论文总字数：18207字

目 录

1 引言 6

1.1 选题的意义与背景 6

1.2 国内外现状 6

2 激光雷达理论基础 7

2.1 激光雷达的工作原理 7

2.2 激光雷达系统噪声 8

2.2.1 背景噪声 8

2.2.2 探测器噪声 8

2.2.3 信噪比 9

3 EMD基本理论 9

3.1基础概念 9

3.1.1 解析信号及其构建目的 9

3.1.2 瞬时频率 9

3.1.3 本征模态函数（IMF） 10

3.2 EMD算法的应用 10

3.3 EMD去噪方法及原理 11

3.3.1 EMD分解过程 11

3.3.2 EMD时间尺度去噪原理 12

3.3.3 EMD阈值去噪原理 12

4 激光雷达回波信号去噪方法研究 13

4.1 激光雷达回波信号仿真模型的建立 13

4.1.1 激光雷达方程 13

4.1.2 噪声模拟分析以及加噪回波信号仿真 14

4.2 EMD去噪实验 14

4.2.1 去噪判定标准 14

4.2.2 EMD分解信号 15

4.2.3 相关分量的选取 16

4.2.4 EMD去噪 16

4.2.5 不同噪声条件下去噪效果比较 18

4.2.6　总结 19

5 结论 20

参考文献 20

致谢 23

基于EMD的激光雷达回波信号去噪方法研究

卢艺

，China

窗体顶端

Abstract: The development of laser radar is inseparable from photoelectric radiation detection and electromagnetic radar technology. The former and the latter are the preconditions for laser radar launch and reception, and the development of the whole machine technology. The laser beam has the advantages of small divergence angle, energy concentration, high detection sensitivity and high resolution, so the laser radar has a high angular resolution and strong anti-interference ability and better concealment. However, in practical applications, there is often a lot of background noise and noise in the circuit system mixed in the laser radar echo signal received by the photodetector. Since the signal strength will be attenuated with the square of the distance, the greater the detection distance, the weaker the signal strength and the more likely the signal will be submerged in all types of noise. Therefore, it is imperative to effectively extract useful signals from various types of noise. The Empirical Mode Decomposition (EMD) can decompose various types of signals. Therefore, when using this method to deal with non-stationary and nonlinear data, it has very obvious advantages. A lot of practice shows that this method can be used to analyze nonlinear and non-stationary sequences. In this dissertation, aiming at the characteristics of noise in laser radar echo signals, the EMD method is used to remove the noise contained in the laser radar echo signals. The method uses EMD to decompose the lidar echo signal into multiple linear, steady-state intrinsic eigenmode functions and a trend term, and removes the noise by removing high-frequency modal components, reconstructing residual components and trend terms.In this paper, the traditional EMD denoising method, EMD hard threshold denoising method and EMD soft threshold denoising method are compared experimentally. The experimental results show that the soft threshold denoising method in the EMD threshold denoising method is smoother than the denoised signal. The signals are most similar, and the denoising effect is best in the three denoising methods studied in this paper.

Keywords: Lidar; Empirical mode decomposition; Denoising

1 引言

选题的意义与背景

激光雷达的发展离不开光电辐射探测以及电磁雷达技术。其中前者和后者分别是激光雷达发射接收的前提和整机技术发展的前提。传统雷达的载波是微波和毫米波段的电磁波，具体功能有：搜查目标；测量各种运动参数，如距离、角位置；测量特征参数，如形状、目标反射率等。相比之下，激光雷达的载波为激光，由于具有较小的发散角，能够将能量聚集在一起，因此在测量上表现出很明显的优势，即灵敏性强和分辨率高，使激光雷达具有较高的角分辨率和较强的抗干扰能力以及较好的隐蔽性。它之所以可以测定不同物质成分，原因在于分子不同，对特定波长的吸收、散射以及荧光等性能均存在差异。由于激光雷达具备如此多优势，所以在环境监测、航天、通信以及导航定位等领域有广泛应用。

剩余内容已隐藏，请支付后下载全文，论文总字数：18207字