<< Chapter < Page Chapter >> Page >
  • If A is case 1a, (square and nonsingular), then
    A + = A - 1
  • If A is case 2a or 2b, (over specified) then
    A + = [ A T A ] - 1 A T
  • If A is case 3a, (under specified) then
    A + = A T [ A A T ] - 1

Figure 2. Four Cases with Analytical Solutions

Fortunately, most practical cases are one of these four but even then, it is generally faster and less error prone to use special techniques on the normal equationsrather than directly calculating the inverse matrix. Note the matrices to be inverted above are all r by r ( r is the rank) and nonsingular. In the other six cases from the ten in Figure 1, these would be singular, so alternate methods suchas SVD must be used [link] , [link] , [link] .

In addition to these four cases with “analytical” solutions, we can pose a more general problem by asking for an optimal approximation with a weighted norm [link] to emphasize or de-emphasize certain components or range of equations.

  • If A is case 2a or 2b, (over specified) then the weighted error pseudoinverse is
    A + = [ A T * W T * W A ] - 1 A T * W T * W
  • If A is case 3a, (under specified) then the weighted norm pseudoinverse is
    A + = [ W T W ] - 1 A T A [ W T W ] - 1 A T - 1

Figure 3. Three Cases with Analytical Solutions and Weights

These solutions to the weighted approxomation problem are useful in their own right but also serve as the foundation to the Iterative Reweighted Least Squares (IRLS)algorithm developed in the next chapter.

Geometric interpretation and least squares approximation

A particularly useful application of the pseudo-inverse of a matrix is to various least squared error approximations [link] , [link] . A geometric view of the derivation of the normal equations can be helpful. If b does not lie in the range space of A , an error vector is defined as the difference between A x and b . A geometric picture of this vector makes it clear that for the length of ε to be minimum, it must be orthogonal to the space spanned by the columns of A . This means that A * ε = 0 . If both sides of [link] are multiplied by A * , it is easy to see that the normal equations of [link] result in the error being orthogonal to the columns of A and, therefore its being minimal length. If b does lie in the range space of A , the solution of the normal equations gives the exact solution of [link] with no error.

For cases 1b, 1c, 2c, 2d, 3a, 3b, and 3c, the homogeneous equation [link] has non-zero solutions. Any vector in the space spanned by these solutions (the null space of A ) does not contribute to the equation error ε defined in [link] and, therefore, can be added to any particular generalized solution of [link] to give a family of solutions with the same approximation error. If the dimension of the nullspace of A is d , it is possible to find a unique generalized solution of [link] with d zero elements. The non-unique solution for these seven cases can be written in the form [link] .

x = A + b + [ I - A + A ] y

where y is an arbitrary vector. The first term is the minimum norm solution given by the Moore-Penrose pseudo-inverse A + and the second is a contribution in the null space of A . For the minimum | | x | | , the vector y = 0 .


To derive the necessary conditions for minimizing q in the overspecified case, we differentiate q = ϵ T ϵ with respect to x and set that to zero. Starting with the error

q = ϵ T ϵ = [ Ax - b ] T [ Ax - b ] = x T A T A x - x T A T b - b T A x + b T b
q = x T A T A x - 2 x T A T b + b T b

and taking the gradient or derivative gives

x q = 2 A T A x - 2 A T b = 0

which are the normal equations in [link] and the pseudoinverse in [link] and [link] .

If we start with the weighted error problem

q = ϵ T W T W ϵ = [ Ax - b ] T W T W [ Ax - b ]

using the same steps as before gives the normal equations for the minimum weighted squared error as

A T W T W A x = A T W T W b

and the pseudoinverse as

x = [ A T W T W A ] - 1 A T W T W b

To derive the necessary conditions for minimizing the Euclidian norm | | x | | 2 when there are few equations and many solutions to [link] , we define a Lagrangian

L ( x , μ ) = | | W x | | 2 2 + μ T ( Ax - b )

take the derivatives in respect to both x and μ and set them to zero.

x L = 2 W T W x + A T μ = 0


μ L = Ax - b = 0

Solve these two equation simultaneously for x eliminating μ gives the pseudoinverse in [link] and [link] result.

x = [ W T W ] - 1 A T A [ W T W ] - 1 A T - 1 b

Because the weighting matrices W are diagonal and real, multiplication and inversion is simple. These equations are used in the Iteratively Reweighted LeastSquares (IRLS) algorithm described in the next chapter.


To deal with measurement error and data noise, a process called “regularization" is sometimes used [link] , [link] , [link] .

Least squares approximation with constraints

The solution of the overdetermined simultaneous equations is generally a least squared error approximation problem. A particularly interesting anduseful variation on this problem adds inequality and/or equality constraints. This formulation has proven very powerful in solving theconstrained least squares approximation part of FIR filter design [link] . The equality constraints can be taken into account by using Lagrange multipliers and the inequality constraints can use theKuhn-Tucker conditions [link] , [link] , [link] . The iterative reweighted least squares (IRLS) algorithm described in the next chaptercan be modified to give results which are an optimal constrained least p-power solution [link] , [link] , [link] .


There is remarkable structure and subtlety in the apparently simple problem of solving simultaneous equations and considerable insight can begained from these finite dimensional problems. These notes have emphasized the l 2 norm but some other such as l and l 1 are also interesting. The use of sparsity [link] is particularly interesting as applied in Compressive Sensing [link] , [link] and in the sparse FFT [link] . There are also interesting and important applications ininfinite dimensions. One of particular interest is in signal analysis using wavelet basis functions [link] . The use of weighted error and weighted norm pseudoinverses provide a base for iterative reweighted leastsquares (IRLS) algorithms.

Questions & Answers

how can chip be made from sand
Eke Reply
are nano particles real
Missy Reply
Hello, if I study Physics teacher in bachelor, can I study Nanotechnology in master?
Lale Reply
no can't
where we get a research paper on Nano chemistry....?
Maira Reply
nanopartical of organic/inorganic / physical chemistry , pdf / thesis / review
what are the products of Nano chemistry?
Maira Reply
There are lots of products of nano chemistry... Like nano coatings.....carbon fiber.. And lots of others..
Even nanotechnology is pretty much all about chemistry... Its the chemistry on quantum or atomic level
no nanotechnology is also a part of physics and maths it requires angle formulas and some pressure regarding concepts
Preparation and Applications of Nanomaterial for Drug Delivery
Hafiz Reply
Application of nanotechnology in medicine
has a lot of application modern world
what is variations in raman spectra for nanomaterials
Jyoti Reply
ya I also want to know the raman spectra
I only see partial conversation and what's the question here!
Crow Reply
what about nanotechnology for water purification
RAW Reply
please someone correct me if I'm wrong but I think one can use nanoparticles, specially silver nanoparticles for water treatment.
yes that's correct
I think
Nasa has use it in the 60's, copper as water purification in the moon travel.
nanocopper obvius
what is the stm
Brian Reply
is there industrial application of fullrenes. What is the method to prepare fullrene on large scale.?
industrial application...? mmm I think on the medical side as drug carrier, but you should go deeper on your research, I may be wrong
How we are making nano material?
what is a peer
What is meant by 'nano scale'?
What is STMs full form?
scanning tunneling microscope
how nano science is used for hydrophobicity
Do u think that Graphene and Fullrene fiber can be used to make Air Plane body structure the lightest and strongest. Rafiq
what is differents between GO and RGO?
what is simplest way to understand the applications of nano robots used to detect the cancer affected cell of human body.? How this robot is carried to required site of body cell.? what will be the carrier material and how can be detected that correct delivery of drug is done Rafiq
analytical skills graphene is prepared to kill any type viruses .
Any one who tell me about Preparation and application of Nanomaterial for drug Delivery
what is Nano technology ?
Bob Reply
write examples of Nano molecule?
The nanotechnology is as new science, to scale nanometric
nanotechnology is the study, desing, synthesis, manipulation and application of materials and functional systems through control of matter at nanoscale
Difference between extinct and extici spicies
Amanpreet Reply
Got questions? Join the online conversation and get instant answers!
Jobilize.com Reply

Get Jobilize Job Search Mobile App in your pocket Now!

Get it on Google Play Download on the App Store Now

Source:  OpenStax, Basic vector space methods in signal and systems theory. OpenStax CNX. Dec 19, 2012 Download for free at http://cnx.org/content/col10636/1.5
Google Play and the Google Play logo are trademarks of Google Inc.

Notification Switch

Would you like to follow the 'Basic vector space methods in signal and systems theory' conversation and receive update notifications?