Problem Statement:

R π/2
Discretize the equation: f (x) = cos x + 0
sin(x + t)f (t) dt.

Step 1: Understand the Equation

The given equation is a Fredholm integral equation of the second kind, where f (x) is
defined implicitly in terms of itself. The goal is to discretize this equation, which typically
involves approximating the integral using a numerical method (like the trapezoidal rule) and
evaluating the equation at discrete points.

Step 2: Choose Discretization Points

For numerical discretization, we select N + 1 points in the interval [0, π/2].
Let these points be equally spaced:

xi = i · h, i = 0, 1, . . . , N,
π
where h = 2N
is the step size.

1
 Step 3: Approximate the Integral Using the Trapezoidal Rule
R π/2
The integral 0
sin(x + t)f (t) dt can be approximated using the trapezoidal rule:
Z π/2 " N −1
#
sin(x + x0 )f (x0 ) + sin(x + xN )f (xN ) X
sin(x + t)f (t) dt ≈ h + sin(x + xj )f (xj ) .
0 2 j=1

Step 4: Discretize the Equation

Evaluate the original equation at each discretization point xi :
" N −1
#
sin(xi + x0 )f (x0 ) + sin(xi + xN )f (xN ) X
f (xi ) = cos(xi ) + h + sin(xi + xj )f (xj ) .
2 j=1

This gives a system of N + 1 equations for the N + 1 unknowns f (x0 ), f (x1 ), . . . , f (xN ).

Step 5: Write the System in Matrix Form

2

Let f = [f (x0 ), f (x1 ), . . . , f (xN )]T . The discretized system can be written as:
f = c + h (Mf ) ,
where:
ˆ c is the vector with entries ci = cos(xi ).
ˆ M is a matrix with entries: (
sin(xi +xj )
2
if j = 0 or j = N,
Mij =
sin(xi + xj ) otherwise.

Step 6: Solve the Linear System

Rearrange the equation to solve for f :
(I − hM)f = c.
This is a linear system that can be solved using numerical methods (e.g., Gaussian elimination or iterative solvers).
 Final Answer:
R π/2
The discretized form of the equation f (x) = cos x + 0
sin(x + t)f (t) dt using the trapezoidal rule is:
" N −1
#
sin(xi + x0 )f (x0 ) + sin(xi + xN )f (xN ) X
f (xi ) = cos(xi ) + h + sin(xi + xj )f (xj ) ,
2 j=1

for i = 0, 1, . . . , N . This results in a linear system (I − hM)f = c, which can be solved numerically for the vector f .

Step-by-Step Construction of the System of Equations for i = 0, 1, 2, . . . , N

We start with the discretized equation at each point xi :
" N −1
#
sin(xi + x0 )f (x0 ) + sin(xi + xN )f (xN ) X
f (xi ) = cos(xi ) + h + sin(xi + xj )f (xj ) .
2 j=1
3

Let’s write this explicitly for each i:

For i = 0:
" N −1
#
sin(x0 + x0 )f (x0 ) + sin(x0 + xN )f (xN ) X
f (x0 ) = cos(x0 ) + h + sin(x0 + xj )f (xj ) .
2 j=1

Simplify: " −1
N
#
sin(2x0 )f (x0 ) + sin(x0 + xN )f (xN ) X
f (x0 ) = cos(x0 ) + h + sin(x0 + xj )f (xj ) .
2 j=1

For i = 1:
" N −1
#
sin(x1 + x0 )f (x0 ) + sin(x1 + xN )f (xN ) X
f (x1 ) = cos(x1 ) + h + sin(x1 + xj )f (xj ) .
2 j=1
 For i = 2:
" N −1
#
sin(x2 + x0 )f (x0 ) + sin(x2 + xN )f (xN ) X
f (x2 ) = cos(x2 ) + h + sin(x2 + xj )f (xj ) .
2 j=1

..
.
For i = N − 1:
" N −1
#
sin(xN −1 + x0 )f (x0 ) + sin(xN −1 + xN )f (xN ) X
f (xN −1 ) = cos(xN −1 ) + h + sin(xN −1 + xj )f (xj ) .
2 j=1

For i = N :
" N −1
#
sin(xN + x0 )f (x0 ) + sin(xN + xN )f (xN ) X
f (xN ) = cos(xN ) + h + sin(xN + xj )f (xj ) .
2 j=1
4

Simplify: " N −1
#
sin(xN + x0 )f (x0 ) + sin(2xN )f (xN ) X
f (xN ) = cos(xN ) + h + sin(xN + xj )f (xj ) .
2 j=1

Matrix Representation
Let f = [f (x0 ), f (x1 ), . . . , f (xN )]T , and c = [cos(x0 ), cos(x1 ), . . . , cos(xN )]T . The system can be written as:

f = c + hMf ,

where M is an (N + 1) × (N + 1) matrix with entries:

(
sin(xi +xj )
2
if j = 0 or j = N,
Mij =
sin(xi + xj ) otherwise.
 Final Linear System
Rearrange to solve for f :
(I − hM)f = c,
where I is the identity matrix. This system can be solved numerically for f .

Summary
The system of equations for i = 0, 1, . . . , N is:
" N −1
#
sin(xi + x0 )f (x0 ) + sin(xi + xN )f (xN ) X
f (xi ) = cos(xi ) + h + sin(xi + xj )f (xj ) .
2 j=1

This forms a linear system (I − hM)f = c, where M is defined as above.

5
 To solve the given equation using the Leibniz rule for differentiation, follow these steps:

Given Equation:
Z x
2
f (x) = x + ex−t f (t) dt
0

Step 1: Differentiate Both Sides with Respect to x

Differentiate the entire equation with respect to x:
Z x 
′ d d
x2 + x−t


f (x) = e f (t) dt
dx dx 0

Step 2: Compute the Derivative of x2

The derivative of x2 is straightforward:
d
6

x2 = 2x


dx

Step 3: Apply the Leibniz Rule to the Integral

The integral term is of the form:
Z b(x)
g(x, t) dt
a(x)

where a(x) = 0, b(x) = x, and g(x, t) = ex−t f (t).

The Leibniz rule states: Z b(x) ! Z b(x)
d ′ ′ ∂
g(x, t) dt = g(x, b(x)) · b (x) − g(x, a(x)) · a (x) + g(x, t) dt
dx a(x) a(x) ∂x

Apply this to our integral:

1. First Term: g(x, x) · b′ (x) = ex−x f (x) · 1 = f (x)

2. Second Term: g(x, 0) · a′ (x) = ex−0 f (0) · 0 = 0 (since a′ (x) = 0)

 3. Integral Term:
∂
ex−t f (t) = ex−t f (t)


∂x
So, the integral becomes: Z x
ex−t f (t) dt
0

Combining these, the derivative of the integral is: Z x

f (x) + ex−t f (t) dt
0

Step 4: Combine All Terms

Now, substitute the derivatives back into the original equation:
Z x
′
f (x) = 2x + f (x) + ex−t f (t) dt
0
7

But from the original equation, we know: Z x

ex−t f (t) dt = f (x) − x2
0
Substitute this into the derivative equation:
f ′ (x) = 2x + f (x) + (f (x) − x2 )
f ′ (x) = 2x + 2f (x) − x2

Step 5: Rewrite the Differential Equation

Rearrange the terms to form a first-order linear ordinary differential equation (ODE):

f ′ (x) − 2f (x) = 2x − x2
 Step 6: Solve the ODE
This is a linear ODE of the form:
f ′ (x) + P (x)f (x) = Q(x)
where P (x) = −2 and Q(x) = 2x − x2 .
The integrating factor µ(x) is: R
µ(x) = e P (x) dx
= e−2x
Multiply both sides of the ODE by µ(x):
e−2x f ′ (x) − 2e−2x f (x) = (2x − x2 )e−2x
The left side is the derivative of e−2x f (x):
d −2x
e f (x) = (2x − x2 )e−2x


dx
Integrate both sides with respect to x: Z
−2x
e f (x) = (2x − x2 )e−2x dx
8

Step 7: Compute the Integral

We need to compute: Z
(2x − x2 )e−2x dx

Use integration by parts: Let u = 2x − x2 , dv = e−2x dx, so du = (2 − 2x)dx, v = − 21 e−2x .

Apply integration by parts: Z Z
u dv = uv − v du
Z Z
2 −2x 1 2 −2x 1
(2x − x )e dx = − (2x − x )e − − e−2x (2 − 2x) dx
2 2
Z
1 1
= − (2x − x2 )e−2x + (2 − 2x)e−2x dx
2 2
Now, compute the remaining integral: Z
(2 − 2x)e−2x dx
 Again, use integration by parts: Let u = 2 − 2x, dv = e−2x dx, so du = −2dx, v = − 21 e−2x .
Z Z
−2x 1 −2x 1
(2 − 2x)e dx = − (2 − 2x)e − − e−2x (−2) dx
2 2
1 1
= − (2 − 2x)e−2x − e−2x
2 2
1
= − (2 − 2x + 1)e−2x
2
1
= − (3 − 2x)e−2x
2
Substitute back: Z  
2 −2x 1 2 −2x 1 1 −2x
(2x − x )e dx = − (2x − x )e + − (3 − 2x)e
2 2 2
1 1
= − (2x − x2 )e−2x − (3 − 2x)e−2x
2 4
 
1 1
9

−2x 2
=e − (2x − x ) − (3 − 2x)
2 4
x2 3 x
 
−2x
=e −x + − +
2 4 2
 2 
−2x x x 3
=e − −
2 2 4

Step 8: Solve for f (x)

Now, substitute the integral back:
x2 x 3
 
−2x −2x
e f (x) = e − − +C
2 2 4
Divide both sides by e−2x :
x2 x 3
f (x) = − − + Ce2x
2 2 4
 Step 9: Determine the Constant C
Use the original equation to find C. Evaluate at x = 0:
Z 0
2
f (0) = 0 + e0−t f (t) dt = 0
0

Substitute x = 0 into the general solution:

0 0 3 3
f (0) = − − + Ce0 = − + C = 0
2 2 4 4
3
C=
4

Final Solution:
x2 x 3 3 2x
f (x) = − − + e
2 2 4 4
10

Verification:
Differentiate f (x) and substitute back into the ODE to ensure consistency.

Answer:
3 x2 x 3
f (x) = e2x + − −
4 2 2 4