CHAPTER 2

THE Z TRANSFORM
DEFINITION
f(t) f*(t)

Let the output of an ideal sampler be denoted as 𝑓 ∗ (𝑡). Then the Laplace
transform of 𝑓 ∗ (𝑡) is 𝐹 ∗ (𝑠), where:
∞

𝐿 𝑓 ∗ (𝑡) = 𝐹 ∗ (𝑠) = ෍ 𝑓(𝑘𝑇)𝑒 −𝑘𝑇𝑠

𝑘=0

If we replace:
𝑒 𝑇𝑠 = 𝑧
Then,
1
𝑠= 𝑙𝑛𝑧
𝑇

2
To consider the z transform of the time function 𝑓(𝑡), consider the sampled
values of 𝑓(𝑡), 𝑓 0 , 𝑓 𝑇 , 𝑓 2𝑇 , ⋯ where T is the sampling period. The z
transform of 𝑓(𝑘) is given by 𝐹(𝑧) where:

𝐹 𝑧 =𝑍 𝑓 𝑡 = 𝑍 𝑓 𝑘𝑡 =𝑍 𝑓 𝑘 = ෍ 𝑓(𝑘)𝑧 −𝑘
𝑘=0

■ This is the one-sided z transform.

■ It is a tool to analyze discrete-time control system.
■ It is similar to Laplace Transform for continuous time system.
■ The dynamics of a linear discrete time system is characterized by a linear
difference equation. To determine the system’s response to a given input,
such a difference equation must be solved.

3
z transform of elementary
functions f(t)

1
Unit step function
t
0
1(𝑡) ; 0≤𝑡
𝑓 𝑡 =ቊ
0 ; 𝑡<0
∞ ∞

𝐹 𝑧 = 𝑍 𝑓(𝑡) = 𝑍 1 𝑡 = ෍ 1 𝑧 −𝑘 = ෍ 𝑧 −𝑘
𝑘=0 𝑘=0

= 1 + 𝑧 −1 + 𝑧 −2 + 𝑧 −3 + ⋯
1 𝑧
= ×
1 − 𝑧 −1 𝑧

𝑧
𝐹 𝑧 =
𝑧−1

4
Unit ramp function
f(t)
𝑡 ; 0≤𝑡
𝑓 𝑡 =ቊ
0 ; 𝑡<0

𝑓 𝑘𝑇 = 𝑘𝑇 𝑘 = 0,1,2,3, … t
0
∞ ∞

𝐹 𝑧 = 𝑍 𝑡 = ෍ 𝑘𝑇𝑧 −𝑘 = 𝑇 ෍ 𝑘𝑧 −𝑘
𝑘=0 𝑘=0

= 𝑇(𝑧 −1 + 2𝑧 −2 + 3𝑧 −3 + ⋯ )

𝑇𝑧 −1 𝑧2
= × 𝑧2
1 − 𝑧 −1 2

𝑇𝑧
𝐹 𝑧 = 2
𝑧−1

5
Polynomial function

𝑓 𝑡 =ቊ
𝑎𝑘 ; 𝑘 = 0,1,2, … where a is a constant
0 ; 𝑘<0

𝐹 𝑧 = 𝑍 𝑎𝑘 = ෍ 𝑎𝑘 𝑧 −𝑘
𝑘=0

= 1 + 𝑎1 𝑧 −1 + 𝑎2 𝑧 −2 + 𝑎3 𝑧 −3 + ⋯

1
=
1 − 𝑎𝑧 −1

𝑧
𝐹 𝑧 =
𝑧−𝑎

6
Exponential function

𝑒 −𝑎𝑡 ; 0≤𝑡
𝑓 𝑡 =ቊ
0 ; 𝑡<0

𝑓 𝑘𝑇 = 𝑒 −𝑎𝑘𝑇 ; 𝑘 = 0,1,2,3, …

𝐹 𝑧 = 𝑍 𝑒 −𝑎𝑘𝑇 = ෍ 𝑒 −𝑎𝑘𝑇 𝑧 −𝑘
𝑘=0

= 1 + 𝑒 −𝑎𝑇 𝑧 −1 + 𝑒 −2𝑎𝑇 𝑧 −2 + 𝑒 −3𝑎𝑇 𝑧 −3 + ⋯

1
=
1 − 𝑒 −𝑎𝑇 𝑧 −1

𝑧
𝐹 𝑧 =
𝑧 − 𝑒 −𝑎𝑇

7
Example 1

1 𝑗𝜔𝑡
sin 𝜔𝑡 ; 0≤𝑡 sin 𝜔𝑡 = 𝑒 − 𝑒 −𝑗𝜔𝑡
𝑓 𝑡 =ቊ 2𝑗
0 ; 𝑡<0
1
𝑍 𝑒 −𝑎𝑡 =
1 − 𝑒 −𝑎𝑇 𝑧 −1

1 𝑗𝜔𝑡
𝐹 𝑧 = 𝑍 sin 𝜔𝑡 = 𝑍 𝑒 − 𝑒 −𝑗𝜔𝑡
2𝑗

1 1 1
= −
2𝑗 1 − 𝑒 𝑗𝜔𝑡 𝑧 −1 1 − 𝑒 −𝑗𝜔𝑡 𝑧 −1

1 ቀ𝑒 𝑗𝜔𝑇 − 𝑒 −𝑗𝜔𝑡 )𝑧 −1
=
2𝑗 1 − 𝑒 𝑗𝜔𝑇 + 𝑒 −𝑗𝜔𝑡 𝑧 −1 + 𝑧 −2

𝑧 −1 sin 𝜔𝑇 𝑧2
= × 𝑧2
1 − 2𝑧 −1 cos 𝜔𝑇 + 𝑧 −2

𝑧 sin 𝜔𝑇
𝐹 𝑧 =
𝑧 2 − 2𝑧 cos 𝜔𝑇 + 1
8
Example 2
1 𝑗𝜔𝑡
cos 𝜔𝑡 ; 0≤𝑡 cos 𝜔𝑡 = 𝑒 + 𝑒 −𝑗𝜔𝑡
𝑓 𝑡 =ቊ 2
0 ; 𝑡<0
1
𝑍 𝑒 −𝑎𝑡 =
1 − 𝑒 −𝑎𝑇 𝑧 −1

1 𝑗𝜔𝑡
𝐹 𝑧 = 𝑍 cos 𝜔𝑡 = 𝑍 𝑒 + 𝑒 −𝑗𝜔𝑡
2

1 1 1
= +
2 1 − 𝑒 𝑗𝜔𝑡 𝑧 −1 1 − 𝑒 −𝑗𝜔𝑡 𝑧 −1

1 2 − (𝑒 −𝑗𝜔𝑇 + 𝑒 𝑗𝜔𝑡 )𝑧 −1
=
2 1 − 𝑒 𝑗𝜔𝑇 + 𝑒 −𝑗𝜔𝑡 𝑧 −1 + 𝑧 −2

1 − 𝑧 −1 cos 𝜔𝑡
=
1 − 2𝑧 −1 cos 𝜔𝑇 + 𝑧 −2

𝑧 2 − 𝑧 cos 𝜔𝑇
𝐹(𝑧) = 2
𝑧 − 2𝑧 cos 𝜔𝑇 + 1

9
Example 3
1
𝐹 𝑠 = Laplace Transform form
𝑠(𝑠 + 1)

Method 1: Convert 𝐹(𝑠) into 𝑓(𝑡), then find 𝐹(𝑧) of 𝑓(𝑡).

Method 2: Expand 𝐹(𝑠) into partial fraction and use z transform table to find
the z transform.

Using Method 1:
The inverse LT of 𝐹(𝑠) is 𝑓 𝑡 where (from LT table), 𝑓 𝑡 = 1 − 𝑒 −𝑡 ; 0 ≤ 𝑡 .
Then;
𝐹 𝑧 = 𝑍 1 − 𝑒 −𝑡
1 1
𝐹(𝑧) = −
1 − 𝑧 −1 1 − 𝑒 −𝑇 𝑧 −1
൫1 − 𝑒 −1 )𝑧 −1 (1 − 𝑒 −𝑇 )𝑧
= =
1 − 𝑧 −1 )(1 − 𝑒 −𝑇 𝑧 −1 𝑧 − 1)(𝑧 − 𝑒 −𝑇

Just as for LT, table of z transform of common function is useful. **Table2-1, page 29,
Ogata.

10
 𝐹(𝑠) 𝑓(𝑡) 𝑓 𝑘𝑇 𝐹(𝑧)

𝛿0 (𝑘)
1. 1, 𝑘=0 1
0, 𝑘≠0

𝛿0 (𝑛 − 𝑘)
2. 1, 𝑛=𝑘 𝑧 −𝑘
0, 𝑛≠𝑘

1 1
3. 1(𝑡) 1(𝑘)
𝑠 1 − 𝑧 −1

1 1
4. 𝑒 −𝑎𝑡 𝑒 −𝑎𝑘𝑇
(𝑠 + 𝑎) 1 − 𝑒 −𝑎𝑇 𝑧 −1

1 𝑇𝑧 −1
5. 𝑡 𝑘𝑇
𝑠2 (1 − 𝑧 −1 )2

2 𝑇 2 𝑧 −1 (1 + 𝑧 −1)
6. 𝑡2 (𝑘𝑇)2
𝑠3 (1 − 𝑧 −1 )3

6 𝑇 3 𝑧 −1 (1 + 4𝑧 −1 + 𝑧 −2 )
7. 𝑡3 (𝑘𝑇)3
𝑠4 (1 − 𝑧 −1 )4

𝑎 (1 − 𝑒 −𝑎𝑇 )𝑧 −1
8. 1 − 𝑒 −𝑎𝑡 1 − 𝑒 −𝑎𝑘𝑇
𝑠(𝑠 + 𝑎) (1 − 𝑧 −1 )(1 − 𝑒 −𝑎𝑇 𝑧 −1)

𝑏−𝑎 (𝑒 −𝑎𝑇 − 𝑒 −𝑏𝑇 )𝑧 −1

9. 𝑒 −𝑎𝑡 − 𝑒 −𝑏𝑡 𝑒 −𝑎𝑘𝑇 − 𝑒 −𝑏𝑘𝑇
(𝑠 + 𝑎)(𝑠 + 𝑏) (1 − 𝑒 −𝑎𝑇 𝑧 −1 )(1 − 𝑒 −𝑏𝑇 𝑧 −1 )

1 𝑇𝑒 −𝑎𝑇 𝑧 −1
10. 𝑡𝑒 −𝑎𝑡 𝑘𝑇𝑒 −𝑎𝑘𝑇
(𝑠 + 𝑎)2 (1 − 𝑒 −𝑎𝑇 𝑧 −1 )2

Table 2.1: z transform of some common functions

11
 𝑠 1 − 1 + 𝑎𝑇 𝑒 −𝑎𝑇 𝑧 −1
11. (1 − 𝑎𝑡)𝑒 −𝑎𝑡 (1 − 𝑎𝑘𝑇)𝑒 −𝑎𝑘𝑇
(𝑠 + 𝑎)2 (1 − 𝑒 −𝑎𝑇 𝑧 −1 )2

2 𝑇 2 𝑒 −𝑎𝑇 1 + 𝑒 −𝑎𝑇 𝑧 −1 𝑧 −1
12. 𝑡 2 𝑒 −𝑎𝑡 (𝑘𝑇)2 𝑒 −𝑎𝑘𝑇
(𝑠 + 𝑎)3 (1 − 𝑒 −𝑎𝑇 𝑧 −1 )3

𝑎 1 − 𝑒 −𝑎𝑡 1 − 𝑒 −𝑎𝑘𝑇 𝑎𝑇 − 1 + 𝑒 −𝑎𝑇 + 1 − 𝑒 −𝑎𝑇 − 𝑎𝑇𝑒 −𝑎𝑇 𝑧 −1 𝑧 −1

13. 𝑡− 𝑘𝑇 −
𝑠 2 (𝑠 + 𝑎) 𝑎 𝑎 𝑎 1 − 𝑧 −1 2 1 − 𝑒 −𝑎𝑇 𝑧 −1

𝑎2 𝑎𝑇 − 1 + 𝑒 −𝑎𝑇 + (1 − 𝑒 −𝑎𝑇 − 𝑎𝑇𝑒 −𝑎𝑇 )𝑧 −1 𝑧 −1

14. 𝑎𝑡 − 1 + 𝑒 −𝑎𝑡 𝑎𝑘𝑇 − 1 + 𝑒 −𝑎𝑘𝑇
𝑠 2 (𝑠 + 𝑎) (1 − 𝑧 −1 )(1 − 𝑒 −𝑎𝑇 𝑧 −1 )

𝑎2 1 1 𝑎𝑇𝑒 −𝑎𝑇 𝑧 −1
15. 1 − (1 + 𝑎𝑡)𝑒 −𝑎𝑡 1 − (1 + 𝑎𝑘𝑇)𝑒 −𝑎𝑘𝑇 − −
𝑠(𝑠 + 𝑎)2 1 − 𝑧 −1 1 − 𝑒 −𝑎𝑇 𝑧 −1 1 − 𝑒 −𝑎𝑇 𝑧 −1 2

𝜔 𝑧 −1 sin 𝜔𝑇
16. sin 𝜔𝑡 sin 𝜔𝑘𝑇
𝑠2 + 𝜔2 1 − 2𝑧 −1 cos 𝜔𝑇 + 𝑧 −2

𝑠 1 − 𝑧 −1 cos 𝜔𝑇
17. cos 𝜔𝑡 cos 𝜔𝑘𝑇
𝑠2 + 𝜔2 1 − 2𝑧 −1 cos 𝜔𝑇 + 𝑧 −2

1 1 −𝑎𝑡 1 −𝑎𝑘𝑇 1 𝑧 −1 𝑒 −𝑎𝑇 sin 𝛽𝑇

18. 𝑒 sin 𝛽𝑡 𝑒 sin 𝛽𝑘𝑇
(𝑠 + 𝑎)2 +𝛽 2 𝛽 𝛽 𝛽 1 − 2𝑧 𝑒 −𝑎𝑇 cos 𝛽𝑇 + 𝑒 −2𝑎𝑇 𝑧 −2
−1

𝑠+𝑎 1 − 𝑧 −1 𝑒 −𝑎𝑇 cos 𝜔𝑇

19. 𝑒 −𝑎𝑡 cos 𝜔𝑡 𝑒 −𝑎𝑘𝑇 cos 𝜔𝑘𝑇
(𝑠 + 𝑎)2 +𝜔 2 1 − 2𝑧 −1 𝑒 −𝑎𝑇 cos 𝜔𝑇 + 𝑒 −2𝑎𝑇 𝑧 −2

(𝐴 + 𝐵𝑧 −1 )𝑧 −1
1 (1 − 𝑧 −1 )(1 − 2𝑧 −1 𝑒 −𝜔𝑇 cos 𝛽𝑇 + 𝑒 −2𝜔𝑇 𝑧 −2 )

𝜔2 + 𝛽 2 − 𝑒 −𝜔𝑡 ቆcos 𝛽𝑡 𝜔 𝜔
20. 1 − 𝑒 −𝜔𝑘𝑇 cos 𝛽𝑘𝑇 + sin 𝛽𝑘𝑇 𝐴 = 1 − 𝑒 −𝜔𝑇 cos 𝛽𝑇 + sin 𝛽𝑇
𝑠 (𝑠 + 𝜔)2 +𝛽 2 𝛽 𝛽
𝜔
+ sin 𝛽𝑡ቇ
𝛽
𝜔
𝐵 = 𝑒 −2𝜔𝑇 + 𝑒 −𝜔𝑇 sin 𝛽𝑇 − cos 𝛽𝑇
𝛽

Table 2.1: z transform of some common functions - continued

12
 𝐴 + 𝐵𝑧 −1 𝑧 −1
(1 − 𝑒 −𝑎𝑇 𝑧 −1 )(1 −
𝑒 −𝑏𝑇 𝑧 −1 )(1 − 𝑧 −1)
1 𝑒 −𝑎𝑡
+
1 𝑎𝑏 𝑎(𝑎 − 𝑏) 1 𝑒 −𝑎𝑘𝑇 𝑒 −𝑏𝑘𝑇 𝑏 1 − 𝑒 −𝑎𝑇 − 𝑎(1 − 𝑒 −𝑏𝑇 )
21. + + 𝐴=
𝑠(𝑠 + 𝑎)(𝑠 + 𝑏) 𝑒 −𝑏𝑡 𝑎𝑏 𝑎(𝑎 − 𝑏) 𝑏(𝑏 − 𝑎) 𝑎𝑏(𝑏 − 𝑎)
+
𝑏(𝑏 − 𝑎)
𝑎𝑒 −𝑎𝑇 1 − 𝑒 −𝑏𝑇 − 𝑏𝑒 −𝑏𝑇 (1 − 𝑒 −𝑎𝑇 )
𝐵=
𝑎𝑏(𝑏 − 𝑎)

1
22. 𝑎𝑘
1 − 𝑎𝑧 −1

𝑎𝑘−1 𝑧 −1
23.
𝑘 = 1,2,3, ⋯ 1 − 𝑎𝑧 −1

𝑧 −1
24. 𝑘𝑎𝑘−1
(1 − 𝑎𝑧 −1 )2

𝑧 −1 (1 + 𝑎𝑧 −1 )
25. 𝑘 2 𝑎𝑘−1
(1 − 𝑎𝑧 −1 )3

𝑧 −1 (1 + 4𝑎𝑧 −1 + 𝑎2𝑧 −2)

26. 𝑘 3 𝑎𝑘−1
(1 − 𝑎𝑧 −1 )4

𝑧 −1 (1 + 11𝑎𝑧 −1 + 11𝑎2𝑧 −2 + 𝑎3 𝑧 −1 )
27. 𝑘 4 𝑎𝑘−1
(1 − 𝑎𝑧 −1 )5

1
28. 𝑎𝑘 cos 𝑘𝜋
1 + 𝑎𝑧 −1

Table 2.1: z transform of some common functions - continued

13
Some theorems and
properties of z transform
Multiplication by a constant:
If 𝐹(𝑧) is the z transform of 𝑓(𝑡), then, 𝑍 𝑎𝑓(𝑡) = 𝑎𝐹 𝑧 , where 𝑎 is a constant.
By definition:
∞ ∞

𝑍 𝑎𝑓(𝑡) = ෍ 𝑎𝑓(𝑘𝑇)𝑧 −𝑘 = 𝑎 ෍ 𝑓(𝑘𝑇)𝑧 −𝑘 = 𝑎𝐹(𝑧)

𝑘=0 𝑘=0

Multiplication by 𝑎𝑘 :
If 𝐹(𝑧) is the z transform of 𝑓(𝑘), then 𝑍 𝑎𝑘 𝑓 𝑘 = 𝐹(𝑎−1 𝑧).
Prove:
∞ ∞

𝑍 𝑎𝑘 𝑓 𝑘 = ෍ 𝑎𝑘 𝑓(𝑘)𝑧 −𝑘 = ෍ 𝑓(𝑘)(𝑎−1 𝑧)−𝑘 = 𝐹(𝑎−1 𝑧)

𝑘=0 𝑘=0

14
Linearity:
If 𝑓(𝑘) and 𝑔(𝑘) are z-transformable and  and  are scalars, then 𝑥(𝑘) formed by a linear
combination:
𝑥 𝑘 = 𝛼𝑓 𝑘 + 𝛽𝑔 𝑘
has the z transform,
𝑋 𝑧 = 𝛼𝐹 𝑧 + 𝛽𝐺(𝑧)
where 𝐹(𝑧) and 𝐺(𝑧) are the z transforms of 𝑓(𝑘) and 𝑔(𝑘) respectively.
Prove:
𝐹 𝑧 =𝑍 𝑓 𝑘 = 𝑍[𝛼𝑓 𝑘 + 𝛽𝑔 𝑘 ]

= ෍ [𝛼𝑓 𝑘 + 𝛽𝑔 𝑘 ] 𝑧 −𝑘
𝑘=0

∞ ∞

= 𝛼 ෍ 𝑓 𝑘 𝑧 −𝑘 + 𝛽 ෍ 𝑔 𝑘 𝑧 −𝑘
𝑘=0 𝑘=0

= 𝛼𝑍[𝑓 𝑘 ] + 𝛽𝑍[𝑔 𝑘 ]

= 𝛼𝐹 𝑧 + 𝛽𝐺(𝑧)

15
Real shifting:
If 𝑓 𝑡 = 0 for 𝑡 < 0 and 𝑓(𝑡) has the z transform 𝐹(𝑧), then
𝑍 𝑓(𝑡 − 𝑛𝑇) = 𝑧 −𝑛 𝐹 𝑧
and
𝑛−1

𝑍 𝑓(𝑡 + 𝑛𝑇) = 𝑧 𝑛 𝐹 𝑧 − ෍ 𝑓(𝑘𝑡)𝑧 −𝑘

𝑘=0

where 𝑛 is zero or a positive integer.

Prove:
∞

𝑍 𝑓(𝑡 − 𝑛𝑇) = ෍ 𝑓(𝑡 − 𝑛𝑇) 𝑧 −𝑘

𝑘=0
∞

= 𝑧 −𝑛 ෍ 𝑓(𝑘𝑇 − 𝑛𝑇)𝑧 −(𝑘−𝑛)

𝑘=0

By defining 𝑚 = 𝑘 − 𝑛, this last equation can be written as follows:

∞

𝑍 𝑓(𝑡 − 𝑛𝑇) = 𝑧 −𝑛 ෍ 𝑓(𝑚𝑇) 𝑧 −𝑚

𝑚=−𝑛

Since 𝑥 𝑚𝑇 = 0 for 𝑚 < 0, we may change the lower limit of the summation from 𝑚 = −𝑛 to
𝑚 = 0. Hence,
∞

𝑍 𝑓(𝑡 − 𝑛𝑇) = 𝑧 −𝑛 ෍ 𝑓(𝑚𝑇) 𝑧 −𝑚 = 𝑧 −𝑛 𝐹(𝑧)

𝑚=0
−𝑛
Thus, multiplication of a z transform by 𝑧 has the effect of delaying the time function 𝑓(𝑡)
by 𝑛𝑇.

16
Example 4:

Given:
1
𝑍 𝑒 −𝑎𝑘𝑇 =
1 − 𝑒 −𝑎𝑇 𝑧 −1
Thus,
1 1
𝑍 𝑒 −𝑎(𝑘−3)𝑇 = 𝑧 −3 =
1 − 𝑒 −𝑎𝑇 𝑧 −1 𝑧 2 (𝑧 − 𝑒 −𝑎𝑇 )
Also,
𝑛−1
1
𝑍 𝑒 −𝑎(𝑘+2)𝑇 = 𝑧 2 − ෍ 𝑒 −𝑎𝑘𝑇 𝑧 −𝑘
1− 𝑒 −𝑎𝑇 𝑧 −1
𝑘=0

1
𝑍 𝑒 −𝑎(𝑘+2)𝑇 = 𝑧2 − 1 − 𝑒 −𝑎𝑇 𝑧 −1
1− 𝑒 −𝑎𝑇 𝑧 −1

17
Initial value theorem:
If 𝑓(𝑡) has the transform 𝐹(𝑧) and if lim 𝐹(𝑧) exists; then, the initial value 𝑓(0)
𝑧→∞
of 𝑓(𝑡) is given by,

𝑓 0 = lim 𝐹(𝑧)
𝑧→∞

Prove:
∞

𝐹 𝑧 = ෍ 𝑓(𝑘)𝑧 −𝑘 = 𝑓 0 + 𝑓 1 𝑧 −1 + 𝑓 2 𝑧 −2 + ⋯
𝑘=0

Letting 𝑧 → ∞ in this last equation, we obtain 𝑓 0 = lim 𝐹(𝑧). The behaviour of

𝑧→∞
the signal in the neighborhood of 𝑡 = 0 or 𝑘 = 0 can thus be determined by the
behaviour of 𝐹(𝑧) at 𝑧 = ∞. The initial value theorem is convenient for checking z
transform calculations for possible errors. Since 𝑓(0) is usually known, a check of
the initial value by lim 𝐹(𝑧) can easily spot errors in 𝐹(𝑧), if any exist.
𝑧→∞

18
Example 4: Initial value theorem

Determine the initial value, 𝑥(0), if the z transform of 𝑥(t) is given by,

൫1 − 𝑒 −𝑇 )𝑧 −1
𝑋(𝑧) =
1 − 𝑧 −1 )(1 − 𝑒 −𝑇 𝑧 −1

By using the initial value theorem,

𝑥(0) = lim 𝑋 𝑧
𝑧→∞
∞
൫1 − 𝑒 −𝑇 )𝑧 −1
= lim
𝑧→∞ 1 − 𝑧 −1 )(1 − 𝑒 −𝑇 𝑧 −1

𝑥(0) = 0

■ Note that 𝑋(𝑧) is the z transform of 𝑥 𝑡 = 1 − 𝑒 −𝑡 and thus 𝑥(0) = 0

19
Final value theorem:
The final value of 𝑓(𝑘), that is, the value of 𝑓(𝑘) as 𝑘 approaches infinity, can be given by,

lim 𝑓(𝑘) = lim 1 − 𝑧 −1 𝐹(𝑧)

𝑘→∞ 𝑧→1

Prove:
∞

𝑍𝑓 𝑘 = 𝐹(𝑧) = ෍ 𝑓(𝑘)𝑧 −𝑘
𝑘=0
∞

𝑍 𝑓 𝑘−1 = 𝑧 −1 𝐹(𝑧) = ෍ 𝑓(𝑘 − 1)𝑧 −𝑘

𝑘=0

Hence,
∞ ∞
−𝑘
෍ 𝑓(𝑘)𝑧 − ෍ 𝑓 𝑘 − 1 𝑧 −𝑘 = 𝐹 𝑧 − 𝑧 −1 𝐹(𝑧)
𝑘=0 𝑘=0

Taking the limit as z approaches unity, we have

∞ ∞
−𝑘
lim ෍ 𝑓(𝑘) 𝑧 − ෍ 𝑓(𝑘 − 1) 𝑧 −𝑘 = lim 1 − 𝑧 −1 𝐹(𝑧)
𝑧→1 𝑧→1
𝑘=0 𝑘=0

Because of the assumed stability condition and the condition that 𝑓 𝑘 = 0 for 𝑘 < 0, the left-hand side of this last
equation becomes
∞

෍ 𝑓 𝑘 − 𝑓(𝑘 − 1) = 𝑓 0 − 𝑓(−1) + 𝑓 1 − 𝑓(0) + 𝑓 2 − 𝑓(1) + ⋯ = 𝑓 ∞ = lim 𝑓(𝑘)

𝑘→∞
𝑘=0

Hence,
lim 𝑓(𝑘) = lim 1 − 𝑧 −1 𝐹(𝑧)
𝑘→∞ 𝑧→1

The final value theorem is very useful in determining the behaviour of 𝒇(𝒌) as 𝒌 → ∞ from its z transform 𝑭(𝒛).

20
Example 4: Final value theorem
By using the final value theorem, determine the final value 𝑥 ∞ of,
1 1
𝑋 𝑧 = −1
− −𝑎𝑇
; 𝑎>0
1−𝑧 1−𝑒 𝑧 −1

By applying the final value theorem to 𝑋 𝑧 , we obtain,

𝑥 ∞ = lim 1 − 𝑧 −1 𝑋(𝑧)
𝑧→1

1 1
= lim 1 − 𝑧 −1 −1
− −𝑎𝑇
𝑧→1 1−𝑧 1−𝑒 𝑧 −1

1 − 𝑧 −1
= lim 1 −
𝑧→1 1 − 𝑒 −𝑎𝑇 𝑧 −1

𝑥 ∞ =1

𝑋 𝑧 is actually the z transform of 𝑥 𝑡 = 1 − 𝑒 −𝑎𝑇 . By substituting 𝑡 = ∞ in this equation, we have 𝑥 ∞ =

lim 1 − 𝑒 −𝑎𝑡 = 1.
𝑡→∞

*Important theorems and properties of the z transform are summarized in Table 2.2, page 38, Ogata

21
 Sequence Transform
∞
1. 𝑓(𝑘) 𝐹 𝑧 = ෍ 𝑓(𝑘)𝑧 −𝑘
𝑘=0

2. 𝑎1 𝑓1 𝑘 + 𝑎2 𝑓2 𝑘 𝑎1 𝐹1 𝑧 + 𝑎2 𝐹2 𝑧

3. 𝑓 𝑘−𝑛 ; 𝑛≥0 𝑧 −𝑛 𝐹(𝑧)

𝑛−1
4. 𝑓 𝑘+𝑛 ; 𝑛≥1 𝑧𝑛 𝐹 𝑧 − ෍ 𝑓(𝑘)𝑧 −𝑘
𝑘=0

5. 𝑒 𝑎𝑘 𝑓(𝑘) 𝐹(𝑧𝑒 −𝑎 )
𝑑𝐹(𝑧)
6. 𝑘𝑓(𝑘) −𝑧
𝑑𝑧
7. 𝑓1 𝑘 ∗ 𝑓2 𝑘 𝐹1 (𝑧)𝐹2 (𝑧)

8. Initial value: 𝑓 0 = 𝑧→∞

lim 𝐹(𝑧)

9. Final value: 𝑓 ∞ = 𝑧→1

lim (𝑧 − 1)𝐹(𝑧), if 𝑓(∞) exists

Table 2.2: Summary of important theorems and properties of z transform

22
23