Review : Matrices & Vectors — Question 8

PDF ↗

Question 8

Work with real matrices and column vectors. Write InI_n for the n×nn\times n identity, ATA^T for transpose, and ∥v∥=vTv\|v\|=\sqrt{v^Tv} for Euclidean length. Show the reasoning behind every classification; do not use eigenvalue methods.

For n=(1,2)Tn=(1,2)^T, define H=I2−2nnTnTn.H=I_2-\frac{2nn^T}{n^Tn}. Consider its action on the triangle with ordered vertices O=(0,0)O=(0,0), a=(3,1)a=(3,1) and b=(1,2)b=(1,2).

Tasks

  1. Compute HH. Prove from a perpendicular decomposition that it reflects the plane across a named line.

  2. Verify HTH=I2H^TH=I_2 and H2=I2H^2=I_2. Deduce preservation of lengths and dot products and describe the inverse operation.

  3. Find the reflected triangle, its area and orientation relative to the original. Check the midpoint and displacement of aa and its image.

  4. Find every vector fixed by HH and every vector sent to its negative. Derive these directions directly from the reflection formula.

Original worksheet page 1: question and worked solution for 5-2-008
Show solutionHide solution

Question 8 – Solution

Strategy. Separate a vector into components parallel and perpendicular to the normal. The formula changes the sign of only the normal component.

Step 1: Identify the reflection axis. Here nTn=5n^Tn=5, so H=15(3−4−4−3).\boxed{H=\frac 15\begin{pmatrix}3&-4\\-4&-3\end{pmatrix}.} Write x=x⟂+x∥x=x_\perp+x_\parallel, where x∥=n(nTx)/5x_\parallel=n(n^Tx)/5 and nTx⟂=0n^Tx_\perp=0. Then Hx=x⟂−x∥Hx=x_\perp-x_\parallel: it reflects across the line x1+2x2=0\boxed{x_1+2x_2=0}, whose direction is (2,−1)T(2,-1)^T.

Step 2: Verify the metric and inverse identities. The matrix is symmetric, and its numerator squared is (250025)\begin{pmatrix}25&0\\0&25\end{pmatrix}. Thus H2=HTH=I2H^2=H^TH=I_2. For any x,zx,z, (Hx)T(Hz)=xTHTHz=xTz(Hx)^T(Hz)=x^TH^THz=x^Tz, preserving dot products and hence lengths. Also H−1=HH^{-1}=H: reflecting twice restores the original vector.

Step 3: Transform and inspect the triangle. The images are HO=O,Ha=(1,−3)T,Hb=(−1,−2)T.HO=O,\qquad Ha=(1,-3)^T,\qquad Hb=(-1,-2)^T. The original ordered edge determinant is 3⋅2−1⋅1=53\cdot 2-1\cdot 1=5, so its area is 5/25/2. The reflected edge determinant is 1(−2)−(−1)(−3)=−51(-2)-(-1)(-3)=-5. Its area remains 5/25/2 but its orientation reverses, consistent with det⁡H=−1\det H=-1. The midpoint of aa and HaHa is (2,−1)(2,-1) on the axis; the displacement Ha−a=(−2,−4)=−2nHa-a=(-2,-4)=-2n is perpendicular to that axis.

Step 4: Classify fixed and reversed vectors. From the formula, Hx=xHx=x exactly when nTx=0n^Tx=0, giving the axis. For Hx=−xHx=-x, the decomposition gives x⟂=0x_\perp=0, so xx is a multiple of nn. The zero vector belongs to both sets. These conclusions follow from the components directly, without a characteristic-polynomial calculation. The plot uses equal axis scales.

See the diagram in the original worksheet below.

Original worksheet page 2: question and worked solution for 5-2-008

Original worksheet layout. Use Enlarge or open the PDF for a closer view.