Solutions of The Second Examination

Engineering Mathematics (II)

Examination Date : November 30, 2020, Time: 13:10~16:00
Problem 1 : Simply answer the following questions(簡答題): (20pts)
1. What is the “Wronskian” of n solutions, y (i ) , i = 1,..., n , of a system of ODEs given by
y' = Ay, y(0) = y0 ∈ R n , t ≥ 0 where the dimension of the constant matrix A is n by n.

Ans:
y1(1) y1( 2)  y1( n)
y2(1) y2( 2)  y2( n)
W ( y , , y ) =
(1) ( n)


yn(1) yn( 2)  yn( n)
2. Consider a system of ODEs given by y' = Ay + g , y (0) = y0 ∈ R n , t ≥ 0 where the dimension of the
constant matrix A is n by n . Give your two methods to find a solution y p
Ans: the method of undetermined coefficients, and the method of variation of parameters, or the
matrix exponential method.

3. Consider a system of ODEs given by y' = Ay, y(0) = y0 ∈ R n , t ≥ 0 where the dimension of the
constant matrix A is n by n . Show that the solution of the system is given y(t ) = ceλt x where λ is one
of eigenvalues, x is the eigenvector corresponding to the eigenvalue λ , and c is a constant.
Ans:
dceλt x
y '(t ) = = λ ceλt x = ceλt (λ x ) = ceλt Ax = A( ceλt x ) = Ay.
dt
This proves that ceλt x is a solution vector of y ' = Ay.

4. Find the particular solution y p of a system of nonhomogeneous linear ODEs given by y' = Ay + g
where the dimension of the constant matrix A is n by n, and g is a continuous time vector , and its
homogeneous solution is given by yh = Y (t )c and c is a constant vector .
Ans:
Let y p = Y (t )u where Y (t ) ' = AY (t ). Then y '(t ) = Y (t ) ' u + Y (t )u ' = AY (t )u + Y (t )u ' = AY (t )u + g
⇒ Y (t )u ' = g ⇒ u ' = Y (t ) −1 g ⇒ u = ∫ Y (t ) −1 gdt.

5. For a system of linear ODEs, y' = Ay, y(0) = y0 ∈ R n , t ≥ 0 , give your two methods to compute the
matrix exponential function, e At .
Ans: eigenvector-eigenvalue decomposition, and the Cayley-Hamilton theorem

∞
3
6. Find the radis of the convergence for the series ∑ ( 4)
m=0
m
x2m .

Ans:
 (3 / 4) m +1 ( x )
2m+2
3 2
By ratio test, lim = x <1
( x)
2m
m →∞
(3 / 4) m
4

2 4 4
⇒ x < ⇒ x< .
3 3
4
Hence, the radius of convergence this problem is R =
3
7. Find a power series solution of the ODE described by y ''+ p ( x) y '+ q ( x) y = 0 where p(x) and q(x) are
analytic at x=x0.
∞
Ans: y( x) = ∑ am ( x − x0 )m
m=0

b( x ) c( x)
8. Find a power series solution of the ODE described by y ''+ y '+ 2 y = 0 where b(x) and c(x) are
x x
analytic at x=0.
∞
Ans: y( x) = ∑ am x m+ r
m=0

∞
9. The gamma function is defined by Γ(v + 1) = ∫ e − t t v dt , v ∈ R ≥ 0 .Show that Γ(v + 1) = vΓ(v) .
0
Ans:
∞ ∞ ∞
Γ(v + 1) = ∫ e − t t v dt = ∫ t v d (−e − t ) = −e − t t v |∞0 − ∫ −e − t vt v −1dt
0 0 0
∞
= 0 + v ∫ e − t t v −1dt = vΓ(v)
0

10. What is a general solution of Bessel’s equation, x 2 y ''+ xy '+ ( x 2 −ν 2 ) y = 0 , for all values of ν and x >
0?
Ans: y(x) = C1Jν(x) + C2Yν(x)..
Problem 2: (20ts)
(a) Consider the RLC circuit in Fig. 1 whose simultaneous ordinary differential equations for the two loop
currents are given by (1) I1 = −4 I1 + 4 I 2 + 12
'
. Use Gaussian Elimination method to solve the
(2) I 2 ' = −1.6 I1 + 1.2 I 2 + 4.8
initial value problem of the circuit where I1 ( 0 ) = I 2 ( 0 ) = 0 . (10pts)
(b) Consider the RL circuit in Fig. 2 whose simultaneous ordinary differential equations for the two lopp
currents are given by  I1 '  =  −3 1.25  I1  + 125 . Use the method of undertermined coefficients
 I ' 1 −1   I 2   0 
 2    
and the vector-matrix technique to solve the initial value problem of the circuit where
I1 ( 0 ) = I 2 ( 0 ) = 0 . (10pts)

Fig.1. RLC circuit in Problem 4(a). Fig.2. RL circuit in Problem 4(b).

Solutions:
(a)
The system of ODEs for the circuit network in Problem 9 is described by
(a) I1' = −4 I1 + 4 I 2 + 12
(b) I 2 ' = −1.6 I1 + 1.2 I 2 + 4.8
which can be rewritten by using the differential operator
( D + 4) I1 − 4 I 2 = 12......................(1)
1.6 I1 + ( D − 1.2) I 2 = 4.8...............(2)
(1) × ( D − 1.2) + (2) × 4 gives (( D + 4)( D − 1.2) + 6.4) I1 = 4.8
which leads to the the characteristic equation as
λ 2 +2.8λ + 1.6 = (λ + 2)(λ + 0.8) = 0.
Solving the equation yields λ = −0.8, λ = −2 which are two distinct real roots.
Thus, the homogeneous solution of y1 is I1h = c1e −0.8t + c2 e −2t .
Moreover, the particular solustion of y1 is I1 p = A, and
1.6A = 4.8 ⇒ y1 p = A = 3.
Thus, the general solution of I1 is I1 = I1h + I1 p = c1e −0.8t + c2 e −2t + 3.
Furthermore, (2) × (D+4)-(1) ×1.6 gives (D 2 + 2.8 D + 1.6 I ) I 2 = 4.8 × 4 − 12 ×1.6 = 0
Similiarly, we have that the same general equation of I 2 is c3e −0.8t + c4 e −2t .
In the last step, we substitute I1 and I 2 into (1) and then obtain
(-0.8c1 + 4c1 − 4c3 )e −0.8t + (−2c2 + 4c2 − 4c4 )e −2t = 0
which leads to that 3.2c1 − 4c3 = 0, and 2c2 − 4c4 = 0, giving c3 = 0.8c1 and c4 = 0.5c2 .
Hence, the general equation of I1 is c1e −0.8t + c2 e −2t + 3..
and the the general equation of I 2 is 0.8c1e −0.8t + 0.5c2 e −2t .
or, the general solution of the homogeneous ODE system is
 I (t )  1   1  3
I (t ) =  1  = c1   e −0.8t + c2 e −2t   +   , c1 , c2 ∈ R.
 I 2 (t )  0.8 0.5 0 
  I (0)  1   1  3 0 
From the initial condition, we have I (0) =  1  = c1   + c2   +   =  
 I 2 (0)  0.8 0.5 0  0 
which gives c1 = 5, and c2 = −8.
The particular solution of the circuit network is
 I (t )  1   1   3 5e −0.8t − 8e −2t + 3
I (t ) =  1  = 5   e −0.8t − 8e −2t   +   =  −0.8t  ,t ≥ 0
 I 2 (t )  0.8 0.5 0   4e − 4e −2t 
3
and we find that I (t ) →   as t → ∞.
0 
(b)
The IVP problem of the circuit can be written in the following matrix-vector from
 I1 '   −3 1.25  I1  125
 I ' =  1 +
−1   I 2   0 
, I1 (0) = 0, I 2 (0) = 0.
 2  
 −3 1.25
From the system matrix  , we obtain the characteristic equation as
1 −1 
1 0   −3 1.25 λ + 3 −1.25
λ − 1  =  −1  = λ +4λ + 3 − 1.25 = λ +4λ + 1.75 = (λ + 3.5)(λ + 0.5) = 0.
2 2

 0 1   −1   λ + 1 
Solving the eigenvalues gives λ = −3.5, λ = −0.5.For λ = −3.5 , its corresponding eigenvector is
found by
1 0   −3 1.25 (1) 0   −0.5 −1.25 (1) 0  5
(−3.5   −  )x =   ⇒   x =   ⇒ x (1) =  
0 1   1 −1  0   −1 −2.5  0   −2 
Similarly, For λ = −0.5 , its corresponding eigenvector is
also found by
1 0   −3 1.25 (2) 0   2.5 −1.25 (2) 0  1 
(−0.5   −  )x =   ⇒   x =   ⇒ x (2) =   .
0 1   1 −1  0   −1 0.5  0 2
Hence, the homogeneous solution of the ODE system is
5 1 
I ( h ) = c1e −3.5t   + c2 e −0.5t  
 −2  2
Due to the nonhomogeneous term in the system, we use the method of undetermined coefficients
c 
to choose I ( p ) =  31 
c32 
which leads to
 I1( p ) '   −3 1.25  c31  125 0   −3c + 1.25c32  125
 ( p)  =     +   , ⇒   =  31 + 
 I 2 '  1 −1  c32   0  0   c31 − c32   0 
Equaling the coefficient of the aformentioned matrix equation obtains
 −3c31 + 1.25c32 + 125
 c − c  = 0.................................(1)
 31 32 
 c  500 / 7  500 / 7 
From (1), we solve that  31  =   .Finally,I ( p ) =  
c32  500 / 7  500 / 7 
The general solution of the ODE system is given by
5 1  500 / 7 
I = I ( h ) + I ( p ) = c1e −3.5t   + c2 e −0.5t   +  
 −2   2  500 / 7 
  I (0)  0 
Including the two initial conditions,  1  =   gives
 I 2 (0)  0 
5 1  500 / 7  0 
c1   + c2   +  = 
 −2   2  500 / 7  0 
 5c + c   −500 / 7   c1   −125 / 21  −125 / 21
which gives  1 2  =  ⇒ = = 
 −2c1 + 2c2   −500 / 7  c2   −875 / 21  −125 / 3 
125 −3.5t  5  125 −0.5t 1  500 1
∴ I = I (h) + I ( p) = − e  − e  + 
21  −2  3  2  7 1
Problem 3: (10pts)
Use the matrix-vector method and the method of undertermined coefficients to find a genral solution of
the nonhomogenous linear system of ODEs with the initial condition given by
 y '  1 4   y1   6t − t 2   y1 (0)   2 
y' = Ay + g =  1  =   y  +  ,
2   =  −1 , t ≥ 0 .
y '
 2  1 1   2  −1 + t − t y (0)
  2   
Solution:
The IVP problem of the ODE system can be written in the following matrix-vector from
 y1 '  1 4   y1   6t − t 2 
 y ' = 1 1   y  +  2
, y1 (0) = 2, y2 (0) = −1.
 2     2   −1 + t − t 
1 4 
From the system matrix   , we obtain the characteristic equation as
1 1 
1 0  1 4  λ − 1 −4 
λ  − 1 1  =  −1 λ − 1 = λ -2λ + 1 − 4 = λ -2λ − 3 = (λ − 3)(λ + 1) = 0.
2 2

 0 1     
Solving the eigenvalues gives λ = 3, λ = −1.For λ = 3 , its corresponding eigenvector is
found by
1 0  1 4  (1) 0   2 −4  (1) 0  2
(3   −  )x =   ⇒   x =   ⇒ x (1) =  
0 1  1 1  0   −1 2  0  1 
Similarly, For λ = −1 , its corresponding eigenvector is
also found by
1 0  1 4  (2) 0   −2 −4  (2) 0  2
(−1   −  )x =   ⇒   x =   ⇒ x (2) =   .
0 1  1 1  0   −1 −2  0   −1
Hence, the homogeneous solution of the ODE system is
2 2
y ( h ) = c1e3t   + c2 e − t  
1   −1
Due to the nonhomogeneous term in the system, we use the method of undetermined coefficients
to choose
c   c   c   c t 2 + c t + c51 
y ( p ) =  31  t 2 +  41  t +  51  =  31 2 41 
c32  c42  c52  c32t + c42t + c52 
which leads to
 y1( p ) '  1 4   y1( p )   6t − t 2 
 ( p)  =    ( p)  +  2
,
y
 2  ' 1 1  y
 2   −1 + t − t 
 c   c  c t + c t + c51 + 4c32t 2 + 4c42t + 4c52   6t − t 2 
2
⇒ 2t  31  +  41  =  31 2 41 + 2
c32  c42   c31t + c41t + c51 + c32t + c42t + c52   −1 + t − t 
2

 2c   c  c + 4c32 − 1 2 c41 + 4c42 + 6   c51 + 4c52 

⇒  31  t +  41  =  31  t +  c + c + 1  t + c + c − 1
 2c32  c42   c31 + c32 − 1   41 42   51 52 
Equaling the coefficients of the second-order matrix equation obtains
c31 + 4c32 − 1
 c + c − 1  = 0.................................(1)
 31 32 
c41 + 4c42 + 6   2c31 
 c + c + 1  =  2c  ........................(2)
 41 42   32 
 c41   c51 + 4c52 
c  = c + c − 1 ..............................(3)
 42   51 52 
  c  1 
From (1), we solve that  31  =   ,
c32  0 
c + 4c42   −4   c41   0 
From (2), it follows that  41 = ⇒ = 
 c41 + c42   −1 c42   −1
c + 4c52  0   c51  0 
From (3), we have  51 = ⇒ = 
 c51 + c52  0  c52  0 
1  2  0   0   t 2 
Finally,y =   t +   t +   =  
( p)

0   −1 0   −t 
The general solution of the ODE system is given by
2  2  t2 
y = y ( h ) + y ( p ) = c1e3t   + c2 e − t   +  
1   −1  −t 
 y (0)   2 
Including the two initial conditions,  1  =   gives
 y2 (0)   −1
2  2  0   2 
c1   + c2   +   =  
1   −1 0   −1
 2c + 2c2   2   c1  0 
which gives  1 = ⇒ = 
 c1 − c2   −1 c2  1 
 2   t 2   2e − t + t 2 
∴ y = y ( h ) + y ( p ) = e−t   +   =  −t 
 −1  −t   −e − t 
Problem 4: (10pts)
Use the matrix-vector method and the method of variation of parameters to find a genral solution of the
nonhomogenous linear system of ODEs given by
 −3 1  −6 
y′ = Ay + g =   y +   e −2t , t ≥ 0
1 -3  2
Solution:
Problem 5: (10pts)
Use the power series method and the method of reduction of order to find a genral solution of the
following second-order ODE with time-varying coefficients, x( x − 1) y ''+ (3 x − 1) y '+ y = 0 .
Solution:
Problem 6: (15pts)
Consider the following Lengendre’s equation, (1 − x 2 ) y ''− 2 xy '+ n( n + 1) y = 0 where n is the parameter.
∞
(a) Use the power series method to solve the equation using y( x) = ∑ am x m where the recurrence relation
m=0

(n − s )(n + s + 1)
of the coefficients are given by as + 2 = − as , s = 0,1, 2,3,... (6pts)
( s + 2)( s + 1)
(b) Show that the power series of the Lengendre’s equation will converge for x < 1 . (3pts)
(c) By setting n to be a positive integer, and choosing the coefficient an of the highest power x m as
(2n)!
an = n 2 , show that the Lengendre polynomial of the order n is
2 (n !)
M
(2n − 2 m)!
Pn ( x) = ∑ ( −1)m x n− 2 m where M=n/2, or (M-1)/2, whichever is an integer. (6pts).
m=0 2 m !(n − m)!(n − 2 m)!
n

Solution:
(a)
Dividing the Lengendre’s equation x2 x x2) and n(n + x2 )
of the new equation are analytic at x = 0, so that we may apply the power series method. Substituting
∞
y( x) = ∑ am x m and its derivatives into the the Lengendre’s equation, and denoting the constant n(n + 1)
m=0

simply by k, we obtain
∞ ∞ ∞
(1 − x 2 ) ∑ m( m − 1)am x m− 2 − 2 x ∑ mam x m−1 + k ∑ am x m = 0.
m=2 m =1 m=0

By writing the first expression as two separate series we have the equation
∞ ∞ ∞ ∞

∑ m(m − 1)a
m=2
m
x m− 2 − ∑ m( m − 1)am x m − ∑ 2 mam x m + ∑ kam x m = 0.
m=2 m =1 m=0
To obtain the same general power xs in all four series, set
m − 2 = s (thus m = 2 + s) in the first series and simply write s instead of m in the other three series. This
gives
∞ ∞ ∞ ∞

∑ (s + 2)(s + 1)as+2 xs − ∑ s(s − 1)as xs − ∑ 2sas xs + ∑ kas xs = 0.

s=0 s=2 s =1 s=0

We obtain the general formula

(n − s)(n + s + 1)
as + 2 = − a ( s = 0,1,).
( s + 2)( s + 1) s
∞
By inserting these expressions for the coefficients into y( x) = ∑ am x m we obtain
m=0

y ( x ) = a0 y1 ( x ) + a1 y2 ( x )
where
n(n − 1) 2 (n − 2)n(n + 1)(n + 3) 4
y1 ( x) = 1 − x + x − +
2! 4!
and
(n − 1)(n + 2) 3 (n − 3)(n − 1)(n + 2)(n + 4) 5
y 2 ( x) = x − x + x − +
3! 5!
(b) By using the ratio test, one obtains that
a (n − s)(n + s + 1)
R = lim s→∞ 1 / (| s+ 2 |) = lim s→∞ 1/| − |= 1
as ( s + 2)( s + 1)
So the radius of convergence of the series is 1, and it implies that x < 1 .

(c)
To makes pn(1) = 1 for every n , we choose the coefficient of the highest power xn as
 (2n)! 1 ⋅ 3 ⋅ 5 (2n − 1)
an = n 2
= (n a positive integer) and an = 1 if n = 0. Then we calculate the other
2 (n !) n!
(n − s)(n + s + 1)
coefficients from as+ 2 = − a ( s = 0,1,). (4), solved for as in terms of as+2, that is,
( s + 2)( s + 1) s
( s + 2)( s + 1) (2n − 2 m)!
as = − as+ 2 (s ≤ n − 2). Thus we obtain an− 2 m = ( −1)m n . Hence,
(n − s)(n + s + 1) 2 m !(n − m)!(n − 2 m)!
M
(2n − 2 m)!
the Lengendre polynomial of the order n is Pn ( x) = ∑ ( −1)m x n− 2 m where M=n/2, or
m=0 2 m !(n − m)!(n − 2 m)!
n

(M-1)/2, whichever is an integer.

Problem 7: (15pts)
Consider the Bessel equation, x 2 y ''+ xy '+ ( x 2 − v 2 ) y = 0, v ∈ R ≥ 0 where the parameter ν is a given real
number which is positive or zero. According to the Frobenius theory, the Bessel equation has a solution
∞
of the form y( x) = ∑ am x m+ r .
m=0

(a) Show that the indicial equation of the ODE is r 2 − v 2 = 0 .(2pts).

1
(b) Show that the coefficient recursion is as = − as − 2 , s = 2,3,... (2pts).
( s + 2v) s
(c) Show that a2 m +1 = 0, m = 0,1, 2,3,.... (2pts).
(d) From part (c), it implies that all odd coeficients are zero, and all even coefficients are given by
( −1)m a0
a2 m = 2 m , m = 1, 2, .
2 m !(ν + 1)(ν + 2)(ν + m)
1
Show that, by assuming a0 = v , one obtain a particular solution of the Bessel equation as
2 Γ(v + 1)
∞
( −1)m x 2 m
Jν ( x) = x ∑ 2 m+ν
ν
(3pts)
m=0 2 m ! Γ(ν + m + 1)
(e) Show that the derivative of the Bessel function Jν(x) with respect to x can be expressed by Jν −1 ( x) or
2ν
Jν +1 ( x) by (1) Jν −1 ( x) + Jν +1 ( x) = J ( x) (3pts), and (2) Jν −1 ( x) − Jν +1 ( x) = 2 Jν ' ( x) (3pts).
x ν

Solutions:
∞
Hence, according to the Frobenius theory, it has a solution of the form y( x) = ∑ am x m+ r (a0
m=0
∞
Substituting y( x) = ∑ am x m+ r and its first and second derivatives into the Bessel’s equation, we obtain
m=0
∞ ∞ ∞ ∞

∑ (m + r )(m + r − 1)am xm+r + ∑ (m + r )am xm+r + ∑ am xm+r +2 −ν 2 ∑ am xm+r = 0. To make all coefficients of
m=0 m=0 m=0 m=0

x , we have
m

∞ ∞

∑ (m + r )(m + r − 1)a
m=0
m
x m+ r → (r )(r − 1)a0 x r + (r + 1)(r )a1 x r +1 + ∑ ( m + 2 + r )( m + 2 + r − 1)am+ 2 x m+ r + 2
m=0
∞ ∞

∑ ( m + r )a
m=0
m
x m+ r ⇒ ra0 x r + (r + 1)a1 x r +1 + ∑ ( m + 2 + r )am+ 2 x m+ r + 2
m=0
∞ ∞
-ν 2 ∑ am x m+ r ⇒ -ν 2 a0 x r −ν 2 a1 x r +1 −ν 2 ∑ am+ 2 x m+ 2 + r
m=0 m=0

We get a general formula for all s,

(i ) (r(r − 1) + r − v 2 )a0 x r = 0, s = 0.
(ii ) ((r + 1)r + (r + 1) − v 2 )a1 x r +1 = 0, s = 1.
(iii ) (( s + r )( s + r − 1)as + ( s + r )as + as− 2 − v 2 as )x r + s = 0, s ≥ 2.

(a) We obtain the indicial equation from (i),(r + ν)(r − ν) = 0. The roots are r1 = ν r2 ν.
(b) Next, we find the coefficient recursion for r = r1 = ν>0.
From (ii) reduces to (2v+1) a1 =0 a1 =0
1
(c) From (iii), we have s(s+2v) as+ as-2 =0 as = - a , m = 1, 2, .
s( s + 2ν ) s− 2
(d) s(s+2v) as+ as-2 =0 implies that all odd coeficients are zero, i.e., a1 = a3 = a5 =…=0.
 a2 m − 2
For the even coeficients, we let s=2m, one obtains a2 m = - , m = 1, 2, .
2 m(ν + m)
2

We then obtain the recursion form for a2 , a4 , a6,…

a
a2 = - 2 0 ,m = 1
2 (ν + 1)
a a0
a4 = - 2 2 = ( −1)2 4 , m=2
2 (ν + 2) 2 2!(ν + 1)(ν + 2)

( −1)m a0
a2 m = , m = 1, 2, .
2 2 m m !(ν + 1)(ν + 2)(ν + m)

1
by letting a0 = , one obtains that
2 Γ(v + 1)
v

( −1)m ( −1)m
a2 m = = , m = 1, 2, .
2 2 m+ν m ! Γ(ν + 1)(ν + 1)(ν + 2) (ν + m) 2 2 m+ν m ! Γ(ν + m + 1)
∞
( −1)m x 2 m
Hence, a particular solution of the Bessel equation is Jν ( x) = xν ∑ 2 m +ν
.
m=0 2 m ! Γ(ν + m + 1)
(e)
(1)
( −1)m x 2 m∞ ∞
( −1)m x 2 m+2ν
x J v ( x) = x x ∑ 2 m+ν
v v ν
= ∑ 2 m+ν .
m =0 2 m ! Γ(ν + m + 1) m =0 2 m ! Γ(ν + m + 1)
(2 m + 2ν )( −1)m x 2 m+2ν −1 ∞ ( −1)m x 2 m+2ν −1
∞
( x J v ( x))' = ∑ 2 m+ν
v
= ∑ 2 m+ν −1
m =0 2 m ! Γ(ν + m + 1) m =0 2 m ! Γ(ν + m)
∞
( −1)m x 2 m+ν −1
= x ∑ 2 m+ν −1
ν
= x v J v-1 ( x)
m =0 2 m ! Γ((ν − 1) + m + 1)
(2)
∞
( −1)m x 2 m ∞
( −1)m x 2 m
x J v ( x) = x
-v -v
x ∑ 2 m+ν
ν
= ∑ 2 m+ν .
m=0 2 m ! Γ(ν + m + 1) m=0 2 m ! Γ(ν + m + 1)
2 m( −1)m x 2 m−1
∞ ∞
( −1)s+1 x 2 s+1
( x J v ( x))' = ∑ 2 m+ν
-v
= ∑ 2 s+ν +1
m=0 2 m ! Γ(ν + m + 1) s=0 2 s ! Γ(ν + s + 2)
( −1)s x 2 s
∞ ∞
( −1)s x 2 s
= − x∑ 2 s+ν +1 = −x x ∑ 2 s+ν +1
-v v+1

s=0 2 s ! Γ(ν + s + 2) s=0 2 s ! Γ(ν + s + 2)

= − x -v J v+1 ( x)