Beruflich Dokumente
Kultur Dokumente
of
VOLUME 1
MATHEMATICS
Solutions Guide
by Eric Lengyel
Chapter 1
1. (a) i j 0, j k 0, and k i 0.
(b) i j k , j i k , j k i, k j i , k i j, and i k j.
(c) i, j, k 1.
a 2 b 2 2a b a 2 b 2 2 a b a b .
2 2
ab
4. Expanding each term with the vector triple product identity, we have
a b c b c a c a b
b a c c a b c b a a b c a c b b c a ,
b y2 b z2 b x b y b x b z
1
6. R 2 b x b y b b
2
x
2
z b y b z .
b
b x b z b y b z b x2 b y2
7. Expanding v ab , we have
v a b
v a b v a b .
a b2
Since a b a 0, the entire last term disappears when we calculate
v ab a , and we are left with
Answers to Exercises 3
v a
v a b a a v a.
a2
AB C ij AB ik Ckj
k 0
p 1
m 1
Ail Blk C kj
k 0 l 0
m 1
p 1
Ail Blk C kj
l 0 k 0
m 1
Ail BC lj
l 0
A BC ij .
9. For a 2 2 matrix, its clear that the determinant is just the product of the two
diagonal entries. To prove the general case by induction, assume that for any
k n , the determinant of a k k matrix is equal to the product of its diagonal
entries. If we calculate the determinant of an n n matrix using expansion by
minors along the last column, then the only nonzero term corresponds to the
n 1, n 1 entry. This is multiplied by the determinant of the upper-left
n 1 n 1 portion of the matrix, which is known to be the product of its
diagonal entries. Since we are now simply multiplying by one more diagonal
entry, we have proven the general case.
10. In the product tA , every entry of A is multiplied by t. This means that every
factor in the Leibniz formula for the determinant is multiplied by t when A is
replaced by tA . Since every term has n factors, the full determinant is multi-
plied by a factor of t n .
a b
11. A matrix is singular when the determinant ad bc is zero, so we need
c d
to nd all cases for which ad bc with a, b, c, and d being either 0 or 1. Of
the 16 possible 2 2 matrices, the following 10 are singular.
4 Answers to Exercises
0 0 1 0 0 1 0 0 0 0
0 0 0 0 0 0 1 0 0 1
1 1 0 1 0 0 1 0 1 1
0 0 0 1 1 1 1 0 1 1
12. Multiply both sides of the equation LM I on the right by R. Then LMR R ,
but MR I , so L R.
13. The following matrix is the inverse of the elementary matrix E that multiplies
row r by t.
1 0
0
E 1 0 1 t 0 row r
0 0 1
The following matrix is the inverse of the elementary matrix E that exchanges
rows r and s. (In this case, E is its own inverse.)
1 0 0 0
0 0 1 0 row r
E
1
0 1 0 0 row s
0 0 0 1
The following matrix is the inverse of the elementary matrix E that adds row
s multiplied by the scalar value t to row r.
Answers to Exercises 5
column s
1 0 0 0
0 1 t 0 row r
E
1
0 0 1 0 row s
0 0 0 1
15. Suppose that M is an n n matrix and that row r of M is all zeros. Let N be
any n n matrix. The r , r entry of the product MN is given by
n 1
MN rr M rk N kr ,
k 0
but each M rk 0, so its impossible to produce anything other than zero. Since
a one is needed in the r , r entry of the identity matrix, M cannot be invertible.
adj M M ij C ki M M kj .
k 0
6 Answers to Exercises
In 0 I n a b a I n a b a
T b T b T a b b T a 1 .
b 1 0 1
Now multiplying by the third matrix produces
In a b a In 0 I n a
T T ,
b b a b b a 1 b 1 0 1 a b
T T
b v yt a b t x
M 1M 00 s v t u
v a xt b a t y
M 1M 11 s v t u
d u ws c d s z
M 1M 22 s v t u
u c zs d c s w
M 1M 33 s v t u
A little regrouping in the numerators gives us
Answers to Exercises 7
b, v, a t ya xb
M 1M 00 s v t u
v, a, b t ya xb
M 1M 11 s v t u
d , u , c s wc z d
M 1M 22 s v t u
u , c , d s wc z d
M 1M 33 s v t u
.
The term v a a is zero, and the remaining two terms in the numerator can-
cel, producing a zero for the 1, 0 entry.
a y b z a z b y c x a z bx a x b z c y a xb y a y b x c z d x ,
and the x component of the right side is
a x d x a y d y a z d z b y c z bz c y bx d x b y d y bz d z c y a z c z a y
c x d x c y d y c z d z a y bz a zb y .
Collecting the terms on the right side containing the factor d x gives us
a x b y c z bz c y bx c y a z c z a y c x a ybz a z b y d x,
and each of these terms matches a term from the left side. We just need to show
that the remaining terms on the right side sum to zero. Those terms are
a y d y a z d z b y c z bz c y b y d y b z d z c y a z c z a y
c y d y c z d z a ybz a zb y ,
8 Answers to Exercises
Chapter 2
1. A matrix M is an involution if M 2 I. The determinant of M must therefore
square to one, which leaves only two possibilities, 1 or 1.
AB 1 B 1A 1 B T A T AB T .
Therefore, AB is orthogonal.
1 a 2 0 0
4. N 0 1 b 2
0 .
0 0 1 c 2
1 s a x2 s 1 s a x a y 1 s a x a z
5. 1 s a x a y 1 s a y s 1 s a y a z .
2
1 s a x a z 1 s a y a z 1 s a z2 s
1 0 tan
6. (a) M skew , i, k 0 1 0 .
0 0 1
Answers to Exercises 9
1 0 0
(b) M skew , j, k 0 1 tan .
0 0 1
1 0 0
(c) M skew , j, i tan 1 0 .
0 0 1
M M
8. .
0 1
v 1 v 2 s 1 v 2 s 2 v 1 s1 s 2 v 1 v 2 ,
2 2 2
q 1q 2
After applying Lagranges identity to the cross product and dot product, this
becomes
Taking square roots of both sides shows that the magnitude of the product is
equal to the product of the magnitudes.
q q q q
10. f q and g q .
2 2
We know that a 2 1, so we are left with the sum of a squared sine and squared
cosine of the same angle, which is always one.
14. Its required that c cos 2 . Applying the relationship cos v 1 v 2 and
the trigonometric identity cos 2 2 1 cos 2, we nd that
1 v1 v 2
c .
2
We must also have sa sin 2 v 1 v 2 sin . This can be simplied using
the identity sin 2 sin 2 cos 2 , and the quaternion can be written as
q 1 2c v 1 v 2 c.
Chapter 3
1. 3
derivatives are
f
2 1 v 1 2 v 1 t1v12 t 2 v 1 v 2
t1
f
2 2 v 2 1 v 2 t 2 v 22 t1 v 1 v 2 .
t 2
Simultaneously setting these to zero (and dropping the factors of two) gives us
the system
v12 v 1 v 2 t1 2 1 v 1
.
v1 v 2 v 22 t 2 1 2 v 2
Answers to Exercises 11
When we negate the bottom row and solve for t1 and t 2, the result is equivalent
to Equation (3.24).
vd n 1 2 d 2n 1 v 2 ,
5. n n and d d n t.
6. v v and m m t v .
8. n v 1 v 2, and either d v 1 m 2 or d v 2 m 1.
9. n v 2 m 2 m 1 and d v, m 1 , m 2 .
v m 2 m1
10. d .
v2
11. v v t 2 0 v 0 v v v t
0 0 0 0 0 .
p v 1 v 2 v 1 m 2 v 2 m 1 v12 v 2 m 2 v 22 v 1 m 1
m 2 v1 v1 v 2
12 Answers to Exercises
Chapter 4
1. The area of region A is a x b x b y a y , the area of region B is
2 x y
1
b b a y , and the area of region C is 12 a y a x b x . The total area of the
parallelogram spanned by a and b is A 2 B 2C , which simplies to the quan-
tity a x b y a y b x .
3. Suppose that A and B are both n 1 -blades. The set of basis vectors compos-
ing A and B can dier by at most one element because it would otherwise
require that we have more than n basis vectors. Therefore, we can write
A C v 1 and B C v 2, where C is an n 2 -blade representing the com-
mon factor between A and B, and v 1 and v 2 are vectors. The sum A B is then
C v 1 v 2 , and this is an n 1 -blade because v 1 v 2 is just a vector. Since
any n 1 -vector can be decomposed into a sum of n 1 -blades, this proves
that all n 1 -vectors must be n 1 -blades.
4. First assume that A is a 2-blade. Then A a b for some vectors a and b, and
clearly, A A a b a b 0.
Now assume that A A 0, and suppose that A is not a 2-blade. Then A can
be written as A a b B , where a and b are vectors, and B is a 2-vector such
that a b B 0. That is, the basis elements of a b are independent of the
basis elements of B, and the fact that the dimensionality is at least four guar-
antees that enough independent basis vectors are available. This means that
A A contains a nonzero term, which is a contradiction, so it must be the case
that A is a 2-blade.
A B 1
gr A gr B n gr A gr B
AB
and
gr A n gr A gr B n gr B
A B 1 A 1 B.
Answers to Exercises 13
A B C A B C A B C.
8. The dot product between L vx , Lvy , Lvz and L mx , L my , Lmz is
a wb x a x b w a y b z a z b y a w b y a y b w a z b x a x b z
a wb z a z b w a x b y a y b x .
When multiplied out, every term cancels, so the dot product is zero, and the
two vectors are therefore orthogonal.
q x p x e 41 q y p y e 42 q z p z e 43
p y q z p z q y e 23 p z q x p x q z e 31 p x q y p y q x e 12
q y rz p y rz p z ry q z ry p y q z p z q y e 234
p x rz q x rz q z rx p z rx p z q x p x q z e 314
q x ry p x ry p y rx q y rx p x q y p y q x e 124
p y q z rx p z q y rx p z q x ry p x q z ry p x q y rz p y q x rz e 123
p q q r r p e 4 p q r e4
1
11.
ab cdf g h
a b c f g h .
k k 1
13. f k .
2
14. Let p be the point closest to the origin on the line L. Then we can write m as
m p p v p v because p and p v are two points on the line. We
know that p and v are perpendicular, so p v 0, and we can write m pv using
the geometric product. Division by v gives us p m v.
16. The even subalgebra of the three-dimensional geometric algebra has the basis
elements 1, e 23 , e 31 , e 12 . The two-dimensional Cliord algebra has the
basis elements 1, e 1 , e 2 , e 12 . In the 3D case, let x e 23, y e 31, and
z e12. In the 2D case, let x e1, y e 2 , and z e 12 . Then in both cases, we
have x 2 y 2 z 2 1, xy z , yz x, and zx y. This is sucient to establish
an isomorphism.