西安工业信号检测与估计SDE_12 ch8 最小二乘估计应用.pptVIP

05-02-29 SDE_12 MLE(c) apl Ch8c 最小二乘估计应用Chapter 8c Apl of Least Squares (LS) 雷斌 gropemind@163.com Review: LSE least squares 8.4 最小二乘估计：加权LSE 线性LS问题的一种扩展形式是加权LS。 加权LSE的一般形式是 它的最小LS误差为 8.6按阶递推最小二乘估计 例：水下射弹减速规律研究 高速摄影检测到4组图像序列，位置为 全弹道速度分析方法:多项式模型的参数估计 模型阶数的确定：5阶多项式 8.7 Sequential LS序贯最小二乘估计 In Last Section 上节: In This Section 本节: ?Data Stays Fixed --?Data Length Increases ?Model Order Increases --?Model Order Stays Fixed 8.7 Sequential LS序贯最小二乘估计 例：WGN中的DC电平 Weighted Sequential LS for DC-Level Case DC电平情况的 加权序贯LS General Sequential LS Result Sequential LS Block Diagram 8.8 Constrained LS 约束最小二乘法 有时我们会遇到未知参数受限的情况Sometimes we know (or believe!) certain values are not allowed for θ Constrained约束 LS Problem Statement综述 LS Cost with a Linear Equality Constraint 线性约束 Constrained Optimization约束最优: Lagrange Multiplier LS Solution with a Linear Equality Constraint线性约束最小二乘估计 Geometry of Constrained Linear LS 8.9 Nonlinear LS 非线性最小二乘法 s(θ)≠Hθ Example of Linearization Trick: EX. 8.9 Example of Separation Trick: Ex.8.10 Iterative Methods迭代法 for Solving Nonlinear LS Newton-Raphson Solution to Nonlinear LS So the Newton-Raphson method becomes: Gauss-Newton Solution to Nonlinear LS First we linearize the model around our current estimate by using a Taylor series and keeping only the linear terms: This gives a form for the LS cost that looks like a linear problem!! Newton-Raphson vs. Gauss-Newton How do these two methods compare? 8.10 Signal Processing Examples of LS We’ll briefly look at two examples from the book… Book Examples 1.Digital Filter Design 数字滤波器设计 2.AR Parameter Estimation for the ARMA Model 3.Adaptive Noise Cancellation 自适应噪声对消器 4.Phase-Locked Loop (used in phase-coherent demodulation) Ex. 8.11 Filter Design by Prony’s LS Method Prony’s Modification to Get Linear Model Key Ideas in Prony LS Example 1. Shows power and flexibility of LS approach显示出最小二乘方法的强大能力和适应性 -There is no noise here!!! ?MVU, ML, etc. are not applicabl