Chapter 2

Quantum Computation

• Qubit: A quantum system which can be in two states, say |0! and |1!.
Therefore, the dimension of the Hilbert space H associated with the system
would be two, i.e. dim(H) = 2, and we denote this Hilbert space as H(2) .

– Any arbitrary state |ψ! ∈ H(2) , can be represented in the basis

{|0!, |1!} as
|ψ! = α|0! + β|1!
where |α|2 + |β|2 = 1 (normalization condition).
– The above normalization condition will always be satisfied if we do
the following parametrization

α = cos θ! and β = eiφ sin θ! .

– So in general,
|ψ! = cos θ! |0! + eiφ sin θ! |1!.
– If θ! = 0, |ψ! = |0!, and θ! = π/2, eiφ |1! where eiφ is a global phase
at θ! = π/2 which has no physical meaning.
– If we rescale θ! = θ/2, then |ψ! = |1! (with a global phase) at θ = π.
Hence, let us now express:
! " ! "
θ θ
|ψ! = cos |0! + eiφ sin |1!
2 2
where 0 ≤ θ ≤ π and say 0 ≤ φ ≤ 2π.
– So now instead of two complex numbers α and β, we have represented
any arbitrary state of a qubit by two real parameters (θ, φ).
– Define: x = sin θ cos φ, y = sin θ sin φ, and z = cos θ. Then the state
|ψ! will be #
1+z x + iy
|ψ! = |0! + $ |1!.
2 2(1 + z)

23
24 CHAPTER 2. QUANTUM COMPUTATION

Z
|0>
|ψ>
θ Y

X |1>
Bloch Sphere

From the normalization condition of |ψ!, we have x2 + y 2 + z 2 = 1

which is an equation of sphere with unit radius. Any point on the
surface of this sphere represents a state of a qubit. This sphere is
called “Bloch sphere”.
– Any unitary transformation of a state |ψ ! ! = U |ψ! can be considered
as a motion of a point on the surface of Bloch sphere.
– Some useful 1-qubit unitary operations or quantum gates :
(1) Pauli matrices :
! " ! " ! "
0 1 0 i 1 0
σx = ; σy = ; and .
1 0 −i 0 0 −1

These matrices are represented in the orthonormal basis {|0!, |1! }

∗ Check the following: (i) operations of σx and σy on {|0!, |1!};
(ii) (σx , σy , σz ) are unitary as well as Hermitian.
(2) Hadamard gate H:
! "
1 1 1 1
H= √ = √ (σx + σz ) .
2 1 −1 2

∗ H|0! = √1 (|0! + |1!)

2
∗ H|1! = √1 (|0! − |1!)
2

(3) Phase shift gate:

! "
1 0
P (α) =
0 e−iα
! "! " ! "
1 0 cos θ2 cos 2θ
P (α)|ψ! = =
0 eiα e sin θ2
iφ
e i(α+φ)
sin θ2
 25

! "
iα/2 e−iα/2 0
P (α) = e
0 eiα/2
We can neglect the global phase term eiα/2 , then the Phase shift can
be written as
%α& ! "
(
P (α) = I cos − i σz sin 2
2 α

(4) Rotation about x, y-axes:

! " ! " ! " ! "
θ θ θ θ
Rx (θ) = I cos −i σx sin and Ry (θ) = I cos −i σy sin
2 2 2 2

(5) Rotation about any arbitrary direction n

':
! " ! "
θ θ
Rn! (θ) = I cos n·−
− i (' →σ ) sin
2 2

' = (nx , ny , nz ) with n2x + n2y + n2z = 1.

where n

^
n
θ
Y

(6) Flip operator:

F = |0!(1| + |1!(0|
This operator flips the state from |0! to |1! and vice versa, i.e.,

F |0! = |1!; F |1! = |0!.

This operator is equivalent to classical N OT gate.

• Multi-qubits (Composite) system/Quantum register: A classical

bit can be in state 0 (low volt state) or 1 (high volt state). Classical
register is a collection of many bits (It can be 8 bits, 16 bits, 32 bits, 64
bits, etc.). A 2-bits register can be in one of the following four states 00
or 01 or 10 or 11. Similarly, a 3-bits register can be in any of the possible
23 = 8 states.
26 CHAPTER 2. QUANTUM COMPUTATION

Quantum systems have one distinct non-classical property, and that is, a
quantum system can be in the superposition of all possible states. For
example, general state of a qubit is the superposition of |0! and |1! states,
i.e., |ψ! = α|0! + β|1!. Mathematically, this state represents a vector in a
two dimensional Hilbert space. Similarly, a 2-qubits quantum register can
be in the following superposition state:

|ψ! = α00 |00! + α01 |01! + α10 |10! + α11 |11!.

Here the corresponding Hilbert space dimension is 22 = 4. Naturally,

3-qubits quantum register is represented in a Hilbert space of dimension
23 = 8, and 2n dimensional Hilbert space is required to represent a n-
qubits quantum register. Here, we observe that Hilbert space dimension
increases exponentially with the number of qubits.

• Bases for multi-qubits system: Single qubit basis states {|0!, |1!} are
represented by 2-dimensional column matrices or column vectors. For
example, ! " ! "
1 0
|0! = and |1! = .
0 1

We expect that we need 22 = 4 dimensional column vectors to represent 4

basis states. We also expect that these 4 basis states should be constructed
from the single-qubit basis states. Therefore, we have to construct 4-
dimensional column vectors from the single-qubit 2-dimensional column
vectors.
We have come across only two different kind of product between any two
given vectors, one is inner product and the other one is outer product.
For example, the inner product of |0! and |0! is
! "
( ) 1
(0|0! = 1 0 = 1,
0

which is a scalar (number). The outer product of these two same vector
is: ! " ! "
1 ( ) 1 0
|0!(0| = 1 0 = .
0 0 0
This is a 2 × 2 matrix. This clearly indicates that we need to develop
or define a new kind of product which increases the dimension of the
column vectors. Let us denote that “new product” by the symbol ⊗, i.e.,
|0! ⊗ |0! = |00! or
 
! " ! " •
1 1  • 
⊗ −→  
 .
0 0 • 
•
 27

We now define this new kind of product which will faithfully satisfy all
the properties of a basis for a 4-dimensional Hilbert space.

! "   ! " 
1 1
! " ! "  1   
 0   0 
1 1
⊗ ≡ =
 ! "   !

" .
0 0    
1 0
0
0 0

We now remove those two brackets inside the big bracket, and write down
the above expression in a more nice way as:

 
! " ! " 1
1 1  0 
⊗ = 
 0 .
0 0
0

Here we have basically done the following: we have multiplied the first
element of the first vector in the left hand side with the second vector
and have placed the resultant vector as the first two elements of the 4-
dimensional vector. Then we have multiplied the second element of the
first vector with the second vector and have placed the resultant vector as
the next two elements of the 4-dimensional vector. This kind of product,
denoted by ⊗, is called ‘Tensor product’ or ‘Krönecker product’. In
general, tensor product of any two arbitrary 2-dimensional vectors will be:

 
! " ! " α1 α2
α1 α2  α1 β2 
⊗ = 
 β1 α2  .
β1 β2
β1 β2

Similarly, for 3-qubits case, we have

 
α1 α2 α3
 α1 α2 β3 
   
α α  α1 β2 α3 
! " ! " ! " ! " 2 3  
α1 α2 α3 α1  α2 β3   α1 β2 β3 
⊗ ⊗ = ⊗ =
 β2 α3  
.

β1 β2 β3 β1  β1 α2 α3 
β2 β3  β1 α2 β3 
 
 β1 β2 α3 
β1 β2 β3
28 CHAPTER 2. QUANTUM COMPUTATION

In Dirac’s notation:
! " ! " ! "
α1 α2 α3 ( ) ( )
⊗ ⊗ = α1 |0!1 + β1 |1!1 ⊗ α2 |0!2 + β2 |1!2
β1 β2 β3 ( )
⊗ α3 |0!3 + β3 |1!3

= α1 α2 |0!1 ⊗ |0!2 + α1 β2 |0!1 ⊗ |1!2

+β1 α2 |1!1 ⊗ |0!2 + β1 β2 |1!1 ⊗ |1!2

= α1 α2 |00! + α1 β2 |01! + β1 α2 |10!

+β1 β2 |11!.

For the first two equalities, we have used subscripts 1 and 2 to identify
the qubits. In the last step we have dropped those subscripts, and now
the location of the single qubit state inside the multi-qubits state identify
the qubit’s numbers. For example, |00! state suggests that the first qubit
is in |0! state and the second qubit is in |1! state; |01! state suggests that
the first qubit is in |0! state and the second qubit is in |1!; and so on. For
3-qubits case,

|ψ! = α1 α2 α3 |000! + α1 α2 β3 |001! + α1 β2 α3 |010! + . . . . . .

Therefore, for n qubits system, where each qubit is represented in its re-
spective Hilbert spaces Hi then the Hilbert space of the composite system
H will be
H = H1 ⊗ H2 ⊗ H3 ⊗ · · · ⊗ Hn = ⊗ni=1 Hi .

• Operators for multi-qubit systems: States of multi-qubits or com-

posite systems have been introduced above. We now define operators for
multi-qubits systems. Consider the simplest composite system, i.e., 2-
(2) (2)
qubits system. Here H = H1 ⊗ H2 here subscripts represent different
qubits and the superscripts represent the dimensions of individual subsys-
tems. Consider an operator A ∈ H. The simplest form of this operator
can be A = A1 ⊗ A2 . This operator A operates on a bipartite state
|ψ! = |ψ1 ! ⊗ |ψ2 ! as following way:

A|ψ! = (A1 ⊗ A2 )(|ψ1 ! ⊗ |ψ2 !

.
= A1 |ψ1 ! ⊗ A2 |ψ2 !

If we consider a bipartite operator A which is a sum of two bipartite

operators, say A = A1 ⊗ A2 + A!1 ⊗ A!2 . Then,

A|ψ! = A1 |ψ1 ! ⊗ A2 |ψ2 ! + A!1 |ψ1 ! ⊗ A!2 |ψ2 !.

We can more generalize and consider the operator

n 0
0 m
(i) (j)
A= γij A1 ⊗ A2 .
i=1 j=1
 29

Then,
n 0
0 m
(i) (j)
A|ψ! = γij A1 |ψ1 ! ⊗ A2 |ψ2 !.
i=1 j=1

One can then further generalize for n-qubits case where A = A1 ⊗ A2 ⊗

· · · ⊗ An and the state |ψ! = |ψ1 ! ⊗ |ψ2 ! ⊗ · · · ⊗ |ψn !, then

A|ψ! = A1 |ψ1 ! ⊗ A2 |ψ2 ! ⊗ · · · ⊗ An |ψn !.

• Two qubits system or bipartite system: We now discuss more detail

about two qubits system which is the simplest composite system. How-
ever, this very simple system shows many properties which are completely
against our intuition. Consider state of two qubits are |ψ1 ! = α1 |0!+β1 |1!
and |ψ2 ! = α2 |0! + β2 |1!. Then the joint state of these two qubits system
will be
|ψ!12 = |ψ1 ! ⊗ |ψ2 !
= (α1 |0! + β1 |1!) ⊗ (α2 |0! + β2 |1!)
= α1 α2 |00! + α1 β2 |01! + α2 β1 |10! + β1 β2 |11!.

If one asks what is the state of the first qubit, then the answer will be
|ψ1 !. On the other hand, if one asks identical question for the second
qubit, then the answer will be |ψ2 !. Let us now consider the following
state:
|ψ! = α|01! + β|10!.
Now if someone asks the same above questions then we cannot identify the
state of the individual qubits. Now two qubits are correlated with each
other, and this state is called entangled state.
• Entanglement: If the state of a bipartite system cannot be expressed as
a tensor product of the states of individual qubits, i.e., |ψ! ,= |ψ1 ! ⊗ |ψ2 !,
then this correlated state is called entangled state. Entanglement is a
pure quantum correlation which has no classical analogue. Following are
examples of two qubits entangled states:

|ψ + ! = √1 (|01! + |10!)
2
|ψ − ! = √1 (|01! − |10!)
2
|φ+ ! = √1 (|00! + |11!)
2
|φ− ! = √1 (|00! − |11!).
2

The above states are called Bell states or EPR states. One can check that
{|ψ ± !, |φ± !} form orthonormal basis for two qubits system. Therefore,
these states can be considered as eigenstates of an observable. As a con-
sequence, instead of the computational basis states {|00!, |01!, |10!, |11!},
one can write any arbitrary two qubits state as

|ψ! = c1 |ψ + ! + c2 |ψ − ! + c3 |φ+ ! + c4 |φ− !.

30 CHAPTER 2. QUANTUM COMPUTATION

This implies that even a two qubits unentangled or product state |ψ! =
|ψ1 !⊗|ψ2 ! can also be written as a superposition of these entangled states.
Therefore, a linear superposition of entangled states can be an unentangled
state.
• Quantum cloning: Cloning or copying is a process by which we want to
create a copy of a given arbitrary state |ψ!. If |ψ! is a qubit state then a
quantum cloning operation will act on this state and create copy of this
state. Suppose we have |0! or |1! state and have a quantum cloner Uc
where Uc is a unitary operator. Initially we have 2 qubits, one is in |0! or
|1! state and the other one in some blank state |blank!. The operator Uc
operates on this two qubits state as a following way:

Uc |0! ⊗ |blank! = |0! ⊗ |0!

Uc |1! ⊗ |blank! = |1! ⊗ |1!.

The cloning operator can clone |0! and |1! states. We now want to make a
clone of an arbitrary state |ψ! = α|0!+β|1!, i.e., we want Uc |ψ!⊗|blank! =
|ψ! ⊗ |ψ!. However, it is not possible to make a clone of an arbitrary state.
This can be shown very easily in following way:

Uc |ψ! ⊗ |blank!

= Uc (α|0! + β|1!) ⊗ |blank!

( ) ( )
= α Uc |0! ⊗ |blank! + β Uc |1! ⊗ |blank!

= α|0! ⊗ |0! + β|1! ⊗ |1!

( ) ( )
,= α|0! + β|1! ⊗ α|0! + β|1! .

Even though, Uc can individually make a clone of |0! and |1!, but any
general state cannot be cloned. This is known as ‘no-cloning theorem’.
• Quantum teleportation:

– Task: Alice and Bob are far away apart. Alice has a qubit C which is
in a general state (may be unknown to Alice) |ψ!C = α|0!C + β|1!C .
She wants to send this state to Bob without physically sending any
qubit.
– Protocol/Process: Alice and Bob are sharing two qubits, A and
B. These two qubits are in an entangled state, say
1
|φ+ !AB = √ (|0!A ⊗ |0!B + |1!A ⊗ |1!B ) .
2
Alice has now two qubits A and C, where A is in an entangled state
with Bob’s qubit B and C is the qubit whose state Alice wants to
 31

teleport to Bob. Therefore, the joint state of the 3-qubits system is:
|Ψ! = |φ+ !AB ⊗ |ψ!C

= √1 (|0!A ⊗ |0!B + |1!A ⊗ |1!B ) ⊗ (α|0!C + β|1!C )

2

= √1 (α|0!A ⊗ |0!B ⊗ |0!C + β|0!A ⊗ |0!B ⊗ |1!C

2

+ α|1!A ⊗ |1!B ⊗ |0!C + β|1!A ⊗ |1!B ⊗ |1!C )

Instead of computational basis, Alice’s qubits A and C can be ex-
pressed in Bell basis. Then the above 3-qubits joint state can be
written as
1
|Ψ! = 21 |φ+ !AC ⊗ (α|0!B + β|1!B ) + |φ− !AC ⊗ (α|0!B − β|1!B )
2
+ |ψ + !AC ⊗ (β|0!B + α|1!B ) + |ψ − !AC ⊗ (β|0!B − α|1!B ) .
If Alice now measures her two qubits in Bell basis, she will find one
of the Bell states and the joint state |Ψ! will collapse in a following
way:
|φ+ ! then : |Ψ! −→ |φ+ !AC ⊗ (α |0!B + β |1!B )

|φ− ! then : |Ψ! −→ |φ− !AC ⊗ (α |0!B − β |1!B )

|ψ + ! then : |Ψ! −→ |ψ + !AC ⊗ (β |0!B + α |1!B )

|ψ − ! then : |Ψ! −→ |ψ − !AC ⊗ (β |0!B − α |1!B )

Note that, after the measurement, Alice’s two qubits have become en-
tangled state, whereas Bob’s qubit is not anymore in entanglement
with Alice’s qubit A. This implies that the loss of entanglement be-
tween the qubits A and B leads to the creation entanglement between
Alice’s two qubits A and C. This property is known as ‘Monogamy
of entanglement’.
The next step of the protocol is the following: Alice will inform
Bob about her measurement outcomes by using any classical channel
(phone call, text message, etc.). Based upon Alice’s measurement
outcome, Bob has to apply following definite unitary operations on
his qubit B:
For : |φ+ !AC : Nothing

|φ− !AC : σz

|ψ + !AC : σx

|ψ − !AC : −σz σx
32 CHAPTER 2. QUANTUM COMPUTATION

After this step, the state of Bob’s qubit will be |ψ! = α |0! + β |1!.
Also note that, in this protocol, the original state of the qubit C is
now destroyed. So it is no more in |ψ! = α |0! + β |1! state, but its
copy is being created at somewhere else (in qubit B). Therefore, the
teleportation protocol does not violate no-cloning principle, because
this protocol does not suggest

|ψ!C ⊗ |initial!B −→ |ψ!C ⊗ |ψ!B ,

rather it is suggesting

|ψ!C ⊗ |initial!B −→ |?!C ⊗ |ψ!B .

The other important points of this protocol are:

(1) There is no transfer of particle physically.
(2) Alice has to inform Bob about her measurement outcomes by
any classical channel. In this process Alice has to transfer 2 bits
of information to Bob classically. Additionally, this also implies no
superluminal transfer of information.
(3) If someone intercept the classical channel used by Alice and Bob
and knows the measurement outcomes, but still he/she cannot dis-
turbs the protocol. Because that person has no access to any of the
qubits.
(4) This protocol has two parts, local operations on their respective
qubits by Alice and Bob, and then use of classical channel. Therefore,
this process is called LOCC, i.e. Local Operation(s) and Classical
Communication.

• Quantum super-dense coding:

– Task: Here also two persons are involved, say Alice and Bob who are
far away apart. The goal of this protocol is to transmit two bits of
information from one person to another person by physically sending
only one qubit.
– Protocol: Alice and Bob are sharing two qubits which are in an
entangled state, say
3 +4
3ψ = √1 (|0! ⊗ |1! + |1! ⊗ |0! ) .
A B A B
2
Alice and Bob were in agreement that the measurement will per-
formed in Bell basis {|φ± ! , |ψ ± !}, i.e., they will only measure the
observable whose eigenstates are {|φ± ! , |ψ ± !}, and they assign
3 4 3 +4 3 4 3 −4 3 4 3 +4 3 4 3 4
300 = 3ψ ; 301 = 3ψ ; 310 = 3φ ; and 311 = 3φ− .

(1) If Alice wants to send the bit string 00, then she will not do
anything with her qubit and send that to Bob.
 33

(2) If she wants to send 01, she will apply σz on her qubit and then
send the qubit to Bob.
(3) If she wants to send 10, she will apply σx on her qubit and then
send the qubit to Bob.
(4) If she wants to send 11, she will apply σx and followed by σz on
her qubit and then send the qubit to Bob.

Therefore, by physically sending one qubit, Alice can convey 2 bits

of information to Bob.