Prove that all Toeplitz tridiagonal M × M M \times M M × M matrices A A A of the form
A = [ a b − b a b ⋱ ⋱ ⋱ − b a b − b a ] A=\left[\begin{array}{rrrrr} a & b & & & \\ -b & a & b & & \\ & \ddots & \ddots & \ddots & \\ & & -b & a & b \\ & & & -b & a \end{array}\right] A = ⎣ ⎢ ⎢ ⎢ ⎢ ⎢ ⎡ a − b b a ⋱ b ⋱ − b ⋱ a − b b a ⎦ ⎥ ⎥ ⎥ ⎥ ⎥ ⎤
share the same eigenvectors ( v ( k ) ) k = 1 M \left(\boldsymbol{v}^{(k)}\right)_{k=1}^{M} ( v ( k ) ) k = 1 M , with the components v m ( k ) = i m sin k m π M + 1 , m = \boldsymbol{v}_{m}^{(k)}=i^{m} \sin \frac{k m \pi}{M+1}, m= v m ( k ) = i m sin M + 1 k m π , m = 1 , … , M 1, \ldots, M 1 , … , M , where i = − 1 i=\sqrt{-1} i = − 1 , and find their eigenvalues.
The advection equation
∂ u ∂ t = ∂ u ∂ x , 0 ⩽ x ⩽ 1 , 0 ⩽ t ⩽ T , \frac{\partial u}{\partial t}=\frac{\partial u}{\partial x}, \quad 0 \leqslant x \leqslant 1, \quad 0 \leqslant t \leqslant T, ∂ t ∂ u = ∂ x ∂ u , 0 ⩽ x ⩽ 1 , 0 ⩽ t ⩽ T ,
is approximated by the Crank-Nicolson scheme
u m n + 1 − u m n = 1 4 μ ( u m + 1 n + 1 − u m − 1 n + 1 ) + 1 4 μ ( u m + 1 n − u m − 1 n ) u_{m}^{n+1}-u_{m}^{n}=\frac{1}{4} \mu\left(u_{m+1}^{n+1}-u_{m-1}^{n+1}\right)+\frac{1}{4} \mu\left(u_{m+1}^{n}-u_{m-1}^{n}\right) u m n + 1 − u m n = 4 1 μ ( u m + 1 n + 1 − u m − 1 n + 1 ) + 4 1 μ ( u m + 1 n − u m − 1 n )
where μ = Δ t ( Δ x ) 2 , Δ x = 1 M + 1 \mu=\frac{\Delta t}{(\Delta x)^{2}}, \Delta x=\frac{1}{M+1} μ = ( Δ x ) 2 Δ t , Δ x = M + 1 1 , and u m n u_{m}^{n} u m n is an approximation to u ( m Δ x , n Δ t ) u(m \Delta x, n \Delta t) u ( m Δ x , n Δ t ) . Assuming that u ( 0 , t ) = u ( 1 , t ) = 0 u(0, t)=u(1, t)=0 u ( 0 , t ) = u ( 1 , t ) = 0 , show that the above scheme can be written in the form
B u n + 1 = C u n , 0 ⩽ n ⩽ T / Δ t − 1 B \boldsymbol{u}^{n+1}=C \boldsymbol{u}^{n}, \quad 0 \leqslant n \leqslant T / \Delta t-1 B u n + 1 = C u n , 0 ⩽ n ⩽ T / Δ t − 1
where u n = [ u 1 n , … , u M n ] T \boldsymbol{u}^{n}=\left[u_{1}^{n}, \ldots, u_{M}^{n}\right]^{T} u n = [ u 1 n , … , u M n ] T and the real matrices B B B and C C C should be found. Using matrix analysis, find the range of μ \mu μ for which the scheme is stable. [Fourier analysis is not acceptable.]