Paper 3, Section II, B

Numerical Analysis
Part II, 2016

(a) Define the Jacobi and Gauss-Seidel iteration schemes for solving a linear system of the form Au=bA \mathbf{u}=\mathbf{b}, where u,bRM\mathbf{u}, \mathbf{b} \in \mathbb{R}^{M} and ARM×MA \in \mathbb{R}^{M \times M}, giving formulae for the corresponding iteration matrices HJH_{J} and HGSH_{G S} in terms of the usual decomposition A=L0+D+U0A=L_{0}+D+U_{0}.

Show that both iteration schemes converge when AA results from discretization of Poisson's equation on a square with the five-point formula, that is when

A=[SIISIISIIS],S=[4114114114]Rm×mA=\left[\begin{array}{rrrrr} S & I & & & \\ I & S & I & & \\ & \ddots & \ddots & \ddots & \\ & & I & S & I \\ & & & I & S \end{array}\right], \quad S=\left[\begin{array}{rrrrr} -4 & 1 & & & \\ 1 & -4 & 1 & & \\ & \ddots & \ddots & \ddots & \\ & & 1 & -4 & 1 \\ & & & 1 & -4 \end{array}\right] \in \mathbb{R}^{m \times m}

and M=m2M=m^{2}. [You may state the Householder-John theorem without proof.]

(b) For the matrix AA given in ()(*) :

(i) Calculate the eigenvalues of HJH_{J} and deduce its spectral radius ρ(HJ)\rho\left(H_{J}\right).

(ii) Show that each eigenvector q\mathbf{q} of HGSH_{G S} is related to an eigenvector p\mathbf{p} of HJH_{J} by a transformation of the form qi,j=αi+jpi,jq_{i, j}=\alpha^{i+j} p_{i, j} for a suitable value of α\alpha.

(iii) Deduce that ρ(HGS)=ρ2(HJ)\rho\left(H_{G S}\right)=\rho^{2}\left(H_{J}\right). What is the significance of this result for the two iteration schemes?

[Hint: You may assume that the eigenvalues of the matrix AA in ()(*) are

λk,=4(sin2x2+sin2y2), where x=kπhm+1,y=πhm+1,k,=1,,m,\lambda_{k, \ell}=-4\left(\sin ^{2} \frac{x}{2}+\sin ^{2} \frac{y}{2}\right), \quad \text { where } x=\frac{k \pi h}{m+1}, y=\frac{\ell \pi h}{m+1}, \quad k, \ell=1, \ldots, m,

with corresponding eigenvectors v=(vi,j),vi,j=sinixsinjy,i,j=1,,m.]\left.\mathbf{v}=\left(v_{i, j}\right), \quad v_{i, j}=\sin i x \sin j y, \quad i, j=1, \ldots, m . \quad\right]