Digital Signal Processing

Lecture 4 – Linear Constant-Coefficient

Difference Equations (LCCDE)

Alp Ertürk
[email protected]
Linear Constant-Coefficient
Difference Equations (LCCDE)
• We calculate the output (response) of an LTI system using
the input and the system’s impulse response

• However, as n gets larger, the convolution sum results in

increased computation time and memory requirement

• Systems for which the output can be represented in terms

of the input signal’s present and past values and the output
signal’s past values reduce these requirements
LCCDE
• Consider the bank account example we discussed in
previous lectures

• The output at a given instant was:

𝑦 𝑛 = 𝑦 𝑛 − 1 + 0.1𝑦 𝑛 − 1 + 𝑥[𝑛]

𝑦 𝑛 = 1.1𝑦 𝑛 − 1 + 𝑥[𝑛]

• This equation is a first degree linear constant coefficient

difference equation (LCCDE)
LCCDE
• All systems cannot be represented in terms of an LCCDE !

• An LCCDE is denoted as:

𝑁 𝑀

෍ 𝑎𝑘 𝑦[𝑛 − 𝑘] = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘]
𝑘=0 𝑘=0

OR:

𝑀 𝑁
1
𝑦𝑛 = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘] − ෍ 𝑎𝑘 𝑦[𝑛 − 𝑘]
𝑎0
𝑘=0 𝑘=1
LCCDE
• In most cases, 𝑎0 is distributed to the other coefficients,
resulting in:
𝑀 𝑁

𝑦 𝑛 = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘] − ෍ 𝑎𝑘 𝑦[𝑛 − 𝑘]
𝑘=0 𝑘=1
LCCDE
• As a special case, if the output of the system does not
depend on past output values:
𝑀

𝑦 𝑛 = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘]
𝑘=0

• If an impulse signal is applied to this system:

𝑀
𝑏 , 0≤𝑛≤𝑀
ℎ 𝑛 = ෍ 𝑏𝑘 𝛿[𝑛 − 𝑘] = ቊ 𝑘
0 , 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
𝑘=0
LCCDE
• In this special case, as the impulse response of the system is
the impulse coefficients, we have:
𝑀 𝑀

𝑦 𝑛 = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘] = ෍ ℎ[𝑘]𝑥[𝑛 − 𝑘]
𝑘=0 𝑘=0

• In other words, the LCCDE and the convolution sum will

have a similar representation in this case
LCCDE
• Example: The output of a system is defined as 𝑦 𝑛 =
𝑥 𝑛 + 𝑥[𝑛 − 1]. Find the output for the input signal 𝑥 𝑛
shown below.
LCCDE
𝑛 < −1 ∶ 𝑥 𝑛 = 0 𝑎𝑛𝑑 𝑦 𝑛 = 0

𝑛 = −1 ∶ 𝑦 −1 = 𝑥 −1 + 𝑥 −2 = 1

𝑛 = 0 ∶ 𝑦 0 = 𝑥 0 + 𝑥 −1 = 3

𝑛 =1 ∶ 𝑦 1 =𝑥 1 +𝑥 0 =2

𝑛 = 2 ∶ 𝑦 2 = 𝑥 2 + 𝑥 1 = −1

𝑛 = 3 ∶ 𝑦 3 = 𝑥 3 + 𝑥 2 = −1

𝑛≥4 ∶ 𝑥 𝑛 = 0 𝑎𝑛𝑑 𝑦 𝑛 = 0
LCCDE
• The output is:
LCCDE: Recursive
• If an LCCDE system’s output depends only on the input’s
present and past values, this system is termed as non-
recursive

• If, on the other hand, the output also depends on the past
values of the output, then this system is termed as recursive

• For recursive systems, knowing only the input values are not
sufficient, and the starting conditions for the system should
be known / provided.
LCCDE: Recursive
• Example: Find the output of the accumulator system for the
input signal 𝑥 𝑛 shown below.
LCCDE: Recursive
• The accumulator system is:

𝑦 𝑛 = ෍ 𝑥[𝑘]
𝑘=−∞

• In LCCDE representation:

𝑦 𝑛 = 𝑦 𝑛 − 1 + 𝑥[𝑛]
LCCDE: Recursive
• Note that for this case we need an initial value for y:

𝑦 0 = 𝑦 −1 + 𝑥 0 , 𝑦 −1 =?

• Assigning or assuming an initial value enables us to solve

this problem.

• For example, let’s assume that the initial conditions have

been provided as 𝑦 −1 = 0. Then, we can solve this
problem in a manner quite similar to the previous example.
LCCDE: Recursive
• The output is:
LCCDE: Recursive
• Example: Consider a system with the input – output
relationship of 𝑦 𝑛 = 𝑎𝑦 𝑛 − 1 + 𝑥 𝑛 .

• If the input signal is 𝑥 𝑛 = 𝑏𝛿[𝑛], find the output for the

initial conditions of:
• 1) 𝑦 −1 = 1
• 2) 𝑦 −1 = 0
LCCDE: Recursive
• 1) 𝑦 −1 = 1 :

𝑦 0 = 𝑎𝑦 −1 + 𝑥 0 = 𝑎 + 𝑏

𝑦 1 = 𝑎𝑦 0 + 𝑥 1 = 𝑎 𝑎 + 𝑏 + 0 = 𝑎2 + 𝑎𝑏

𝑦 2 = 𝑎𝑦 1 + 𝑥 2 = 𝑎3 + 𝑎2 𝑏

𝑦 3 = 𝑎𝑦 2 + 𝑥 3 = 𝑎4 + 𝑎3 𝑏

...
LCCDE: Recursive

𝑦 −1 = 𝑎𝑦 −2 + 𝑥 −1 = 1 ⇒ 𝑦 −2 = 𝑎−1

𝑦 −2 = 𝑎𝑦 −3 + 𝑥 −2 = 𝑎−1 ⇒ 𝑦 −3 = 𝑎−2

𝑦 −4 = 𝑎−3

𝑦 −5 = 𝑎−4

...
LCCDE: Recursive
• Overall, the system output is:

𝑎 𝑛+1 + 𝑎 𝑛 𝑏 𝑛≥0
𝑦 𝑛 =ቊ
𝑎𝑛+1 𝑛<0

= 𝑎𝑛+1 + 𝑎𝑛 𝑏𝑢[𝑛]

• Note that this system is non-linear and time varying!

𝑥 𝑛 = 0 ⇒ 𝑦 𝑛 = 𝑎𝑛+1

𝑥 𝑛 = 𝑏𝛿 𝑛 − 𝑛0 ⇒ 𝑦 𝑛 = 𝑎𝑛+1 + 𝑎𝑛−𝑛0 𝑏𝑢[𝑛 − 𝑛0 ]

LCCDE: Recursive
• 2) 𝑦 −1 = 0 :

𝑦 0 = 𝑎𝑦 −1 + 𝑥 0 = 0 + 𝑏 = 𝑏

𝑦 1 = 𝑎𝑦 0 + 𝑥 1 = 𝑎 𝑏 + 0 = 𝑎𝑏

𝑦 2 = 𝑎𝑦 1 + 𝑥 2 = 𝑎2 𝑏

𝑦 3 = 𝑎𝑦 2 + 𝑥 3 = 𝑎3 𝑏

...
LCCDE: Recursive

𝑦 −1 = 𝑎𝑦 −2 + 𝑥 −1 = 0 ⇒ 𝑦 −2 = 0

𝑦 −2 = 𝑎𝑦 −3 + 𝑥 −2 = 0 ⇒ 𝑦 −3 = 0

𝑦 −4 = 0

𝑦 −5 = 0

...
LCCDE: Recursive
• Overall, in this case the system output is:

𝑦 𝑛 = 𝑎𝑛 𝑏𝑢[𝑛]

• Note that this system is linear and time-invariant

LCCDE: Homogeneous-Particular
𝑁 𝑀

෍ 𝑎𝑘 𝑦[𝑛 − 𝑘] = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘]
𝑘=0 𝑘=0

• If 𝑥 𝑛 = 0 :

෍ 𝑎𝑘 𝑦[𝑛 − 𝑘] = 0
𝑘=0

• This equation is the system’s textural / homogeneous

equation. Solving this equation for 𝑦 𝑛 gives us the
homogeneous response / solution, 𝑦ℎ 𝑛
LCCDE: Homogeneous-Particular
𝑁 𝑀

෍ 𝑎𝑘 𝑦[𝑛 − 𝑘] = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘]
𝑘=0 𝑘=0

• The system’s response for any particular input 𝑥 𝑛 , is

termed as the particular response, and denoted by 𝑦𝑝 𝑛

• A system’s total response can be obtained by:

𝑦 𝑛 = 𝑦ℎ 𝑛 + 𝑦𝑝 𝑛
LCCDE: Zero-input and zero-state
• However, finding the system’s output for a particular input
signal can be challenging

• An alternative is to represent the system response as:

𝑦 𝑛 = 𝑦𝑧𝑖 𝑛 + 𝑦𝑧𝑠 𝑛

• Zero-input response ( 𝑦𝑧𝑖 𝑛 ) is obtained for 𝑥 𝑛 = 0 ,

whereas the zero-state response (𝑦𝑧𝑠 𝑛 ) is the system’s
response to an input signal x[n] under stable / static starting
conditions
LCCDE: Zero-input response
• A system’s zero-input response is obtained when the input
to the system is zero (x[n]=0)

• In this case, the system may have an output due to its

internal status (stored energies, starting conditions, ... etc.)

෍ 𝑎𝑘 𝑦𝑧𝑖 [𝑛 − 𝑘] = 0
𝑘=0
LCCDE: Zero-input response
𝑁

෍ 𝑎𝑘 𝑦𝑧𝑖 [𝑛 − 𝑘] = 𝑎0 𝑦𝑧𝑖 𝑛 + 𝑎1 𝑦𝑧𝑖 𝑛 − 1 + ⋯ + 𝑎𝑁 𝑦𝑧𝑖 𝑛 − 𝑁 = 0

𝑘=0

• For this property to be satisfied, 𝑦𝑧𝑖 has to be an exponential in the form

of:

𝑦𝑧𝑖 𝑛 = 𝑐𝛾 𝑛 ⇒ 𝑦𝑧𝑖 𝑛 − 𝑛0 = 𝑐𝛾 𝑛−𝑛0 = 𝛾 −𝑛0 𝑐𝛾 𝑛

⇒ ෍ 𝑎𝑘 𝑦𝑧𝑖 [𝑛 − 𝑘] = 𝑎0 𝑐𝛾 𝑛 + 𝑎1 𝑐𝛾 𝑛−1 + ⋯ + 𝑎𝑁 𝑐𝛾 𝑛−𝑁 = 0

𝑘=0

⇒ 𝑎0 + 𝑎1 𝛾 −1 + ⋯ + 𝑎𝑁 𝛾 −𝑁 = 0
LCCDE: Zero-input response
• This equation is an N. degree polinome, and hence has N number of
roots

𝑄 𝛾 = 𝛾 − 𝛾1 𝛾 − 𝛾2 … 𝛾 − 𝛾𝑁 = 0

• 𝑄 𝛾 is the characteristic equation of the system, 𝛾1 , ..., 𝛾𝑁 are the

roots (eigenvalues)

• Zero-input response consists of the linear combination of all the roots:

𝑦𝑧𝑖 𝑛 = 𝑐1 𝛾1𝑛 + 𝑐2 𝛾2𝑛 + ⋯ + 𝑐𝑁 𝛾𝑁𝑛

LCCDE: Zero-input response
• If, on the other hand, there are multiple instances of the same root,
then the characteristic equation and the zero-input response of the
system become, respectively:

𝑄 𝛾 = 𝛾 − 𝛾1 𝑟 𝛾 − 𝛾𝑟+1 𝛾 − 𝛾𝑟+2 … 𝛾 − 𝛾𝑁 = 0

𝑦𝑧𝑖 𝑛 = 𝑐1 + 𝑐2 𝑛 + 𝑐3 𝑛2 + ⋯ + 𝑐𝑟 𝑛𝑟−1 𝛾1𝑛 + 𝑐𝑟+1 𝛾𝑟+1

𝑛
+ ⋯ + 𝑐𝑁 𝛾𝑁𝑛

• Note: If the starting conditions were stable / static, the zero-input

response would be zero, i.e. 𝑦𝑧𝑖 𝑛 = 0
LCCDE: Zero-input response
• Example: Find the zero-input response for the system defined by the
difference equation 𝑦 𝑛 − 5𝑦 𝑛 − 1 + 6𝑦 𝑛 − 2 = 2𝑥 𝑛 + 𝑥 𝑛 − 1
The starting conditions are provided as 𝑦 −1 = 1 and 𝑦 −2 = 0

• We obtain the characteristic equation as:

1 − 5𝛾 −1 + 6𝛾 −2 = 0

⇒ 1 − 3𝛾 −1 1 − 2𝛾 −1 = 0

• The zero-input response is hence obtained as:

⇒ 𝑦𝑧𝑖 𝑛 = 𝑐1 3𝑛 + 𝑐2 2𝑛
LCCDE: Zero-input response
• The starting conditions 𝑦 −1 = 1 and 𝑦 −2 = 0 are caused by the
zero-input response, therefore:

𝑦𝑧𝑖 −1 = 𝑐1 3−1 + 𝑐2 2−1 = 1

𝑦𝑧𝑖 −2 = 𝑐1 3−2 + 𝑐2 2−2 = 0

⇒ 𝑐1 = 9 , 𝑐2 = −4

⇒ 𝑦𝑧𝑖 𝑛 = 9 3𝑛 − 4 2𝑛 = 3𝑛+2 − 2𝑛+2

LCCDE: Zero-input response
• Example: Find the zero-input response for the system defined by the
difference equation 𝑦 𝑛 − 3𝑦 𝑛 − 1 − 4𝑦 𝑛 − 2 = 3𝑥[𝑛] . The
starting conditions are provided as 𝑦 −1 = 5 and 𝑦 −2 = 0

• We obtain the characteristic equation as:

1 − 3𝛾 −1 − 4𝛾 −2 = 0

⇒ 1 + 𝛾 −1 1 − 4𝛾 −1 = 0

• The zero-input response is hence obtained as:

⇒ 𝑦𝑧𝑖 𝑛 = 𝑐1 −1 𝑛 + 𝑐2 4𝑛
LCCDE: Zero-input response
• 𝑦 −1 = 5 and 𝑦 −2 = 0:

𝑦𝑧𝑖 −1 = 𝑐1 −1 −1 + 𝑐2 4−1 = 5

𝑦𝑧𝑖 −2 = 𝑐1 −1 −2 + 𝑐2 4−2 = 0

⇒ 𝑐1 = −1 , 𝑐2 = 16

⇒ 𝑦𝑧𝑖 𝑛 = −1 −1 𝑛 + 16 4𝑛 = −1 𝑛+1 + 4𝑛+2

LCCDE: Zero-input response
• Example: Find the zero-input response for the system defined by the
difference equation 𝑦 𝑛 + 6𝑦 𝑛 − 1 + 9𝑦 𝑛 − 2 = 𝑥 𝑛 − 𝑥[𝑛 − 1].
The starting conditions are provided as 𝑦 −1 = −1/3 and 𝑦 −2 =
− 2/9

• We obtain the characteristic equation as:

1 + 6𝛾 −1 + 9𝛾 −2 = 0

⇒ 1 + 3𝛾 −1 1 + 3𝛾 −1 = 0

• The zero-input response is hence obtained as:

⇒ 𝑦𝑧𝑖 𝑛 = 𝑐1 + 𝑐2 𝑛 −3 𝑛
LCCDE: Zero-input response
• 𝑦 −1 = −1/3 and 𝑦 −2 = −2/9:

𝑦𝑧𝑖 −1 = 𝑐1 − 𝑐2 −3 −1 = −1/3

𝑦𝑧𝑖 −2 = 𝑐1 − 2𝑐2 −3 −2 = −2/9

⇒ 𝑐1 = 4 , 𝑐2 = 3

⇒ 𝑦𝑧𝑖 𝑛 = 4 + 3𝑛 −3 𝑛
LCCDE: Impulse response
• LCCDE representation is:

𝑁 𝑀

෍ 𝑎𝑘 𝑦[𝑛 − 𝑘] = ෍ 𝑏𝑘 𝑥[𝑛 − 𝑘]
𝑘=0 𝑘=0

• Defining the delay operators as 𝐷 𝑛0 𝑥[𝑛] = 𝑥[𝑛 − 𝑛0 ], we denote:

𝑁 𝑀

𝑄 𝐷 −1 = ෍ 𝑎𝑘 𝐷 −𝑘 𝑎𝑛𝑑 𝑃 𝐷 −1 = ෍ 𝑏𝑘 𝐷 −𝑘
𝑘=0 𝑘=0

• which leads to the representation:

𝑄 𝐷 −1 𝑦 𝑛 = 𝑃 𝐷 −1 𝑥[𝑛]
LCCDE: Impulse response
• For a unit impulse input, we have:

𝑄 𝐷 −1 ℎ 𝑛 = 𝑃 𝐷 −1 𝛿[𝑛]

• Note that we investigate the effect of starting conditions on zero-input

response, so that we can assume a stable / static starting condition in
this case. Therefore, h[-1] = h[-2] = h[-3] = ... = 0

𝑃 𝐷 −1
ℎ𝑛 = −1
𝛿[𝑛]
𝑄 𝐷

• 𝑃 𝐷 −1 involves M delays, and 𝑄 𝐷 −1 involves N delays, which means

that after M-N delays of the impulse response, the impulse input’s
effect is finished. (But the characteristic modes remain)
LCCDE: Impulse response
• As a result, the impulse response is of the form:

𝑀−𝑁

ℎ 𝑛 = ෍ 𝐴𝑘 𝛿[𝑛 − 𝑘] + 𝑦𝑑 𝑛 𝑢[𝑛]
𝑘=0

• Note that 𝑦𝑑 𝑛 is not equal to 𝑦𝑧𝑖 𝑛 !

LCCDE: Impulse response
• Depending on M and N values, this representation becomes:

ℎ 𝑛 = 𝑦𝑑 𝑛 𝑢 𝑛 , 𝑀<𝑁

ℎ 𝑛 = 𝐴0 𝛿 𝑛 + 𝑦𝑑 𝑛 𝑢 𝑛 , 𝑀=𝑁

𝑀−𝑁

ℎ 𝑛 = ෍ 𝐴𝑘 𝛿[𝑛 − 𝑘] + 𝑦𝑑 𝑛 𝑢 𝑛 , 𝑀>𝑁
𝑘=0

• Let’s see how to find the impulse reponse through some examples:
LCCDE: Impulse response
• Example: Find the impulse response for the system defined by:

𝑦 𝑛 − 5𝑦 𝑛 − 1 + 6𝑦 𝑛 − 2 = 2𝑥 𝑛 − 𝑥[𝑛 − 1]

The characteristic equation is:

1 − 5𝛾 −1 + 6𝛾 −2 = 1 − 3𝛾 −1 1 − 2𝛾 −1 = 0

𝑦𝑑 𝑛 = 𝑐1 3𝑛 + 𝑐2 2𝑛
LCCDE: Impulse response
𝑦 𝑛 − 5𝑦 𝑛 − 1 + 6𝑦 𝑛 − 2 = 2𝑥 𝑛 − 𝑥[𝑛 − 1]

Note that the equation involves two delays on the left side and one on the
right side, which means N = 3 and M = 2

As M < N, we have:

ℎ 𝑛 = 𝑦𝑑 𝑛 𝑢 𝑛 = 𝑐1 3𝑛 + 𝑐2 2𝑛 𝑢[𝑛]
LCCDE: Impulse response
Trying y 𝑛 − 5𝑦 𝑛 − 1 + 6𝑦 𝑛 − 2 = 2𝑥 𝑛 − 𝑥[𝑛 − 1] for some
values of 𝑛 with the impulse input:

𝑛 = 0 ⇒ ℎ 0 − 5ℎ −1 + 6ℎ −2 = 2𝛿 0 − 𝛿 −1

⇒ ℎ 0 −0+0=2−0 ⇒ ℎ 0 =2

𝑛 = 1 ⇒ ℎ 1 − 5ℎ 0 + 6ℎ −1 = 2𝛿 1 − 𝛿 0

⇒ ℎ 1 − 10 + 0 = 0 − 1 ⇒ ℎ 1 = 9

𝑛 = 2 ⇒ ℎ 2 − 5ℎ 1 + 6ℎ 0 = 2𝛿 2 − 𝛿 1

⇒ ℎ 2 − 45 + 12 = 0 − 00 ⇒ ℎ 2 = 33
LCCDE: Impulse response
ℎ 𝑛 = 𝑦𝑑 𝑛 𝑢 𝑛 = 𝑐1 3𝑛 + 𝑐2 2𝑛 𝑢[𝑛]

ℎ 0 = 𝑐1 30 + 𝑐2 20 𝑢 𝑛 = 2

⇒ 𝑐1 + 𝑐2 = 2

ℎ 1 = 𝑐1 31 + 𝑐2 21 𝑢 𝑛 = 9

⇒ 3𝑐1 + 2𝑐2 = 9

⇒ 𝑐1 = 5 , 𝑐2 = −3

⇒ℎ 𝑛 = 5 3𝑛 + −3 2𝑛 𝑢[𝑛]
LCCDE: Impulse response
• Example: Find the impulse response for the system defined by:

𝑦 𝑛 − 0.8𝑦 𝑛 − 1 + 0.12𝑦 𝑛 − 2 = 0.3𝑥 𝑛 − 0.4𝑥 𝑛 − 1 + 0.2𝑥 𝑛 − 2

The characterisctic equation is:

1 − 0.8𝛾 −1 + 0.12𝛾 −2 = 1 − 0.6𝛾 −1 1 − 0.2𝛾 −1 = 0

𝑦𝑑 𝑛 = 𝑐1 0.6𝑛 + 𝑐2 0.2𝑛
LCCDE: Impulse response
𝑦 𝑛 − 0.8𝑦 𝑛 − 1 + 0.12𝑦 𝑛 − 2 = 0.3𝑥 𝑛 + 0.4𝑥 𝑛 − 1 + 0.2𝑥 𝑛 − 2

Note that the equation involves two delays on the left side and two on the
right side, which means N = 3 and M = 3

As M = N, we have:

ℎ 𝑛 = 𝐴0 𝛿 𝑛 + 𝑦𝑑 𝑛 𝑢 𝑛 = 𝐴0 𝛿 𝑛 + 𝑐1 0.6𝑛 + 𝑐2 0.2𝑛 𝑢[𝑛]

LCCDE: Impulse response
Trying for some values of 𝑛 with the impulse input in 𝑦 𝑛 − 0.8𝑦 𝑛 − 1 +
0.12𝑦 𝑛 − 2 = 0.3𝑥 𝑛 + 0.4𝑥 𝑛 − 1 + 0.2𝑥 𝑛 − 2 :

𝑛 = 0 ⇒ ℎ 0 − 0.8ℎ −1 + 0.12ℎ −2 = 0.3𝛿 0 + 0.4𝛿 −1 + 0.2𝛿 −2

⇒ ℎ 0 − 0 + 0 = 0.3 + 0 + 0 ⇒ ℎ 0 = 0.3

𝑛 = 1 ⇒ ℎ 1 − 0.8ℎ 0 + 0.12ℎ −1 = 0.3𝛿 1 + 0.4𝛿 0 + 0.2𝛿 −1

⇒ ℎ 1 − 0.8 0.3 + 0 = 0 + 0.4 + 0 ⇒ ℎ 1 = 0.64

𝑛 = 2 ⇒ ℎ 2 − 0.8ℎ 1 + 0.12ℎ 0 = 0.3𝛿 2 + 0.4𝛿 1 + 0.2𝛿 0

⇒ ℎ 2 − 0.8 0.64 + 0.12 0.3 = 0 + 0 + 0.2 ⇒ ℎ 2 = 0.676

LCCDE: Impulse response
ℎ 𝑛 = 𝐴0 𝛿 𝑛 + 𝑐1 0.6𝑛 + 𝑐2 0.2𝑛 𝑢[𝑛]

ℎ 0 = 𝐴0 𝛿 0 + 𝑐1 0.60 + 𝑐2 0.20 𝑢[𝑛] = 0.3

⇒ 𝐴0 + 𝑐1 + 𝑐2 = 0.3

ℎ 1 = 𝐴0 𝛿 1 + 𝑐1 0.61 + 𝑐2 0.21 𝑢[𝑛] = 0.64

⇒ 0.6𝑐1 + 0.2𝑐2 = 0.64

ℎ 2 = 𝐴0 𝛿 2 + 𝑐1 0.62 + 𝑐2 0.22 𝑢[𝑛] = 0.676

⇒ 0.36𝑐1 + 0.04𝑐2 = 0.676

LCCDE: Impulse response
⇒ 𝐴0 = 1.67, 𝑐1 = 2.28 , 𝑐2 = −3.65

ℎ 𝑛 = 𝐴0 𝛿 𝑛 + 𝑐1 0.6𝑛 + 𝑐2 0.2𝑛 𝑢[𝑛]

⇒ ℎ 𝑛 = 1.67𝛿 𝑛 + 2.28 0.6𝑛 + −3.65 0.2𝑛 𝑢[𝑛]

LCCDE: Zero-state response
• Zero-state response is defined as the system’s response to an input signal
x[n] under stable / static starting conditions, and is obtained by:

𝑦𝑧𝑠 𝑛 = 𝑥 𝑛 ∗ ℎ[𝑛]

• The total response from a system is obtained as the summation of zero-state

and zero-input responses:

𝑦 𝑛 = 𝑦𝑧𝑖 𝑛 + 𝑦𝑧𝑠 𝑛

• Note that we prefer LTI systems overall, in which case zero-input response
will be zero, and the system’s total output is equal to the zero-state response