDF /
- echo envelope /
- five parameters /
- anti-interference
-
Table 1. Five parameters data table of sample piece
wavelength/μm kb kd kr a b error/% 0.905 1.436 0.109 26.36 0.411 -12.85 3.07 0.86 3.201 0.047 0.681 0.595 -98.04 4.71 Table 2. Objective test data record sheet under ideal environment
targettype range number of measurement echo signal amplitude/mV left 45° center 0° right 45° small object 3m No.1 240 460 250 No.2 260 400 270 No.3 220 280 310 medium target 5m No.1 210 300 130 No.2 150 300 110 No.3 460 600 280 large target 7m No.1 130 200 160 No.2 110 110 110 No.3 280 350 250 Table 3. Objective test data record sheet under cloud and smoke environment
target type range number of measurement echo signal amplitude/mV left 45° center 0° right 45° small object 3m No.1 250 360 210 No.2 260 280 250 No.3 210 220 210 medium target 5m No.1 310 260 230 No.2 150 210 180 No.3 160 230 240 large target 7m No.1 200 190 180 No.2 130 130 130 No.3 130 140 130 -
[1] ZHANG H, WANG Y T. Simulation technology of laser fuze echo signal based on monte carlo method [J]. Journal of System Simulation, 2004, 16(8):1624-1626(in Chinese). [2] TAN Y Y, ZHANG H, ZHA B T. Simulation method of underwater laser fuze echo based on bidirectional reflection function [J]. Acta Photonica, 2016, 45(12):1207002(in Chinese). doi: 10.3788/gzxb20164512.1207002 [3] ZHANG J G, LIANG X G, TANG J. A method for calculating target echo signal of laser fuze [J]. Guidance and Fuze, 2008, 29(1):22-27(in Chinese). [4] CHEN Sh Sh, ZHANG H, XU X B, et al. Study on echo characteristics of pulsed laser fuze detection of planar targets [J]. Journal of Military Engineering, 2018, 39(6):12-15(in Chinese). [5] NIU Q P, GAO Ch. Study on target echo range image characteristics of laser fuze [J]. Computer Measurement and Control, 2013, 21(11): 128-129(in Chinese). [6] WANG B, LIN J X, TONG G D, et al. Echo simulation of complex target laser fuze [J]. Guidance and Fuze, 2012, 33(4): 24-30(in Chinese). doi: 10.4028/www.scientific.net/AMR.571.332 [7] QIAN R Ch, GUI Y N, DONG W B, et al. Variable coefficient correlation detection of laser fuze pulse echo [J]. Chinese Journal of Detection and Control, 2014, 36(5):1-5(in Chinese). [8] MA Y. Study on optimization modeling and application of BRDF on near-field target surface of laser fuze[D].Nanjiang: Nanjing University of Technology, 2015: 11-14(in Chinese). [9] WEI T F. Statistical modeling and application of BRDF optimization [D].Xi'an: University of Electronic Science and Technology, 2012: 22-26(in Chinese). [10] LI H P, LI G Y, CAI Zh J, et al. Full waveform lidar echo decomposition method [J]. Journal of Remote Sensing, 2019, 23 (1): 89-98(in Chinese). [11] LU M, KONG D H, SU Y D. Target identification method of anti-radiation missile laser fuze based on PCA [J]. Chinese Journal of Detection and Control, 2019, 41(1):15-18(in Chinese). [12] LI J, HUANG Zh. Analysis and research on echo signal processing method of laser fuze [J]. Communications Technology, 2009, 42(11):217-218(in Chinese). [13] WANG H J, GUO H J, LI J H. Optimization algorithm of ultrasonic echo signal envelop correlation delay estimation [J]. Computer Engineering and Application, 2012, 48(20):154-157(in Chinese). [14] CHENG Z, XIA M, LI W. Underwater lidar echo extraction based on variable forgetting factor RLS algorithm [J]. Advances in Laser and Optoelectronics, 2016, 53(1):32-39(in Chinese). [15] WANG T, SHEN Y H, YAO J Q. Study on de-noising of laser radar echo signal based on wavelet threshold method [J]. Laser Technology, 2019, 43(1):63-68(in Chinese). doi: 10.1016/j.optlaseng.2013.02.011 [16] YU X, MIN M, ZHANG X Y, et al. Effects of typical filters on the echo signal of satellite borne high spectral resolution lidar channel at 532nm [J]. Infrared and Laser Engineering, 2018, 47(12):122-131(in Chinese). [17] DAI C, WANG Y Q, XU F. 3-D laser radar echo decomposition based on particle swarm optimization [J]. Laser Technology, 2016, 40(2): 284-287(in Chinese). [18] CHEN H M, LIU Y, ZHU X W, et al. Simulation and analysis of echo characteristics of frequency-modulated continuous wave laser fuze [J]. Chinese Journal of Military Engineering, 2015, 36(12):2247-2253(in Chinese). [19] LING J J, LI G, ZHANG R B. Modeling and simulation of polarization spectrum BRDF[J].Spectroscopy and Spectral Analysis, 2016, 36(1):42-46(in Chinese).