Skip to main content

Colored Noise Setting

  • 3856 Accesses

Abstract

Consider the linear stochastic system with the following state-space description.

This is a preview of subscription content, access via your institution.

Buying options

Chapter
USD   29.95
Price excludes VAT (USA)
  • DOI: 10.1007/978-3-319-47612-4_5
  • Chapter length: 11 pages
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
eBook
USD   44.99
Price excludes VAT (USA)
  • ISBN: 978-3-319-47612-4
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
Softcover Book
USD   59.99
Price excludes VAT (USA)
Hardcover Book
USD   79.99
Price excludes VAT (USA)
Fig. 5.1

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Charles K. Chui .

Exercises

Exercises

  1. 5.1.

    Let \(\{\underline{\beta }_{k}\}\) be a sequence of zero-mean Gaussian white noise and \(\{\mathbf {v}_{k}\}\) a sequence of observation data as in the system (5.1). Set

    $$\begin{aligned} \underline{\tilde{\beta }}_{k}=\left[ \begin{array}{c} 0\\ \underline{\beta }_{k} \end{array}\right] ,\qquad \mathbf {v}^{k}=\left[ \begin{array}{c} \mathbf {v}_{0}\\ \vdots \\ \mathbf {v}_{k} \end{array}\right] , \end{aligned}$$

    and define \(L(\mathbf {x},\ \mathbf {v})\) as in (4.6). Show that

    $$\begin{aligned} L\left( \underline{\tilde{\beta }}_{k},\ \mathbf {v}^{k}\right) =0. \end{aligned}$$
  2. 5.2.

    Let \(\{\underline{\gamma }_{k}\}\) be a sequence of zero-mean Gaussian white noise and \(\mathbf {v}^{{k}}\) and \(L(\mathbf {x},\ \mathbf {v})\) be defined as above. Show that

    $$\begin{aligned} L\left( \mathbf {v}_{k-1},\ \mathbf {v}^{k-1}\right) =\mathbf {v}_{k-1} \end{aligned}$$

    and

    $$\begin{aligned} L\left( \underline{\gamma }_{k},\mathbf {v}^{k-1}\right) =0. \end{aligned}$$
  3. 5.3.

    Let \(\{\underline{\gamma }_{k}\}\) be a sequence of zero-mean Gaussian white noise and \(\mathbf {v}^{{k}}\) and \(L(\mathbf {x},\ \mathbf {v})\) be defined as in Exercise 5.1. Furthermore, set

    $$\begin{aligned} \hat{\mathbf {z}}_{k-1}=L\left( \mathbf {z}_{k-1},\ \mathbf {v}^{k-1}\right) \qquad and \qquad \mathbf {z}_{k-1}=\left[ \begin{array}{c} \mathbf {x}_{k-1}\\ \underline{\xi }_{k-1} \end{array}\right] . \end{aligned}$$

    Show that

    $$\begin{aligned} \langle \mathbf {z}_{k-1}-\hat{\mathbf {z}}_{k-1},\ \underline{\gamma }_{k}\rangle =0. \end{aligned}$$
  4. 5.4.

    Let \(\{\underline{\beta }_{k}\}\) be a sequence of zero-mean Gaussian white noise and set

    $$\begin{aligned} \underline{\tilde{\beta }}_{k}=\left[ \begin{array}{c} 0\\ \underline{\beta }_{k} \end{array}\right] . \end{aligned}$$

    Furthermore, define \(\hat{\mathbf {z}}_{k-1}\) as in Exercise 5.3. Show that

    $$\begin{aligned} \langle \mathbf {z}_{k-1}-\hat{\mathbf {z}}_{k-1},\underline{\tilde{\beta }}_{k}\rangle =0. \end{aligned}$$
  5. 5.5.

    Let \(L(\mathbf {x},\ \mathbf {v})\) be defined as in (4.6) and set \(\hat{\mathbf {z}}_{0}=L(\mathbf {z}_{0},\ \mathbf {v}_{0})\) with

    $$\begin{aligned} \mathbf {z}_{0}=\left[ \begin{array}{c} \mathbf {x}_{0}\\ \underline{\xi }_{0}\end{array}\right] . \end{aligned}$$

    Show that

    $$\begin{aligned}&\quad Var(\mathbf {z}_{0}-\hat{\mathbf {z}}_{0})\\&=\left[ \begin{array}{cc} Var(\mathbf {x}_{0}) &{}\quad 0\\ -[Var(\mathbf {x}_{0})]C_{0}^{\top }[C_{0}Var(\mathbf {x}_{0})C_{0}^{\top }+R_{0}]^{-1}C_{0}[Var(\mathbf {x}_{0})] &{}\quad \\ 0 &{}\quad Q_0 \end{array}\right] . \end{aligned}$$
  6. 5.6.

    Verify that if the matrices \(M_{k}\) and \(N_{k}\) defined in (5.1) are identically zero for all k, then the Kalman filtering algorithm given by (5.185.20) reduces to the one derived in Chaps. 2 and 3 for the linear stochastic system with uncorrelated system and measurement white noise processes.

  7. 5.7.

    Simplify the Kalman filtering algorithm for the system (5.1) where \(M_{k}=0\) but \(N_{k}\ne 0\).

  8. 5.8.

    Consider the tracking system (5.22) with colored input (5.23).

    1. (a)

      Reformulate this system with colored input as a new augmented system with Gaussian white input by setting

      $$\begin{aligned}\begin{gathered} \underline{X}_{k}=\left[ \begin{array}{c} \mathbf {x}_{k}\\ \underline{\xi }_{k}\\ \eta _{k} \end{array}\right] ,\quad \underline{\zeta }_{k}=\left[ \begin{array}{c} 0\\ \underline{\beta }_{k+1}\\ \gamma _{k+1} \end{array}\right] ,\\ A_{c}=\left[ \begin{array}{ccc} A &{}\quad I &{}\quad 0\\ 0 &{}\quad F &{}\quad 0\\ 0 &{}\quad 0 &{}\quad g \end{array}\right] \quad and \quad C_{c}=[C \ 0\ 0\ 0\ 1]. \end{gathered}\end{aligned}$$
    2. (b)

      By formally applying formulas (3.25) to this augmented system, give the Kalman filtering algorithm to the tracking system (5.22) with colored input (5.23).

    3. (c)

      What are the major disadvantages of this approach?

Rights and permissions

Reprints and Permissions

Copyright information

© 2017 Springer International Publishing AG

About this chapter

Cite this chapter

Chui, C.K., Chen, G. (2017). Colored Noise Setting. In: Kalman Filtering. Springer, Cham. https://doi.org/10.1007/978-3-319-47612-4_5

Download citation