Received 8 January 2024, accepted 31 January 2024, date of publication 6 February 2024, date of current version 15 February 2024.

Digital Object Identifier 10.1109/ACCESS.2024.3362955

Quantum Computing: Circuits, Algorithms,

and Applications
MUHAMMAD ALI SHAFIQUE1 , ARSLAN MUNIR 2, (Senior Member, IEEE), AND IMRAN LATIF3
1 Department of Electrical and Computer Engineering, Kansas State University, Manhattan, KS 66506, USA
2 Department of Computer Science, Kansas State University, Manhattan, KS 66506, USA
3 Brookhaven National Laboratory, U.S. Department of Energy, Upton, NY 11973, USA

Corresponding author: Arslan Munir ([email protected])

Muhammad Ali Shafique is also affiliated with the University of Engineering and Technology (UET), Lahore.

ABSTRACT Quantum computing, a transformative field that emerged from quantum mechanics and
computer science, has gained immense attention for its potential to revolutionize computation. This paper
aims to address the fundamentals of quantum computing and provide a comprehensive guide for both novices
and experts in the field of quantum computing. Beginning with the foundational principles of quantum
computing, we introduce readers to the fundamental concepts of qubits, superposition, entanglement,
interference, and noise. We explore quantum hardware, quantum gates, and basic quantum circuits. This
study offers insight into the current phase of quantum computing, including the noisy intermediate-
scale quantum (NISQ) era and its potential for solving real-world problems. Furthermore, we discuss the
development of quantum algorithms and their applications, with a focus on famous algorithms like Shor’s
algorithm and Grover’s algorithm. We also touch upon quantum computing’s impact on various industries,
such as cryptography, optimization, machine learning, and material science. By the end of this paper, readers
will have a solid understanding of quantum computing’s principles, applications, and the steps involved in
developing quantum circuits. Our goal is to provide a valuable resource for those eager to embark on their
quantum computing journey and for researchers looking to stay updated on this rapidly evolving field.

INDEX TERMS Quantum computing, entanglement, interference, quantum circuits, quantum algorithms,
quantum applications.

I. INTRODUCTION rely on the difficulty of factoring large numbers by classical

Quantum computing technology uses different approaches computers. A distinct difference in computational power
to solve certain computational problems, demonstrating of classical versus quantum computers can be illustrated
greater efficiency compared to classical computing systems. with the evaluation of the time required to crack encryption
Recent experimental outcomes are remarkable, hinting at the schemes like Rivest-Shamir-Adleman (RSA) that rely on the
possibility of quantum computers becoming commercially difficulty of finding prime factors of large numbers [7]. While
available in the near future [1], [2], [3], [4]. A prominent traditional computers would require billions of years for such
example of quantum computing’s ability lies in Shor’s a task, quantum computers could potentially solve it in a short
algorithm, renowned for its capability to factor large numbers time [8], [9].
efficiently [5]. This algorithm in 1994 marked a pivotal step Quantum computers share some components with classical
in the advancement of quantum computing by enabling the computers, such as registers, gates, and memory elements.
determination of prime factors of large numbers by quantum However, their underlying physical structures are funda-
computers [6], and has also caused consternation in cryp- mentally distinct and unique. Quantum computations unfold
tography field as many public key cryptography algorithms within quantum registers, where qubits can exist in the state of
superposition and entanglement. These unique characteristics
The associate editor coordinating the review of this manuscript and make quantum computers fundamentally different from
approving it for publication was Siddhartha Bhattacharyya . traditional classical computers.
2024 The Authors. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
For more information, see https://creativecommons.org/licenses/by-nc-nd/4.0/
22296 VOLUME 12, 2024
M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

Another distinct difference of quantum computing versus with their environment and lose their quantum properties
classical computing is computational units such as bits. Bits (superposition, entanglement, and, interference) over time
in classical computing are restricted to zero or one whereas affecting the outcome of quantum circuits. A controlled
quantum computing employ units (qubits) that are capable of quantum environment has led to debates about achieving
existing in states of zero, one, or any intermediate value [10], reliable quantum computers [16]. It also suggests that
[11], [12]. This unique attribute grants quantum computers quantum computers can outperform classical if certain
the remarkable ability to simultaneously follow multiple conditions are met [17].
computational paths within a single calculation, which is not While early noisy quantum computers have been used to
possible by classical computers without repeated iterations. implement algorithms such as Shor’s, Grover’s, and Deutsch–
Jozsa’s, the prevailing high error rates and noise prevent the
A. QUANTUM COMPUTING INTRODUCTION AND HISTORY scaling of these algorithms [18], [19], [20], [21]. In order
to achieve fault-tolerant quantum computation, substantial
Quantum computing, in contrast to classical computing,
improvements are required in quantum computers to control
is a relatively recent development. Its origins can be traced
and protect the qubits sufficiently for reliable algorithms.
back to the late 1970s when it initially appeared in science
These improvements can be made with hardware modifica-
fiction, subsequently attracting significant attention from the
tions or the use of error-correcting codes. Shor introduced
media. It was in 1981 that Richard Feynman is credited with
Quantum Error Correction (QEC) in 1995, showing that
pioneering the concept of a quantum computer. He proposed
information from one logical qubit can be encoded onto
the idea that quantum computers could efficiently simulate
multiple physical qubits, protecting it from errors [22]. Shor’s
quantum systems that could avoid the exponential resource
work demonstrated the possibility of executing quantum
requirements for classical computers. Classical computers
computations reliably with noisy quantum hardware [23].
encounter substantial difficulties when attempting to simulate
Further research has revealed that noise and quantum
quantum systems. Feynman, along with visionaries like Yuri
scaling relate to each other. If errors and noise are below
Manin and Paul Benioff, recognized the vast potential of
a certain threshold, it’s theoretically possible to scale up
quantum computers in the realm of complicated computing
problems. In 1985, David Deutsch formalized the concept of quantum computers to larger sizes [24], [25]. Many types of
a quantum computer, marking a significant milestone in the error-correcting techniques have been developed [26], [27],
field of quantum computing. Furthermore, he distinguished [28], but studies indicate that millions of physical qubits are
between quantum simulators and programmable quantum needed to achieve useful quantum computers [29]. Despite
devices. this, various algorithms have claimed quantum supremacy,
In subsequent years, significant achievements were made showcasing computations on quantum devices likely surpass
in the field of quantum computing, revealing its potential classical computers’ capabilities in a reasonable time frame.
to surpass classical counterparts in terms of computational The research works of IBM, Xanadu, and Google’s Quantum
efficiency. It became increasingly clear that quantum com- AI team are prominent accomplishments in the field of
puters could offer solutions for specific computing problems quantum computing [30], [31], [32], [33]. While these
efficiently. Notably, Simon and Shor made remarkable achievements are significant, they have limitations in scaling
contributions by developing algorithms that demonstrated up quantum computations due to noise and errors.
speed enhancements for particular problem sets, including The term NISQ stands for ‘‘Noisy Intermediate-Scale
the field of prime factorization and cryptography. Seth Lloyd Quantum.’’ It refers to a phase in quantum computing
further enriched the supremacy of quantum computers by where quantum computers are not yet completely error-
introducing an algorithm for simulating a wide range of corrected but are large enough to perform computations
quantum systems on quantum computers. beyond classical computers’ capabilities. NISQ devices are
In summary, quantum computing is getting better with time characterized by the presence of errors due to noise, but
and has the potential to solve certain computing problems they are sufficiently reliable for solving certain problems
more efficiently than classical computers. This is shown by more efficiently than classical computers [34]. This phase
various quantum algorithms, which highlight how powerful represents a transitional period in the advancement of
quantum computing can be. quantum computing technology and quantum supremacy.
Our main contributions in this paper are as follows:
• We discuss the fundamentals of quantum computing
B. NOISY INTERMEDIATE-SCALE QUANTUM (NISQ) including qubits, superposition, entanglement, interfer-
From the beginning, there has been a doubt whether a quan- ence and noise.
tum computer could surpass the capabilities of a classical • We elaborate building blocks of quantum circuits such
computer. Many of these doubts originate from concerns as various quantum gates followed by their usage in
about the complexity of quantum computer design and and different quantum circuits.
difficulty of controlling quantum computation devices [13], • We study different quantum algorithms including Shor’s
[14], [15]. These concerns are primarily related to the algorithm and Grover’s algorithm that manifest the
concept of decoherence, where quantum systems interact enormous potential of quantum computing.

VOLUME 12, 2024 22297

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

• We explore the design of hybrid classical-quantum vector representing this wavefunction in Hilbert space
machine learning models. requires an infinite dimensional vector notation. This infinite
• We discuss various quantum simulators and quantum dimensional vector state of the qubit in Hilbert dual space
applications in different fields, such as cryptography, is shown using Dirac’s bra-ket notation, which was created
optimization, chemistry, finance, and energy. by Paul Dirac in 1939 [36]. However, it can also be a finite-
• We discuss the deployment requirements and current dimensional vector having two states, on/off or spin-up/spin-
challenges for quantum computers. down, which can be shown in two-dimensional Hilbert space.
The remainder of this paper is organized in the following In this notation, two-dimensional state vectors |1⟩ (read ket
manner. Section II provides the fundamental concepts of one) and |0⟩ (read ket zero) are used for qubit.
qubits, superposition, entanglement, interference, and noise.  
Section III explores the building blocks of quantum circuits 1
|0⟩ = 1|0⟩ + 0|1⟩ → (1)
such as quantum gates and measurement. In Section IV, vari- 0
 
ous types of quantum circuits are explained with numerical 0
|1⟩ = 0|0⟩ + 1|1⟩ → (2)
examples. Section V discusses quantum algorithms which 1
include famous Shor’s algorithm and Grover’s algorithm. In the equation (1), ket zero shows that the qubit is at an
Section VI elaborates the utilization of quantum computing in off or spin-down state. Here, the first element represents the
machine learning and discusses the design of hybrid classical- probability amplitude of off or spin down, and the second
quantum machine learning models. Popular quantum simula- element shows the probability amplitude of on or spin up.
tors and their features are explored in Section VII. Quantum Probability amplitude can be a complex value and it is used
hardware and its deployment requirements are discussed to compute the probabilities of vector states. Additionally,
in Section VIII. Section IX discusses quantum computing in Dirac’s notation, the bra is a complex conjugate transpose
applications in various fields, such as cryptography, optimiza- of a ket. For example, ⟨φ| (read bra of φ) is a complex
tion, machine learning, finance, and energy sector. Lastly, conjugate transpose vector of ket ψ. The inner product of
Section X concludes the article and highlights future research these two vectors ⟨φ|ψ⟩ is a scalar value [37]. The symbol
directions. ‘‘|⟩’’ denotes a column vector, and is known as a ‘‘ket’’. The
‘‘bra’’ (⟨|) form is a row vector and it is shown below:
II. QUANTUM COMPUTING FUNDAMENTALS
 
In lieu of bits in classical computing, quantum computing ⟨0| = 1⟨0| + 0⟨1| → 1 0 (3)
utilizes qubits, which can exist in multiple states at the same ⟨1| = 0⟨0| + 1⟨1| → 0 1
 
(4)
time—a phenomenon known as superposition. Quantum
entanglement signifies a unique connection between qubits The ket notation is widely used in quantum computing as
and quantum interference can alter the outcome of the qubits. bra-ket representation
√ of the qubit.
√ The following two states
Quantum computers also face a challenge called quantum (|0⟩ + |1⟩)/ 2 and (|0⟩ − |1⟩)/ 2 are also commonly used in
noise, which can lead to loss of quantum properties, such quantum calculations and these are sometimes written as |+⟩
as superposition, entanglement, and interference, and can and |−⟩, respectively.
affect the outcome of a quantum system. This section aims A single qubit is also called a two-level quantum system
to provide a simple yet comprehensive understanding of because it is a linear combination of two state basis, 0 and 1.
quantum computing fundamentals. Below is the common form of a single qubit in bra-ket
notation.
A. QUBITS  
1
   
0 v
In classical computing, a bit is analogous to a binary light |v⟩ = v0 |0⟩ + v1 |1⟩ = v0 + v1 = 0 (5)
0 1 v1
switch, capable of assuming only two discrete states: 0 or 1,
without any intermediary values. In contrast, for quantum Here vo and v1 are complex coefficients to measure
computing, a quantum bit (qubit) operates more like a probability amplitudes. The probability and the phase of each
dimmer switch. It possesses not just the 0 and 1 states but computational state basis for a qubit can be computed as
also the ability to exist in an intermediate state, which is follows:
a linear combination of the 0 and 1 states, weighted by For state basis |0⟩ with complex coefficient vo = x + i ∗ y
specific coefficients. These coefficients are used to calculate probability amplitude = |vo | (6)
the probability of measuring either the 0 or 1 state when p
measured. |vo | = (x + i · y) ∗ (x − i · y)
q
= x 2 + y2 (7)
1) BRA-KET NOTATION
2
Qubit is a quantum computing particle that has a wave-like probability = |vo | (8)
nature with wavefunction ψ(x) that satisfies the Schrödinger y
phase(rad) = tan−1 (9)
equation. Theoretically, this wavefunction exists in an infinite x
dimensional Hilbert dual space [35]. Therefore, the state phase (degree) = phase (rad) ∗ (180/π) (10)

22298 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

The probability amplitude is used to calculate the probability If we have n qubits, we will need to keep track of 2n
of each state basis of the qubit which helps in the complex probability amplitudes. As we can see, these vectors
measurement of the qubit state. Similarly, phase is used for grow exponentially with the number of qubits. This is the
quantifying interference. The concepts of measurement and reason quantum computers with large numbers of qubits are
interference are explained in the following sections of the so difficult to simulate in classical computers. A modern
paper. If the complex coefficients are normalized, then they laptop can easily simulate a general quantum state of around
represent the probability of the qubit for 0 and 1 state 20 qubits, but simulating 100 qubits is too difficult even for
the largest supercomputers.
|vo |2 + |v1 |2 = 1 (11)
This is known as the normalization constraint since all 2) BLOCH SPHERE NOTATION
two-level systems must obey this quality to function as a The Bloch sphere is a mathematical representation of a given
qubit. quantum state of a qubit, with which researchers can pinpoint
For two or multiple qubits, the tensor product (or and manipulate various such states within the sphere to their
Kronecker product) is used to compute the resultant states of advantage. Three qubits |1⟩, |−⟩ and, |γ ⟩ are shown in Bloch
the quantum system. The tensor product is denoted by the sphere representation in Figure 1.
symbol ⊗. Let us consider two qubits |a⟩ and |b⟩ as
   
a0 b B. QUANTUM SUPERPOSITION
|a⟩ = and |b⟩ = 0 (12)
a1 b1 In classical computing, a bit possesses a binary nature,
The tensor product of the two qubits is exclusively adopting either a state of 1 or 0. Correspondingly,
in a 2-bit classical system, only one state can exist at a given
|x⟩ = |a⟩ ⊗ |b⟩ = |ab⟩ (13) time among four distinct states that is 00, 01, 10, and 11.
  
b0     This conceptual framework can be extended to n-bit classical
a ∗ a0 b0 x0
 0 b1  
 systems with 2n states but only one state exists at a given time
a b
  0 1  x1 
   
|x⟩ =  (14) representing the state of the classical system.
   = a1 b0  = x2 
b  Conversely, in quantum computing, a single quantum
a1 ∗ 0

a1 b1 x3
b1 bit (qubit) can exist in the state of 0, 1, or any linear
|x⟩ = a0 b0 |00⟩ + a0 b1 |01⟩ + a1 b0 |10⟩ + a1 b1 |11⟩ (15) combination of these states as shown in Figure 2. This
phenomenon is called superposition which enables qubits to
|x⟩ = x0 |00⟩ + x1 |01⟩ + x2 |10⟩ + x3 |11⟩ (16)
exist in a combination of the states. Upon measurement, the
and the normalization constraint rule for the two qubits will superposition collapses, and the final outcome is determined
be the same as follows: depending on the probability distribution of the qubit states.
Quantum superposition is the ability of a qubit to be in
|a0 b0 |2 + |a0 b1 |2 + |a1 b0 |2 + |a1 b1 |2 = 1 (17) multiple states simultaneously until it is measured.
 
c
Similarly for 3-qubits, if |c⟩ = 0 then the tensor product
c1 C. QUANTUM ENTANGLEMENT
of the three qubits is In classical computers, the state of a bit can vary inde-
|y⟩ = |ab⟩ ⊗ |c⟩ = |abc⟩ (18) pendently, that is, the state of a bit is not influenced by
   the state of another bit. However, in quantum computing,
c0
a b
 0 0 ∗ the probability of a qubit state can be affected by the
c1 
      change of another qubit state probability. This phenomenon is
  a0 bo c0 y0
   called entanglement [38]. In quantum circuits, entanglement
 c0  a0 bo c1  y1 
a0 b1 ∗      is created through quantum gates by performing specific
 c1  a0 b1 c0  y2 
   

  a0 b1 c1  y3  operations on the qubits that result in inseparable states
|y⟩ =    =  =  (19) of qubits as shown in equations (21), (22), (23), and (24).
a1 b0 ∗ c0  a1 bo c0  y4 
     
Regardless of the physical distance between the entangled
 c1  a1 bo c1 
 
 y5 
 

  a1 b1 c0  y6  qubits, a change in one qubit state probability can change the
   probability distribution of all qubits in the entangled quantum
a1 b1 ∗ c0  a1 b1 c1 y7
 
system [39].
 c1 
Quantum entanglement is a phenomenon that occurs when
two or more particles become correlated in such a way that
The same method will be used to combine n qubits, and the state of one qubit is dependent on the state of the other
normalization constraint rules for n-qubits will be given as in qubit, regardless of the distance between them. If the state
equation (20). of one qubit changes in the entangled system, then the states
X of all other qubits will be affected. There are specific states
|vi |2 = 1 (20)
in 2-qubit systems, which are called Bell’s states or EPR

VOLUME 12, 2024 22299

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

FIGURE 1. Bloch sphere representation of three different qubits.

FIGURE 2. Representation of qubit with the state of 0 in (a), 1 in (b), and FIGURE 3. Two qubits in non-entangled (a) and entangled quantum
superposed states in (c). states (b).

(Einstein–Podolsky–Rosen) pairs, which exhibit entangled a quantum wavefunction satisfying the Schrödinger equation
properties and cannot be written in separable states as given as shown in Figure 4. A wavefunction is a mathematical
below: description of the quantum state that consists of complex
1 probability amplitudes, and the corresponding probabilities
|φ⟩ = √ (|0⟩|0⟩ + |1⟩|1⟩) (21) of quantum system states.
2
1
|φ ′ ⟩ = √ (|0⟩|0⟩ − |1⟩|1⟩) (22)
2
1
|φ ′′ ⟩ = √ (|0⟩|1⟩ + |1⟩|0⟩) (23)
2
1
|φ ′′′ ⟩ = √ (|0⟩|1⟩ − |1⟩|0⟩) (24)
2
In the 2-qubit system, as shown in Figure 3a, each of the
qubits is in a superposition state but qubits are not entangled. FIGURE 4. Wave-like nature of a qubit.
Therefore, the probabilities of all superposed qubit states are
independent of each other. When these qubits are entangled When we have multiple qubits, their wavefunctions are
as shown in Figure 3b, then the change in the probability of added together to give an overall wavefunction describing
one qubit affects the probabilities of the entangled qubits. the resultant states of a quantum system. This adding process
The blue color in Figure 3b shows that the two qubits are of wavefunctions is called interference. It is a fundamental
not independent particles. They are entangled and their states phenomenon that arises from the wave-like nature of quantum
are dependent on each other. This entanglement results in the particles, such as electrons or photons and it distinguishes
change of probability distribution of the state of the entangled quantum systems from classical systems.
quantum system, even if the entangled qubits are far away In quantum computing, when two quantum wavefunctions
from each other. overlap, they can interfere with each other constructively or
destructively. This results in a change in the resultant wave-
D. QUANTUM INTERFERENCE function of the quantum system that affects the probability
Qubit is represented with bra-ket notation or Bloch sphere but distribution of its quantum states as shown in Figure 5.
this is just a mathematical representation of the qubit state. Interference can also be a challenge in quantum computing
In reality, the qubit has a wave-like nature that is described by due to the phenomenon of decoherence. Decoherence is

22300 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

III. BUILDING BLOCKS OF QUANTUM CIRCUITS

The two key building blocks of quantum circuits are quantum
gates and measurement. Quantum gates, akin to classical
logic gates, play an important role in quantum circuits. These
gates manipulate qubits facilitating the transformation of
FIGURE 5. Quantum interference in two quantum wavefunctions.
quantum information. This section elaborates various types
of quantum gates, such as Hadamard gate, CNOT gate,
and Pauli gates, which form the foundational components
the loss of quantum coherence, which is the property of a of quantum circuits. Furthermore, this section discusses
quantum state to maintain its superposition and entanglement, the critical process of quantum measurement, where the
due to the interactions with environment and thermalization. superposition of qubits collapses into definite states.
This loss of coherence leads to a breakdown of interfer-
ence effects and making quantum computation error-prone. A. HOW YOU PERFORM COMPUTATION WITH QUBITS
Quantum error correction techniques are used to mitigate the Practically all classical computers work in a similar way.
impact of the external environment and preserve the delicate They use a group of bits to store information in a binary form
quantum interference necessary for quantum computation. within memory. The state of these bits can be altered using
Overall, interference is a foundational concept in quantum logical gates like AND, OR, NOT, and NAND gates.
computing, allowing quantum systems to perform computa- Quantum computing utilizes a model similar to classical
tions by updating the probability distributions of the quantum computing known as the gate model or circuit model as shown
states. This concept solves certain problems in ways that are in Figure 6. Within this framework, a set of qubits is used
not achievable using classical computing methods. and their states can be superposed and entangled with each
other. Multiple gates are available for conducting operations
E. QUANTUM NOISE on these qubits, thereby altering the probability distribution
Quantum noise refers to the uncertainty and fluctuations that of states within a quantum circuit. Quantum algorithms are
arise in quantum systems due to the probabilistic nature of constructed by applying a sequence of gates to qubits in a
quantum mechanics. It is a challenge in quantum systems specific order. Ultimately, the measurement is executed at
even at low temperatures. the end of the quantum circuit, collapsing the superposition
In classical systems, noise is often associated with random states and revealing the quantum circuit’s state based on the
variations in signals or disturbances caused by external probability distribution of the states.
factors. When a quantum system is in a superposition state,
its outcome upon measurement is not deterministic but is
determined by the probability distribution of the quantum
states. Noise and error can affect the outcome due to the
quantum system’s interaction with the external environment.
It can lead to loss of quantum properties (superposition,
entanglement, and, interference) over time affecting the
outcome of quantum circuits.
Quantum noise has several manifestations in quantum
systems, and it can impact various aspects of quantum
computing. Some common examples of quantum noise
include: FIGURE 6. Quantum circuit example.
• Measurement Noise: When measuring a quantum sys-
tem, the act of measurement can cause a quantum
system to lose the quantum superposition and collapse
B. QUANTUM GATES
the quantum state into one of its states, introducing
uncertainty in the outcome due to the probabilistic nature Quantum gates are basic building blocks in quantum
of the measurement process. computing that manipulate the state of the qubits in quantum
• Decoherence: Interactions with the environment can
circuits. They are analogous to classical logic gates in
cause quantum systems to lose quantum superposition, classical computing but leverage the principles of quantum
entanglement, and interference, affecting the perfor- mechanics for processing quantum information. Quantum
mance of quantum algorithms. gates are summarized in Table 1 with symbol and matrix
Quantum noise poses a significant challenge for quantum representation.
computing. To address this challenge, researchers have been
working on quantum error correction techniques, which 1) IDENTITY GATE
are essential for preserving the quantum states against the The identity gate is a single-qubit gate. It does not modify
detrimental effects of measurement noise and decoherence. the state of a qubit; therefore, it is represented by an identity

VOLUME 12, 2024 22301

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

matrix as shown in equation (25). 3) TWO-QUBIT CONTROLLED GATES

  Controlled gates are two-qubit gates, where the first qubit
10 acts as a control and the second qubit is a target qubit.
I= → (25)
01 The controlled-NOT gate (CNOT) also known as controlled-
X gate (CX), operating on two qubits, performs the NOT
The identity gate is helpful during mathematical computation. operation on the second qubit (target qubit) only when the
This gate is used to compute the resultant gate matrix in multi- first qubit (control qubit) is |1⟩ otherwise leaves the second
qubit circuits. The identity matrix is an involutory matrix, qubit unchanged. It is represented by the matrix CNOT.
meaning that the identity matrix is equal to its inverse matrix
as shown in equation (26).  
1 0 0 0
0 1 0 0
I = I −1 I2 = I (26) CNOT = 
0
 → (34)
0 0 1
0 0 1 0
2) SINGLE-QUBIT PAULI GATES
The Pauli gates (X , Y , and Z ) are based on Pauli matrices Controlled gates can be generalized using the universal
(σx , σy , σz ), which are useful to modify the state or phase of matrix U which is a single-qubit matrix.
the qubit. Pauli gates are single-qubit gates and they rotate  
u00 u01
the state of the qubit around the x, y, and z axes of the Bloch U= (35)
u10 u11
sphere by π radians.
• Pauli-X gate is the quantum equivalent of the NOT gate The controlled-U gate acting on two qubit can be defined
for classical computers. It maps |0⟩ to |1⟩ and |1⟩ to |0⟩. as CU.
It is also called a qubit-flip gate.
 
10 0 0
• Pauli-Y maps |0⟩ to i|1⟩ and |1⟩ to −i|0⟩. 0 1 0 0 
• Pauli-Z maps |1⟩ to −|1⟩ and leaves the state |0⟩ CU =  
0 0 u00 u01  (36)
unchanged. Due to this nature, Pauli Z is also called a 0 0 u10 u11
phase-flip gate.
These matrices are usually represented as where U is one of the Pauli matrices that is X, Y, and Z. Based
on the respective Pauli matrices, we can have ‘‘controlled-
X’’, ‘‘controlled-Y’’, or ‘‘controlled-Z’’ gates. The shortened
  symbols of these controlled gates are CX, CY, and CZ
01
X= → (27) respectively.
10
  Controlled-Y gate (CY) performs the Y operation on the
0 −i second qubit only when the first qubit is |1⟩, and otherwise
Y = → (28)
i 0 leaves it unchanged.
 
1 0 
100 0

Z= → (29)
0 −1 0 1 0 0 
CY = 0 0 0 −i →
 (37)
The Pauli matrices are anti-commute as shown in 00 i 0
equation (30). Two matrices A and B are considered to be
anti-commute if AB = −BA. Controlled-Z gate (CZ) performs the Z operation (phase
flip) on the second qubit only when the first qubit is |1⟩, and
ZX = −XZ = iY (30) otherwise leaves it unchanged.

The Pauli matrices are traceless matrices meaning the sum 

1 0 0 0

of the eigenvalues of Pauli matrices is zero. 0 1 0 0
CZ = 
0
 → (38)
  0 1 0
−1 0 0 0 −1
eig(X ) = eig(Y ) = eig(Z ) = (31)
1
The Controlled gates are Hermitian matrices as shown in
The Pauli matrices are also involutory, meaning that Pauli equation (39). A square matrix (complex or real) is said to be
matrices are equal to their inverse matrices as shown in a Hermitian matrix if it is equal to its own conjugate transpose
equation (32), and the square of the Pauli matrices is the matrix.
identity matrix as shown in equation (33),.
CX = CX T CY = CY T CZ = CZ T (39)
−1 −1 −1
X =X Y =Y Z =Z (32) The Controlled gates are also involutory, meaning that
2 2 2 2
X =Y =Z =I (33) Controlled gates matrices are equal to their inverse matrices

22302 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

as shown in equation (40), and the square of a Con- 6) TWO-QUBIT SWAP GATE
trolled gates matrices is the identity matrix as shown in A swap gate performs a swap operation between two qubits.
equation (41). It is a fundamental quantum gate used in quantum computing
CX = CX −1 CY = CY −1 CZ = CZ −1 (40) with the primary purpose of exchanging the states of two
2 2 2 2
qubits.
CX = CY = CZ = I (41)
SWAP|qubit1, qubit2⟩ = |qubit2, qubit1⟩ (51)
4) SINGLE-QUBIT PHASE SHIFT GATES
The phase shift gates are quantum gates that introduce a phase It is represented by the matrix:
shift to the quantum state of a qubit. These are one-qubit
gates and these are used to alter the phase of the qubit’s  
1 0 0 0
state without changing its probability amplitudes. They map 0 0 1 0
the states |0⟩ → |0⟩ and |1⟩ → eiϕ |1⟩. The probability of SWAP = 
0
 → (52)
1 0 0
measuring a |0⟩ or |1⟩ does not change after applying this
0 0 0 1
gate. Generally, the phase shift gate is generally represented
by the matrix P:
  The SWAP matrix is an involutory matrix as shown in
1 0 equation (53).
P= (42)
0 eiϕ
where ϕ is the phase shift with the period 2π. Some common SWAP = SWAP−1 SWAP2 = I (53)
examples of phase shift gates are,
π
• T gate where ϕ = 4 7) THREE-QUBIT TOFFOLI (CCNOT) GATE
The Toffoli gate is named after Tommaso Toffoli and is also
π
 
1 0
T = P( ) = π → (43) known as CCNOT gate (Controlled-Controlled-NOT gate) or
4 0 ei 4 Deutsch gate D(π/2). The Toffoli gate is like a CNOT gate
• S gate, though S notation is sometimes used for SWAP with two control qubits and one target qubit. The target qubit
gate where ϕ = π2 will be inverted if the first and second qubits are in |1⟩ state.
π

1 0
 It is represented by the matrix CCNOT as given below
S = P( ) = π → (44)
2 0 ei 2 
10000000


• The Pauli-Z gate where ϕ = π 0 1 0 0 0 0 0 0

 
0 0 1 0 0 0 0 0
 
  0 0 0 1 0 0 0 0
1 0 CCNOT =   → (54)
Z = P(π) = → (45) 0 0 0 0 1 0 0 0

0 eiπ 0 0 0 0 0 1 0 0
 
0 0 0 0 0 0 0 1
00000010
5) SINGLE-QUBIT HADAMARD GATE
The Hadamard or Walsh-Hadamard gate, named after The CCNOT matrix is an involutory matrix as shown in
Jacques Hadamard and Joseph L. Walsh, is applied on a single equation (55).
qubit. It creates superposition states with an equal probability
distribution for a given qubit. It maps the state of the qubit as CCNOT = CCNOT −1 CCNOT 2 = I (55)
follows:
|0⟩ + |1⟩
|0⟩ → √ (46)
2 C. MEASUREMENT
|0⟩ − |1⟩ In quantum computing, measurement is a fundamental
|1⟩ → √ (47)
2 operation that provides a way to extract information from a
quantum system. When a quantum measurement is performed
The Hadamard gate is represented by a matrix H:
  on a qubit, it yields a classical result. This measurement
1 1 1 outcome collapses the superposition states of the qubit and
H=√ → (48)
2 1 −1 provides either 0 or 1 based on the probability distribution
The Hadamard matrix is a traceless and involutory matrix of the qubit states. Measurement provides information about
as shown in equation (49) and equation (50) respectively. the quantum state of the qubit at the time of measurement.
 √  Measurements convert multiple (superposition) probabilistic
− 2
eig(H ) = √ (49) states to one absolute state and collapse other states and
2 superpositions. This is known as the measurement problem
H = H −1 H2 = I (50) of quantum mechanics.

VOLUME 12, 2024 22303

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

TABLE 1. Common quantum logic gates.

IV. BASIC QUANTUM CIRCUITS number of states for the 1-qubit quantum system, that is, two.
A quantum circuit is a series of qubits and gates. Qubits The output state of the quantum system can be calculated
can be in superposed or entangled states. Gates are used to using the matrix product of the gate matrix and the vector
change the state of qubits. Several gates perform different notation of the qubit. NOT gate and Hadamard gate are
operations that are summarized in Table 1. Quantum gates are examples of single gate applied on a single qubit as shown
represented in matrix form, and the qubit’s states are denoted in Figures 7a and 7b, respectively.
in vector notation. The overall state of the quantum system
can be calculated using the matrix product between the gate’s
matrices and qubit vectors. Common examples of quantum B. MULTIPLE-QUBIT GATE ON MULTIPLE QUBITS
circuits are summarized in Figure 7. In a multiple qubits gate operated on multiple qubits, the order
of the gate matrix depends on the number of qubits it operates.
A. SINGLE-QUBIT GATE ON SINGLE QUBIT For a multiple qubits gate applied on two qubits, the order of
In a single-qubit gate operated on a single qubit, the order the gate matrix would be 4 × 4 which is equal to the total
of the gate matrix will be 2 × 2 which is equal to the total number of states for a 2-qubit quantum system. Similarly, for

22304 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

a multiple qubits gate operated on three qubits, the order of using the matrix product of the input quantum state and the
the gate matrix would be 8 × 8 which is equal to the total corresponding gate matrix as shown in equation (66).
number of states for the 3-qubit quantum system and so on. In Figure 7e, 1-qubit Hadamard and NOT gates are applied
The resultant input quantum state for multiple qubits can be on a 2-qubit quantum system. The output state of the quantum
calculated using the Kronecker product (or tensor product) of system is computed using the second method as explained
all qubits of the quantum system as shown in equation (58). above. If the number of single-qubit gates are less than the
For two or multiple qubits, the tensor product is used to find total number of qubits, then identity gate can be used for the
the resultant quantum states. The tensor product is denoted computation of the resultant multiple-qubit gate.
by the symbol ⊗.
Let us consider qubits |a⟩ and |b⟩: D. ENTANGLED CIRCUITS
 
a
 
b In classical systems, only one state exists at a time but in
|a⟩ = o and |b⟩ = o (56) quantum computing, all states can exist simultaneously. This
a1 b1
combination of the states is called superposition in quantum
The resultant input quantum state for multiple qubits can systems. These superposed states continue if there is no
be calculated using the tensor product of all qubits of the external influence. However, when the superposed states are
quantum system. measured externally, the superposition collapses and the final
|α⟩ = |a⟩ ⊗ |b⟩ (57) output state is recorded based on the probability distribution
   of the qubit states. Superposed states of the qubit can also be
bo  
a o ∗ ao bo entangled states or not. It can be understood with an example.
 b1  


 = ao b1 
 Let |v⟩ represents a 2-qubit quantum system as
|α⟩ = |ab⟩ =      a1 bo  (58)
b  |v⟩ = a0 b0 |00⟩ + a0 b1 |01⟩ + a1 b0 |10⟩ + a1 b1 |11⟩ (67)
a1 ∗ o

a1 b1
b1 |v⟩ = |a⟩ ⊗ |b⟩ = v0 |00⟩ + v1 |01⟩ + v2 |10⟩ + v3 |11⟩ (68)
|α⟩ = ao bo |00⟩ + ao b1 |01⟩ + a1 bo |10⟩ + a1 b1 |11⟩ (59)
where |v0 |2 ,|v1 |2 ,|v2 |2 , and |v3 |2 are the probabilities of |v⟩
|α⟩ = αo |00⟩ + α1 |01⟩ + α2 |10⟩ + α3 |11⟩ (60)
qubit states.
The output state of the quantum system can be calculated If |v⟩ can be written as
using the matrix product of the gate matrix and vector
|v⟩ = (a0 |0⟩ + a1 |1⟩) × (b0 |0⟩ + b1 |1⟩) = |a⟩ ⊗ |b⟩ (69)
notation of the input quantum state of the gate. SWAP gate
and CNOT gate are examples of multiple-qubit gates applied then |v⟩ is not entangled because |v⟩ can be written as
on multiple qubits shown in Figures 7c and 7d respectively. separable states of qubits |a⟩ and |b⟩. The probabilities
of qubit |a⟩ and |b⟩ states are independent of each other.
C. LOWER ORDER QUBIT GATE ON HIGHER # QUBITS However, if |v⟩ cannot be written into separable states then
In this case, the order of the gate matrix is less than the these are entangled states. Figure 7f shows an example of an
total states of the quantum system. In such a scenario, two entangled quantum circuit with calculations and probabilities
methods are possible. In the first method, the lower order amplitude plot.
qubit gates g0 and g1 are applied to the qubits individually.
The output state |C⟩ of the quantum system is computed using 1) NON-ENTANGLED SYSTEM EXAMPLE
the Kronecker product of the output vectors |c0⟩ and |c1⟩ Let us take an example of a quantum system that is
which are calculated from each lower order qubit gate as not entangled and see the effect of measurement on the
shown in equation (63). probabilities of 2-qubit quantum states:
|c0⟩ = g0 × |a⟩ (61) 1
|φ1⟩ = √ (|0a ⟩|0b ⟩ + |0a ⟩|1b ⟩) (70)
|c1⟩ = g1 × |b⟩ (62) 2
|C⟩ = |c0⟩ ⊗ |c1⟩ (63) the probability of |a⟩ as |0a ⟩ before measuring |b⟩ is
The second method is to compute the resultant input state |v⟩ 1 1
prob(|0a ⟩) = ( √ )2 + ( √ )2 = 1 (71)
of the quantum system using the Kronecker product of the 2 2
qubits, that is:
Suppose if we measure the state of |b⟩ as |1b ⟩ then the
|v⟩ = |a⟩ ⊗ |b⟩ (64) superposed states of |φ1⟩ will be collapsed and we end up
G = g0 ⊗ g1 (65) with
|C⟩ = G × |v⟩ (66) |φ1⟩ = |0a ⟩|1b ⟩ (72)
The resultant quantum gate G can also be calculated using Now the probability of |a⟩ as |0a ⟩ after measuring |b⟩ is
the Kronecker product of lower order qubit gates as shown
in equation (65). The output quantum state can be found prob(|0a ⟩) = (1)2 = 1 (73)

VOLUME 12, 2024 22305

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

The probability of |a⟩ states have not been affected by the B. BERNSTEIN-VAZIRANI ALGORITHM
change in |b⟩ state therefore it is a non-entangled quantum The Bernstein-Vazirani algorithm is a quantum algorithm
system. designed to find a hidden binary string in a function. It uses
quantum principles to extract information about the hidden
2) ENTANGLED SYSTEM EXAMPLE string and determine it with fewer queries than classical
Now let us take an example of a quantum system that algorithms, even in the presence of noise and imperfect equip-
is entangled and see the effect of measurement on the ment [43]. As the number of bits in the secret string increases,
probabilities of 2-qubit quantum states: the probability of correctly guessing the string becomes less
dependent on the type of disorder and more reliant on the cen-
1
|φ2⟩ = √ (|0a ⟩|0b ⟩ + |1a ⟩|1b ⟩) (74) ter and spread of the disorder [44]. The classical algorithm,
2 on the other hand, becomes inefficient for long strings, even
The probability of |a⟩ as |0a ⟩ before measuring |b⟩ is in a noiseless scenario [45]. Overall, the Bernstein-Vazirani
algorithm outperforms classical algorithms in most cases.
1 This algorithm has potential applications in cryptography
prob(|0a ⟩) = ( √ )2 = 0.5 (75)
2 and demonstrates the advantages of quantum computing for
certain problems.
Suppose if we measure the state of |b⟩ as |1b ⟩ then the
superposed states of |φ2⟩ will be collapsed and we end up
with C. SIMON’S ALGORITHM
Simon’s algorithm is a quantum algorithm for obtaining the
|φ2⟩ = |1a ⟩|1b ⟩ (76) period of a vectorial Boolean function with polynomial time
complexity. The algorithm achieves exponential speedup
Now the probability of |a⟩ as |0a ⟩ after measuring |b⟩ is over classical algorithms. It has applications in quantum
cryptanalysis and cryptography. Simon’s algorithm has been
prob(|0a ⟩) = 0 (77)
utilized to study the autocorrelation spectrum and Walsh
Here, the probability of |a⟩ state has been changed by the spectrum of Boolean functions [45]. It has also been
change in |b⟩ state, therefore, it is an entangled quantum applied in the design of a lightweight encryption algorithm
system. called SIMON-GCM for IoT security, which combines the
SIMON cipher block and Galois/Counter Mode (GCM) [46].
V. QUANTUM ALGORITHMS Additionally, this algorithm has been analyzed for different
Quantum algorithms hold unique capabilities and appli- use cases in cryptanalysis [47].
cations that outperform classical counterparts. From
Shor’s algorithm for factoring large numbers to Grover’s D. SHOR’S ALGORITHM
algorithm for searching databases, we discuss various One of the most compelling quantum algorithm is Shor’s
quantum algorithms that have the potential to revolutionize algorithm [5], which is a quantum algorithm for finding the
computation. prime factors of an integer. It was developed in 1994 by
the American mathematician Peter Shor [48]. It has the
A. DEUTSCH-JOZSA ALGORITHM potential to break widely used encryption schemes, such
The Deutsch-Jozsa algorithm is a quantum algorithm that as RSA, which rely on the difficulty of factoring large
solves the Deutsch-Jozsa problem, which involves deter- numbers. It leverages the quantum properties of superposition
mining whether a given Boolean function is balanced or and entanglement to perform the factorization process
constant. The algorithm can solve this problem with just exponentially faster than the best-known classical algorithms.
one step, providing a significant speedup compared to In the integer factorization problem, given an integer N =
classical algorithms. It was proposed by David Deutsch and p × q for some prime numbers p and q, the main goal is to
Richard Jozsa in 1992 and is one of the early examples find the prime factors p and q. The traditional
√ trial division
of a quantum algorithm that provides a significant speedup method has a time complexity of about O( N ) and the best
over classical algorithms. The algorithm generalizes the classical algorithm is the general number field sieve (GNFS),
Deutsch algorithm to handle multiple degrees of freedom which has a sub-exponential time complexity. For GNFS, the
and can be derived from the quantum Fourier transform time complexity is roughly O(exp((64/9)1/3 × (logN )1/3 ×
algorithm [40]. Additionally, the algorithm has been used as (log ∗ logN )2/3 )) [49]. Shor’s quantum algorithm solves
an educational experiment to demonstrate qubit fundamental this problem substantially faster, in time O(logN )3 ). It’s
concepts and algorithmic challenges in quantum science important to note that the implementation of Shor’s algorithm
and technology [41]. Experimental implementations of the becomes more complex for large numbers. In practice, Shor’s
Deutsch-Jozsa algorithm have also been performed on IBM’s algorithm demonstrates the potential of quantum computing
quantum computer, showcasing its efficiency compared to to break widely used encryption methods, emphasizing the
classical techniques [42]. need for post-quantum cryptography techniques.

22306 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

FIGURE 7. Basic circuits of quantum systems.

VOLUME 12, 2024 22307

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

E. GROVER’S ALGORITHM addressing the set of supervised and unsupervised learning

Grover’s algorithm is a quantum algorithm designed to find a tasks [11], [58], [59], [60]. These algorithms offer solutions
specific value within unsorted (or unstructured) databases or that are faster compared to their classical counterparts for
lists exponentially faster than classical algorithms. It provides certain problems. For example, the quantum technique for
a quadratic speedup,√allowing it to search the target item solving systems of linear equations [61] is utilized for clas-
in approximately O( N ) steps instead of the O(N ) queries sification problems such as perceptron or linear regression
required classically where N is the number of items in the training.
database. Grover’s algorithm is a fundamental demonstration
of quantum computing’s potential for solving certain search B. SECOND GENERATION OF QUANTUM MACHINE
and optimization problems faster than classical computers. LEARNING
It was proposed by Lov Grover in 1996 [50]. Since classical With the advent of NISQ processors, the second generation
algorithms for NP-complete problems (nondeterministic of QML has emerged [62], [63], [64]. Differing from the first
polynomial time problems) require exponentially many steps, generation, this new trend in QML relies on heuristic methods
and Grover’s algorithm provides at most a quadratic speedup; due to the increased computational capabilities of quantum
this suggests that Grover’s algorithm by itself will not provide hardware particularly in the field of deep learning [65]. These
polynomial-time solutions for NP-complete problems [51]. novel quantum algorithms utilize parameterized quantum
circuits, often referred to as Parameterized Quantum Circuits
VI. DESIGNING HYBRID CLASSICAL-QUANTUM (PQCs) or Quantum Neural Networks (QNNs) [66] in deep
MACHINE LEARNING MODELS learning applications. Similar to classical deep learning,
Machine learning involves creating models or algorithms that QNN parameters are optimized with respect to a cost function
are capable of learning patterns within datasets. Machine through black-box optimization heuristics [67] or gradient-
learning includes regression, classification, segmentation, based methods [68]. This optimization aims to facilitate the
object detection, etc. After learning from a dataset, these learning of data representations.
models become capable of predicting outcomes for unseen
data. Originally, analytical approaches were used in Machine C. HYBRID MACHINE LEARNING
learning [52]; however, recently, heuristic-based methods Quantum processors are still fairly small and noisy, therefore
have become more dominant due to the abundance of to improve machine learning performance effectively, NISQ
data and computational resources [53]. These methods have processors will need to work with classical co-processors in
achieved substantial success in the field of deep learning [54], hybrid mode. Scalability, training time, and model accuracy
[55]. Deep learning methods develop data representation in are the key factors in the field of deep learning [69] and these
the form of high-dimensional vectors using parameterized are the limitations of classical deep learning methods. These
layers. Optimization techniques are used to fine-tune these issues can also be addressed with the combination of classical
parameters and fit the deep learning model. and quantum processors. The abstract model of the classical-
Concurrently, academia and industry have witnessed a quantum deep neural network is shown in Figure 8.
growing interest in quantum computing [34]. Quantum com-
puting utilizes quantum mechanics principles such as super-
position, entanglement, and quantum parallelism to improve
the performance of classical machine learning algorithms.
Quantum computers exhibit these novel behaviors that are
often challenging to simulate using classical computers.
These novel behaviors of quantum computers make them
capable of solving certain problems more efficiently as
compared to classical systems. Quantum computers have the
FIGURE 8. Abstract structure for hybrid classical-quantum model. ψ
potential to accelerate tasks such as optimization [56] and represents quantum model (QNN) parameters.
cryptanalysis [57].
Liu et al. have proposed a hybrid quantum-classical
A. FIRST GENERATION OF QUANTUM MACHINE convolutional neural network (QCCNN), similar to convo-
LEARNING lutional neural networks (CNNs) in classical deep learning
Quantum computing is capable of enhancing the machine but adapted to quantum computing to improve the process
learning design process. A pivotal development leading to the of feature mapping [70]. Another study [71] has proposed
utilization of quantum computers in machine learning is their a hybrid quantum-classical graph convolutional network
ability to speed up linear algebra operations exponentially as (QGCNN) for learning high-energy physics data. The
state space grows. These quantum machine learning (QML) proposed framework has shown an advantage over classical
algorithms, based on quantum-accelerated linear algebra multilayer perceptron and convolutional neural networks in
techniques, form the first generation of QML algorithms, the aspect of the number of parameters with comparable

22308 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

accuracy. This paper Liang et al. [72] have proposed the computing namely superconducting, photonic, trapped ion,
idea of a hybrid quantum–classical neural network with neutral atoms, and quantum dots. Superconducting quantum
deep residual learning (Res-HQCNN). They have designed computing is the widely used type of quantum supercomput-
a residual block structure with a quantum neural network and ing. Due to highly customized end use, and the limitations of
the corresponding training algorithm. the resources such as Quantum hardware, quantum experts
believe that the design, delivery, and deployment of this
VII. QUANTUM SIMULATORS type of computing can be challenging. The topology that is
Quantum simulators are used to test and debug quantum the most studied and vetted of the superconducting circuit
circuits with realistic noise models. Following are five approaches is the nearest-neighbor cavity coupled transmon
notable quantum simulators: qubits [78].
Quantum computing circuits operate at superconducting
A. QISKIT AER temperatures. The qubits are sensitive to environmental inter-
Qiskit Aer [73] is a high-performance quantum computing ference such as electromagnetic and thermal fluctuations;
simulator provided by IBM. It provides interfaces to run hence the system needs to be maintained at near absolute zero
quantum circuits with various noise models. It also includes temperature. Depending on the type of quantum computing,
state vector and density matrix simulators for realistic a temperature of 10-20mK is required using a state-of-the-
simulations. It can also utilize graphical processing units art dilution refrigerator, the heart of quantum computing
(GPUs) to improve simulation performance. infrastructure.
Qubits are created with superconducting materials.
B. QUTIP (QUANTUM TOOLBOX IN PYTHON) At extremely low (near absolute zero) temperatures, the
QuTiP [74] is an open-source quantum computing framework superconducting circuits act as superconducting material as
for Python. It is designed for simulating the dynamics of open they carry current at low electrical resistivity. A Josephson
quantum systems. This framework depends on the Scipy, junction is an assembly of two weakly coupled supercon-
Cython, and Numpy packages. In addition, Matplotlib is used ductors with a thin layer of insulator. At superconducting
for graphical output. QuTiP is user-friendly and free of any temperature, electric current flows through the Josephson
licensing fees, therefore, it is considered suitable for learning junction [79] with no applied voltage. To generate a qubit,
quantum computing in the classroom. a linear capacitor made of a superconducting material
and an insulator is tied together to superconducting wires
C. CIRQ with Josephson junction in the conducting loop to form
Cirq [75] is an open-source interface designed for program- an artificial atom which holds the qubit. At near absolute
ming quantum computers. It is a Python software library zero temperature, the qubit has no thermal energy in the
developed for creating and simulating quantum circuits on surroundings, so it stays in a stable state. The qubits are then
quantum processors. Cirq also includes a simulator to run entangled using a microwave pulse with a range depending
quantum algorithms for testing and debugging. upon the strength of the magnetic field of the qubit or its
resonance frequency. The quantum computer setup consists
D. PROJECTQ of logical and physical qubits. Physical qubits are the actual
ProjectQ [76] is an open-source programming interface quantum bits implemented in quantum hardware. They are
for quantum computers. It supports compilation framework the primary building blocks of quantum computers, typically
for various types of quantum hardware such as IBM found in quantum systems like superconducting circuits,
Quantum Experience chip, Azure Quantum, AWS Braket, photons, or trapped ions. Physical qubits are susceptible to
AQT devices, or ion-trap quantum (IonQ) devices. ProjectQ errors and noise due to the external environment. Several
also includes a simulator for testing and debugging quantum physical qubits or the actual quantum hardware are grouped
algorithms. together to form a single logical qubit. Logical qubits are a
higher-level concept that represents quantum information that
E. PYQUIL is protected against noise and errors. The logical qubits are
pyQuil [77] is a Python library for writing quantum programs used for error and noise detection.
using Quil, the quantum instruction language developed The superconducting quantum computing cooling setup
at Rigetti Computing, Inc. It comes with a quantum mainly consists of the following components and is shown
virtual machine (QVM) for simulating and testing quantum in Figure 9,
circuits. • Dilution refrigerator (Cryostat with pulse tube cooler)
• Gas handling system with gas/liquid mixture pumps,
VIII. QUANTUM COMPUTING DEPLOYMENT valves and controls
REQUIREMENTS • Liquid Nitrogen Dewars
Quantum computing promises to outperform classical com- • 3 He and 4 He storage tanks
puting by solving certain problems. Though still in the • Cryo compressor
development phase, there are many approaches to quantum • Cryostat control panel

VOLUME 12, 2024 22309

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

not require frequent maintenance, however, proactive pre-

ventative maintenance can ensure the longevity of the entire
quantum computing system. The parts are manufactured
by multiple vendors; therefore assembly, integration, and
commissioning of the entire system can be challenging. The
quantum computing operators need to learn and obtain hands-
on experience to ensure the optimum functioning of the setup.

IX. APPLICATIONS OF QUANTUM COMPUTING

FIGURE 9. Quantum computing cooling infrastructure layout. Quantum computing has the potential to transform many
areas including astrophysics, aerospace, pharmaceuticals,
drug discovery, artificial intelligence, cybersecurity, and
The quantum qubits need to be stored at near absolute secure communication. It has the potential to perform certain
zero temperatures for which a mixture of 3 He and 4 He along tasks exponentially faster than classical computing. Quantum
with nitrogen is used in the dilution refrigerator [80]. Liquid computing can be used to enhance algorithms and discover
nitrogen is used for 80K cooling and the He mixture is structures and patterns efficiently. In this section, we will dive
to achieve 2K cooling. The liquid nitrogen is also used to into the applications of quantum computing in more detail.
cool the 3 He return from the system and to remove any Quantum computing applications are summarized in Table 2.
air impurities. Due to its sensitive nature, the presence of
any AC and DC magnetic fields may be checked in the A. CRYPTOGRAPHY
room housing the cryostat prior to its installation. Helium Cryptography involves the study of secure communication,
and its isotope used in the cryogenic cooling process may and quantum computing has the potential to undermine
potentially leak and pose issues with the depletion of oxygen numerous encryption techniques employed in classical cryp-
in space. A helium leak detector with oxygen deficiency tography, including RSA and elliptic curve cryptography.
hazard alarm sensors is recommended to be installed in However, quantum computing can also be used to create
quantum computing lab space. new cryptographic protocols that are resistant to quantum
The gas handling system (GHS) is an essential element of a attacks. For example, it can be used for secure key distribution
quantum cooling system. It comprises a cabinet that contains over quantum channels, which have advantages over classical
the 3 He and 4 He mixture reservoir, pumps for circulation, channels in terms of detecting eavesdropping [81]. Addition-
valves, chilled water, liquid nitrogen connection, and controls ally, Quantum cryptography schemes can protect wireless
(mostly pneumatic). The operation of GHS is considerably sensor networks from attacks by hackers [92]. Quantum
noisy, so it is recommended to place it in separate rooms computing has emerged as a compelling complement to
alongside the cryostat compressors. Also, vibration isolators classical technologies for applications in security and com-
are recommended under the Dilution refrigeration assembly. munications [93]. In a nutshell, quantum computing offers
The ambient space of the quantum setup as shown in Figure 9 promising solutions for cryptanalysis as well as for enhancing
may be maintained at 68-70F dry bulb and 30-50% relative the security and efficiency of cryptographic systems.
humidity levels.
Unlike conventional high-performance computing (HPC) B. OPTIMIZATION
using graphics processing units (GPUs) and central pro- Optimization is the process of finding the best solution to a
cessing units (CPUs) with thermal design power (TDP) of given problem from a set of solutions. Quantum computers
more than 1000W, a superconducting quantum system relies can efficiently solve certain optimization problems that
on low power consumption. For operational reliability of are complicated tasks for classical computers. Quantum
the quantum computer, clean power from a UPS source is algorithms such as Grover search, quantum phase estima-
generally recommended. The nominal voltage ranges are tion, quantum annealing, quantum approximate optimization
120V-480V. Depending on the size of the qubit system, the algorithm, and variational quantum eigensolver can be
Cryo compressor may need a dedicated chiller to provide used to solve optimization problems [94]. Additionally,
lower-temperature cooling water generally at a higher delta quantum computing can be applied in flight path optimization
> 30F across the compressor. within the aerospace engineering domain, where quantum
The requirements for connectivity are generally scalable computing can tackle computational challenges and improve
and flexible as the need and user community grows over time. performance over classical algorithms [95].
A microwave amplifier, signal synthesizers, a multi-channel
arbitrary pulse sequencer, 1 gigabyte of DDR3 SDRAM, and C. CHEMISTRY
a gigabit Ethernet interface for high-speed data upload are Quantum computing has numerous applications in pharma-
generally required for a qubit setup. ceuticals and drug discovery. It can be used for generative
There are many moving components in the cryogenics chemistry and quantum chemistry simulations [96]. By using
of the Quantum Infrastructure. Most of the components do quantum computing, drug companies can potentially save

22310 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

TABLE 2. Summary of quantum applications with algorithm/model used.

time and money by accelerating the drug discovery process, applications, which can help in understanding wave-particle
leading to a more efficient and productive pharmaceutical interactions and nonlinear plasma dynamics [101].
industry [97]. Quantum computing can also reduce costs and
time in drug development by decreasing the number of neces-
F. FINANCE
sary biochemical experiments [98]. Quantum computing has
been used in protein structure prediction, molecular docking, Quantum computing has numerous applications in finance.
and quantum simulation [99]. Although current quantum Quantum algorithms may improve performance of tasks such
devices are still susceptible to external noise and error, but as risk assessment, portfolio optimization, and derivative
hybrid quantum-classical techniques are well suited for drug pricing, offering solutions that outperform classical coun-
discovery and development. terparts. It can be used for financial interest rate data [89].
It has also been applied to develop financial models, such
as churn prediction and credit risk assessment, where it
D. MACHINE LEARNING has demonstrated better performance compared to traditional
Quantum computing has shown potential to improve machine methods [84], [102]. Quantum computing offers significant
learning model accuracy. Quantum machine learning algo- benefits in terms of computational speed and accuracy,
rithms such as quantum neural networks, have been applied making it a valuable tool in the finance field [103].
in the context of image classification for iris, sonar, breast
cancer, idiopathic pulmonary fibros, CIFAR-10 and MNIST
X. CONCLUSION AND FUTURE OF QUANTUM
datasets [82], [83], [100]. These quantum models have shown
COMPUTING
benefits over classical models, including reduced training
time and improved model accuracy. While there is still a This paper has offered a thorough exploration of the
need for further development of quantum hardware, quantum field of quantum computing, making its complexities more
machine learning holds great potential for improving the understandable. We have covered the fundamental concepts
efficiency of classical machine learning methods. of quantum mechanics and the evolving quantum computing
landscape with practical applications. This paper has dis-
cussed noisy intermediate-scale quantum (NISQ) along with
E. ENERGY quantum computing fundamentals, such as qubits’ notations,
Quantum computing has various applications in the field quantum superposition, quantum entanglement, quantum
of energy. One area of application is in power and energy interference, and quantum noise. We have elaborated building
systems, where quantum computing algorithms like Grover’s blocks of quantum systems, such as quantum gates, quantum
algorithm can be applied to solve problems more efficiently measurements, and basic quantum circuits. We have pro-
than traditional algorithms. Another application area is vided an overview of various quantum algorithms, such as
battery profiling, where quantum computing can be used the Deutsch-Jozsa algorithm, Bernstein-Vazirani algorithm,
to model the degradation profile of the battery [85]. Addi- Shor’s algorithm, and Grover’s algorithm. We have discussed
tionally, quantum computing can contribute to simulating the design of hybrid classical-quantum machine learning
linear and nonlinear dynamics in fusion energy science algorithms. We have further discussed the deployment

VOLUME 12, 2024 22311

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

requirements of superconducting quantum computing, the [11] S. Lloyd, M. Mohseni, and P. Rebentrost, ‘‘Quantum algorithms for
most prevalent form of quantum supercomputing. supervised and unsupervised machine learning,’’ 2013, arXiv:1307.0411.
[12] T. F. Rønnow, Z. Wang, J. Job, S. Boixo, S. V. Isakov, D. Wecker,
In quantum computing, some areas are getting significant J. M. Martinis, D. A. Lidar, and M. Troyer, ‘‘Defining and detecting
attention from researchers. One key focus area is making quantum speedup,’’ Science, vol. 345, no. 6195, pp. 420–424, 6195.
quantum computers more robust by improving fault tolerance [13] I. L. Chuang, R. Laflamme, P. W. Shor, and W. H. Zurek, ‘‘Quantum
computers, factoring, and decoherence,’’ Science, vol. 270, no. 5242,
and error correction. Another important area is optimizing pp. 1633–1635, Dec. 1995.
quantum algorithms to make them more efficient for solving [14] W. G. Unruh, ‘‘Maintaining coherence in quantum computers,’’ Phys. Rev.
different kinds of problems. Machine learning researchers are A, Gen. Phys., vol. 51, no. 2, pp. 992–997, Feb. 1995.
[15] B. Georgeot and D. L. Shepelyansky, ‘‘Quantum chaos border for
also using quantum and classical-quantum hybrid approaches
quantum computing,’’ Phys. Rev. E, Stat. Phys. Plasmas Fluids Relat.
to learn complex patterns in high-dimensional data. Addi- Interdiscip. Top., vol. 62, no. 3, pp. 3504–3507, Sep. 2000.
tionally, researchers are looking into ways to make quantum [16] G. Kalai, ‘‘The argument against quantum computers,’’ in Quantum,
communication more secure through the development of Probability, Logic: The Work and Influence of Itamar Pitowsky. Springer,
2020, pp. 399–422.
various protocols like quantum key distribution and quantum [17] R. Babbush, J. R. McClean, M. Newman, C. Gidney, S. Boixo, and
cryptography. H. Neven, ‘‘Focus beyond quadratic speedups for error-corrected quan-
Quantum computing has the ability to solve problems tum advantage,’’ PRX Quantum, vol. 2, no. 1, Mar. 2021, Art. no. 010103.
[18] L. M. K. Vandersypen, M. Steffen, G. Breyta, C. S. Yannoni,
that are beyond the capabilities of any classical computer, M. H. Sherwood, and I. L. Chuang, ‘‘Experimental realization of shor’s
a capability referred to as quantum supremacy. Achieving quantum factoring algorithm using nuclear magnetic resonance,’’ Nature,
true quantum supremacy would be a major milestone in the vol. 414, no. 6866, pp. 883–887, Dec. 2001.
[19] S. Gulde, M. Riebe, G. P. T. Lancaster, C. Becher, J. Eschner, H. Häffner,
field of quantum computing. However, with such powerful F. Schmidt-Kaler, I. L. Chuang, and R. Blatt, ‘‘Implementation of the
technology, it also raises issues such as privacy and security. Deutsch–Jozsa algorithm on an ion-trap quantum computer,’’ Nature,
Researchers will need to consider these issues as quantum vol. 421, no. 6918, pp. 48–50, Jan. 2003.
[20] L. DiCarlo, J. M. Chow, J. M. Gambetta, L. S. Bishop,
computing technology grows. B. R. Johnson, D. I. Schuster, J. Majer, A. Blais, L. Frunzio, S. M.
In conclusion, we recognize the remarkable progress in Girvin, and R. J. Schoelkopf, ‘‘Demonstration of two-qubit algorithms
quantum computing and its potential to address complicated with a superconducting quantum processor,’’ Nature, vol. 460, no. 7252,
pp. 240–244, Jul. 2009.
problems with efficiency more than classical computers.
[21] T. Van der Sar, Z. Wang, M. Blok, H. Bernien, T. Taminiau, D. Toyli,
This tutorial is intended to be a valuable resource for those D. Lidar, D. Awschalom, R. Hanson, and V. Dobrovitski, ‘‘Decoherence-
beginning their journey into quantum computing and for protected quantum gates for a hybrid solid-state spin register,’’ Nature,
vol. 484, no. 7392, pp. 82–86, 7392.
researchers to stay informed about this exciting field. The
[22] P. W. Shor, ‘‘Scheme for reducing decoherence in quantum computer
future of quantum computing is promising, and we can memory,’’ Phys. Rev. A, Gen. Phys., vol. 52, no. 4, pp. 2493–2496,
expect to see significant breakthroughs in scientific research, Oct. 1995.
engineering, and technology in the coming years. [23] P. Shor, ‘‘Fault-tolerant quantum computation,’’ in Proc. 37th Conf.
Found. Comput. Sci., 1996, pp. 56–65, doi: 10.1109/SFCS.1996.548464.
[24] E. Knill, R. Laflamme, and W. H. Zurek, ‘‘Resilient quantum computa-
REFERENCES tion,’’ Science, vol. 279, no. 5349, pp. 342–345, Jan. 1998.
[1] R. Barends, J. Kelly, A. Megrant, A. Veitia, D. Sank, E. Jeffrey, [25] J. Preskill, ‘‘Reliable quantum computers,’’ Proc. Roy. Soc. London. Ser.
T. C. White, J. Mutus, A. G. Fowler, and B. Campbell, ‘‘Superconducting A, Math., Phys. Eng. Sci., vol. 454, no. 1969, pp. 385–410, Jan. 1998.
quantum circuits at the surface code threshold for fault tolerance,’’ Nature, [26] F. Gaitan, Quantum Error Correction and Fault Tolerant Quantum
vol. 508, no. 7497, pp. 500–503, 7497. Computing. Boca Raton, FL, USA: CRC Press, 2008.
[2] J. Biamonte, P. Wittek, N. Pancotti, P. Rebentrost, N. Wiebe, and [27] D. A. Lidar and T. A. Brun, Quantum Error Correction. Cambridge, U.K.:
S. Lloyd, ‘‘Quantum machine learning,’’ Nature, vol. 549, no. 7671, Cambridge Univ. Press, 2013.
pp. 195–202, 7671. [28] S. G. Glisic and B. Lorenzo, ‘‘Quantum error correction,’’ in Artificial
[3] S. Debnath, N. M. Linke, C. Figgatt, K. A. Landsman, K. Wright, and Intelligence and Quantum Computing for Advanced Wireless Networks.
C. Monroe, ‘‘Demonstration of a small programmable quantum computer Cham, Switzerland: Springer, 2022, pp. 451–498.
with atomic qubits,’’ Nature, vol. 536, no. 7614, pp. 63–66, Aug. 2016. [29] A. G. Fowler, M. Mariantoni, J. M. Martinis, and A. N. Cleland, ‘‘Surface
[4] N. Ofek, A. Petrenko, R. Heeres, P. Reinhold, Z. Leghtas, B. Vlastakis, codes: Towards practical large-scale quantum computation,’’ Phys. Rev.
Y. Liu, L. Frunzio, S. Girvin, and L. Jiang, ‘‘Extending the lifetime of a A, Gen. Phys., vol. 86, Sep. 2012, Art. no. 032324.
quantum bit with error correction in superconducting circuits,’’ Nature, [30] F. Arute, K. Arya, R. Babbush, D. Bacon, J. C. Bardin, R. Barends,
vol. 536, no. 7617, pp. 441–445, Aug. 2016. R. Biswas, S. Boixo, F. G. Brandao, and D. A. Buell, ‘‘Quantum
[5] P. W. Shor, ‘‘Polynomial-time algorithms for prime factorization and supremacy using a programmable superconducting processor,’’ Nature,
discrete logarithms on a quantum computer,’’ SIAM Rev., vol. 41, no. 2, vol. 574, no. 7779, pp. 505–510, 7779.
pp. 303–332, Jan. 1999. [31] H.-S. Zhong, H. Wang, Y.-H. Deng, M.-C. Chen, L.-C. Peng, Y.-H. Luo,
[6] L. G. S. Imre, Advanced Quantum Communications: An Engineering J. Qin, D. Wu, X. Ding, and Y. Hu, ‘‘Quantum computational advantage
Approach. Hoboken, NJ, USA: Wiley, 2013. using photons,’’ Science, vol. 370, no. 6523, pp. 1460–1463, Dec. 2020.
[7] R. L. Rivest, A. Shamir, and L. Adleman, ‘‘A method for obtaining digital [32] J. Chow, O. Dial, and J. Gambetta. IBM Quantum Breaks the 100-
signatures and public-key cryptosystems,’’ Commun. ACM, vol. 21, no. 2, Qubit Processor Barrier. Accessed: Dec. 10, 2023. [Online]. Available:
pp. 120–126, Feb. 1978. https://research.ibm.com/blog/127-qubit-quantum-processor-eagle
[8] P. W. Shor, ‘‘Polynomial-time algorithms for prime factorization [33] L. S. Madsen, F. Laudenbach, M. F. Askarani, F. Rortais, T. Vincent,
and discrete logarithms on a quantum computer,’’ SIAM J. Com- J. F. Bulmer, F. M. Miatto, L. Neuhaus, L. G. Helt, and M. J. Collins,
put., vol. 26, no. 5, pp. 1484–1509, 1997. [Online]. Available: ‘‘Quantum computational advantage with a programmable photonic
http://dx.doi.org/10.1137/S0097539795293172 processor,’’ Nature, vol. 606, no. 7912, pp. 75–81, Jun. 2022.
[9] J. Proos and C. Zalka, ‘‘Shor’s discrete logarithm quantum algorithm for [34] J. Preskill, ‘‘Quantum computing in the NISQ era and beyond,’’ Quantum,
elliptic curves,’’ 2004, arXiv:ph/030114. vol. 2, p. 79, Aug. 2018.
[10] T. Albash and D. A. Lidar, ‘‘Adiabatic quantum computation,’’ Rev. [35] B. Schumacher, ‘‘Quantum coding,’’ Phys. Rev. A, Gen. Phys., vol. 51,
Modern Phys., vol. 90, no. 1, 2018, Art. no. 015002. no. 4, pp. 2738–2747, 1995, doi: 10.1103/PhysRevA.51.2738.

22312 VOLUME 12, 2024

M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

[36] P. A. M. Dirac, ‘‘A new notation for quantum mechanics,’’ Math. Proc. [63] S. McArdle, T. Jones, S. Endo, Y. Li, S. C. Benjamin, and
Cambridge Phil. Soc., vol. 35, no. 3, pp. 416–418, 1939. X. Yuan, ‘‘Variational ansatz-based quantum simulation of imag-
[37] J. C. Solem and L. C. Biedenharn, ‘‘Understanding geometrical phases inary time evolution,’’ npj Quantum Inf., vol. 5, no. 1, p. 75,
in quantum mechanics: An elementary example,’’ Found. Phys., vol. 23, Sep. 2019.
no. 2, pp. 185–195, Feb. 1993. [64] Z. Jiang, J. McClean, R. Babbush, and H. Neven, ‘‘Majorana loop
[38] V. Vedral and M. B. Plenio, ‘‘Basics of quantum computation,’’ Prog. stabilizer codes for error mitigation in fermionic quantum simulations,’’
Quantum Electron., vol. 22, no. 1, pp. 1–39, Jan. 1998. Phys. Rev. Appl., vol. 12, no. 6, Dec. 2019, Art. no. 064041.
[39] J. L. Hevia, G. Peterssen, C. Ebert, and M. Piattini, ‘‘Quantum [65] M. Mohseni, P. Read, H. Neven, S. Boixo, V. Denchev, R. Babbush,
computing,’’ IEEE Softw., vol. 38, no. 5, pp. 7–15, Sep. 2021, doi: A. Fowler, V. Smelyanskiy, and J. Martinis, ‘‘Commercialize quantum
10.1109/MS.2021.3087755. technologies in five years,’’ Nature, vol. 543, no. 7644, pp. 171–174,
[40] P. Aradyamath and R. Ujjinimatad, ‘‘Quantum computing concepts with Mar. 2017.
deutsch jozsa algorithm,’’ JOIV, Int. J. Informat. Visualizat., vol. 3, no. 1, [66] H. Chen, L. Wossnig, S. Severini, H. Neven, and M. Mohseni, ‘‘Universal
pp. 59–68, Jan. 2019. discriminative quantum neural networks,’’ Quantum Mach. Intell., vol. 3,
[41] R. De, R. Moberly, C. Beery, J. Juybari, and K. Sundqvist, ‘‘Multi-qubit no. 1, pp. 1–11, Jun. 2021.
size-hopping deutsch-jozsa algorithm with qubit reordering for secure [67] G. Verdon, M. Broughton, J. R. McClean, K. J. Sung, R. Babbush,
quantum key distribution,’’ in Proc. IEEE Int. Conf. Quantum Comput. Z. Jiang, H. Neven, and M. Mohseni, ‘‘Learning to learn with quantum
Eng. (QCE), Feb. 2021, pp. 473–474. neural networks via classical neural networks,’’ 2019, arXiv:1907.05415.
[42] A. N. Oliveira, E. V. B. de Oliveira, A. C. Santos, and C. J. Villas-Boas, [68] R. Sweke, F. Wilde, J. Meyer, M. Schuld, P. K. Faehrmann, B. Meynard-
‘‘Quantum algorithms in IBMQ experience: Deutsch-jozsa algorithm,’’ Piganeau, and J. Eisert, ‘‘Stochastic gradient descent for hybrid quantum-
2021, arXiv:2109.07910. classical optimization,’’ Quantum, vol. 4, p. 314, Aug. 2020.
[43] A. Gupta, P. Ghosh, K. Sen, and U. Sen, ‘‘Effects of noise on performance [69] M. Broughton, G. Verdon, T. McCourt, A. J. Martinez, J. H. Yoo,
of bernstein-vazirani algorithm,’’ 2023, arXiv:2305.19745. S. V. Isakov, P. Massey, R. Halavati, M. Y. Niu, and A. Zlokapa,
[44] P. Fernández and M. A. Martin-Delgado, ‘‘Homomorphic encryption of ‘‘TensorFlow quantum: A software framework for quantum machine
the K=2 bernstein-vazirani algorithm,’’ 2023, arXiv:2303.17426. learning,’’ 2020, arXiv:2003.02989.
[45] X. Zhou, D. Qiu, and L. Lou, ‘‘Distributed exact quantum algorithms for [70] J. Liu, K. H. Lim, K. L. Wood, W. Huang, C. Guo, and H.-L. Huang,
bernstein-vazirani and search problems,’’ 2023, arXiv:2303.10670. ‘‘Hybrid quantum-classical convolutional neural networks,’’ Sci. China
[46] X. Cheng, Y. Xu, K. Wang, Y. Zhang, B. Li, and Z. Zhang, ‘‘Lightweight Phys., Mech. Astron., vol. 64, no. 9, 2021, Art. no. 290311.
and flexible hardware implementation of authenticated encryption [71] S. Yen-Chi Chen, T.-C. Wei, C. Zhang, H. Yu, and S. Yoo,
algorithm SIMON-galois/counter mode,’’ Int. J. Circuit Theory Appl., ‘‘Hybrid quantum-classical graph convolutional network,’’ 2021,
vol. 51, no. 12, pp. 5951–5961, Dec. 2023. arXiv:2101.06189.
[47] X. Bonnetain, ‘‘Tight bounds for simon’s algorithm,’’ in Proc. 7th [72] Y. Liang, W. Peng, Z.-J. Zheng, O. Silvén, and G. Zhao, ‘‘A hybrid
Int. Conf. Cryptol. Inf. Secur. Latin Amer., Bogota, Colombia. Cham, quantum–classical neural network with deep residual learning,’’ Neural
Switzerland: Springer, 2021, pp. 3–23. Netw., vol. 143, pp. 133–147, Nov. 2021.
[48] P. W. Shor, ‘‘Algorithms for quantum computation: Discrete logarithms [73] Qiskit Contributors, ‘‘Qiskit: An Open-Source Framework for Quantum
and factoring,’’ in Proc. 35th Annu. Symp. Found. Comput. Sci., 1994, Computing,’’ 2023, doi: 10.5281/zenodo.2573505.
pp. 124–134. [74] P. D. Nation and J. R. Johansson. Qutip. Accessed: Dec. 25, 2023.
[49] J. P. Buhler, H. W. Lenstra, and C. Pomerance, ‘‘Factoring integers with [Online]. Available: https://qutip.org/
the number field sieve,’’ in The Development of the Number Field Sieve. [75] Cirq. Developers, Cirq. Accessed: Jan. 2, 2024. [Online]. Available:
Berlin, Germany: Springer, 1993, pp. 50–94. https://github.com/quantumlib/Cirq
[50] L. K. Grover, ‘‘A fast quantum mechanical algorithm for database [76] Projectq—An Open Source Software Framework for Quantum
search,’’ in Proc. 8th Annu. ACM Symp. Theory Comput., 1996, Computing. Accessed: Dec. 27, 2023. [Online]. Available:
pp. 212–219. https://github.com/ProjectQ-Framework/ProjectQ
[51] M. A. Nielsen and I. L. Chuang, Quantum Computation and Quantum [77] R. S. Smith, M. J. Curtis, and W. J. Zeng, ‘‘A practical quantum instruction
Information. Cambridge, U.K.: Cambridge Univ. Press, 2010. set architecture,’’ 2016, arXiv:1608.03355.
[52] K. P. Murphy, Machine Learning: A Probabilistic Perspective. [78] X. Y. Jin, K. Cicak, Z. Parrott, S. Kotler, F. Lecocq, J. Teufel,
Cambridge, MA, USA: MIT Press, 2012. J. Aumentado, E. Kapit, and R. W. Simmonds, ‘‘Versatile parametric
[53] Y. LeCun, Y. Bengio, and G. Hinton, ‘‘Deep learning,’’ Nature, vol. 521, coupling between two statically decoupled transmon qubits,’’ 2023,
no. 7553, pp. 436–444, May 2015. arXiv:2305.02907.
[54] A. Krizhevsky, I. Sutskever, and G. E. Hinton, ‘‘ImageNet classification [79] R. Newrock, C. Lobb, U. Geigenmuller, and M. Octavio, ‘‘The two-
with deep convolutional neural networks,’’ in Proc. Adv. Neural dimensional physics of Josephson-junction arrays,’’ Solid State Physics-
Inf. Process. Syst., vol. 25, F. Pereira, C. J. Burges, L. Bottou, and New York-Academic Press, vol. 54, pp. 266–512, Jan. 2000.
K. Q. Weinberger, Eds. Curran Associates, 2012. [Online]. Available: [80] H. Dang, R. Zha, J. Tan, T. Zhang, J. Li, N. Li, B. Zhao, Y. Zhao, H. Tan,
https://proceedings.neurips.cc/paper_files/paper/2012/file/c399862d3b9 and R. Xue, ‘‘Investigations on a 3.3 K four-stage stirling-type pulse tube
d6b76c8436e924a68c45b-Paper.pdf cryocooler. Part B: Experimental verifications,’’ Cryogenics, vol. 105,
[55] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. Cambridge, Jan. 2020, Art. no. 103015.
MA, USA: MIT Press, 2016. [81] H. Rana and N. Verma, ‘‘Enhanced quantum key distribution using hybrid
[56] E. Farhi, J. Goldstone, and S. Gutmann, ‘‘A quantum approximate channels and natural random numbers,’’ 2020, arXiv:2007.14298.
optimization algorithm,’’ 2014, arXiv:1411.4028. [82] S. Adhikary, S. Dangwal, and D. Bhowmik, ‘‘Supervised learning with
[57] P. Shor, ‘‘Algorithms for quantum computation: Discrete logarithms and a quantum classifier using multi-level systems,’’ Quantum Inf. Process.,
factoring,’’ in Proc. 35th Annu. Symp. Found. Comput. Sci. Los Alamitos, vol. 19, no. 3, pp. 1–12, Mar. 2020.
CA, USA: IEEE Computer Society, 1994, pp. 124–134. [83] G. Sergioli, G. Russo, E. Santucci, A. Stefano, S. E. Torrisi, S. Palmucci,
[58] S. Lloyd, M. Mohseni, and P. Rebentrost, ‘‘Quantum principal component C. Vancheri, and R. Giuntini, ‘‘Quantum-inspired minimum distance
analysis,’’ Nat. Phys., vol. 10, no. 9, pp. 631–633, 2014. classification in a biomedical context,’’ Int. J. Quantum Inf., vol. 16, no. 8,
[59] P. Rebentrost, M. Mohseni, and S. Lloyd, ‘‘Quantum support vector Dec. 2018, Art. no. 1840011.
machine for big data classification,’’ Phys. Rev. Lett., vol. 113, no. 13, [84] G. Hellstem, ‘‘Hybrid quantum network for classification of finance and
Sep. 2014, Art. no. 130503. MNIST data,’’ in Proc. IEEE 18th Int. Conf. Softw. Archit. Companion
[60] I. Kerenidis and A. Prakash, ‘‘Quantum recommendation systems,’’ 2016, (ICSA-C), Mar. 2021, pp. 1–4.
arXiv:1603.08675. [85] A. P. Ngo, N. Le, H. T. Nguyen, A. Eroglu, and D. T. Nguyen,
[61] B. Duan, J. Yuan, C.-H. Yu, J. Huang, and C.-Y. Hsieh, ‘‘A survey on HHL ‘‘A quantum neural network regression for modeling lithium-ion battery
algorithm: From theory to application in quantum machine learning,’’ capacity degradation,’’ in Proc. IEEE Green Technol. Conf. (GreenTech),
Phys. Lett. A, vol. 384, no. 24, Aug. 2020, Art. no. 126595. Apr. 2023, pp. 164–168.
[62] L. Zhou, S.-T. Wang, S. Choi, H. Pichler, and M. D. Lukin, ‘‘Quantum [86] T. Kanimozhi, S. Sridevi, M. Valliammai, J. Mohanraj, and V. Kumar,
approximate optimization algorithm: Performance, mechanism, and ‘‘Quantum regression model for the prediction of surface plasmon
implementation on near-term devices,’’ Phys. Rev. X, vol. 10, no. 2, resonance sensor behaviour,’’ in Proc. Workshop Recent Adv. Photon.
Jun. 2020, Art. no. 021067. (WRAP), 2022, pp. 1–2.

VOLUME 12, 2024 22313

 M. A. Shafique et al.: Quantum Computing: Circuits, Algorithms, and Applications

[87] S. Rainjonneau, I. Tokarev, S. Iudin, S. Rayaprolu, K. Pinto, ARSLAN MUNIR (Senior Member, IEEE)
D. Lemtiuzhnikova, M. Koblan, E. Barashov, M. Kordzanganeh, received the M.A.Sc. degree in ECE from the
M. Pflitsch, and A. Melnikov, ‘‘Quantum algorithms applied to satellite University of British Columbia (UBC), Vancouver,
mission planning for Earth observation,’’ IEEE J. Sel. Topics Appl. Earth Canada, in 2007, and the Ph.D. degree in
Observ. Remote Sens., pp. 1–13, 2023. ECE from the University of Florida (UF),
[88] N. F. Bar, H. Yetis, and M. Karakose, ‘‘An approach based on quantum Gainesville, FL, USA, in 2012. He is currently
reinforcement learning for navigation problems,’’ in Proc. Int. Conf. Data
an Associate Professor with the Department
Analytics Bus. Ind. (ICDABI), Oct. 2022, pp. 593–597.
of Computer Science, Kansas State University.
[89] E. Dri, A. Aita, T. Fioravanti, G. Franco, E. Giusto, G. Ranieri,
D. Corbelletto, and B. Montrucchio, ‘‘Towards an end-to-end approach He was a Postdoctoral Research Associate with
for quantum principal component analysis,’’ in Proc. IEEE Int. Conf. the Electrical and Computer Engineering (ECE)
Quantum Comput. Eng. (QCE), vol. 2, Sep. 2023, pp. 1–6. Department, Rice University, Houston, TX, USA, from May 2012 to June
[90] S. Lok Tsang, M. T. West, S. M. Erfani, and M. Usman, ‘‘Hybrid 2014. From 2007 to 2008, he was a Software Development Engineer with
quantum-classical generative adversarial network for high-resolution the Embedded Systems Division, Mentor Graphics Corporation. His current
image generation,’’ IEEE Trans. Quantum Eng., vol. 4, pp. 1–19, 2023. research interests include embedded and cyber-physical systems, secure
[91] B. Jaderberg, L. W. Anderson, W. Xie, S. Albanie, M. Kiffner, and and trustworthy systems, parallel computing, quantum computing, artificial
D. Jaksch, ‘‘Quantum self-supervised learning,’’ Quantum Sci. Technol., intelligence, and computer vision. He received many academic awards,
vol. 7, no. 3, Jul. 2022, Art. no. 035005. including the doctoral fellowship from Natural Sciences and Engineering
[92] A. Montanaro, ‘‘Quantum algorithms: An overview,’’ NPJ Quantum Inf., Research Council (NSERC) of Canada. He earned gold medals for best
vol. 2, no. 1, pp. 1–8, Jan. 2016. performance in electrical engineering, gold medals, and academic roll of
[93] V. Hassija, V. Chamola, A. Goyal, S. S. Kanhere, and N. Guizani,
honor for securing rank one in pre-engineering provincial examinations (out
‘‘Forthcoming applications of quantum computing: Peeking into the
future,’’ IET Quantum Commun., vol. 1, no. 2, pp. 35–41, Dec. 2020. of approximately 300,000 candidates).
[94] Y. Wang, J. E. Kim, and K. Suresh, ‘‘Opportunities and challenges of
quantum computing for engineering optimization,’’ J. Comput. Inf. Sci.
Eng., vol. 23, no. 6, Dec. 2023.
[95] H. Makhanov, K. Setia, J. Liu, V. Gomez-Gonzalez, and
G. Jenaro-Rabadan, ‘‘Quantum computing applications for flight
trajectory optimization,’’ 2023, arXiv:2304.14445.
[96] A. Pyrkov, A. Aliper, D. Bezrukov, Y.-C. Lin, D. Polykovskiy, P.
Kamya, F. Ren, and A. Zhavoronkov, ‘‘Quantum computing for near-
term applications in generative chemistry and drug discovery,’’ Drug
Discovery Today, vol. 28, no. 8, Aug. 2023, Art. no. 103675.
[97] V. Mahesh and S. Shijo, ‘‘Accelerating drug discovery with quantum
computing,’’ in Evolution and Applications of Quantum Computing.
Wiley, 2023, pp. 175–181.
[98] P.-H. Wang, J.-H. Chen, Y.-Y. Yang, C. Lee, and Y. J. Tseng, ‘‘Recent
advances in quantum computing for drug discovery and development,’’
IEEE Nanotechnol. Mag., vol. 17, no. 2, pp. 26–30, Apr. 2023.
[99] M. Avramouli, I. K. Savvas, A. Vasilaki, and G. Garani, ‘‘Unlocking
the potential of quantum machine learning to advance drug discovery,’’
Electronics, vol. 12, no. 11, p. 2402, May 2023.
[100] H. Xiong, X. Duan, Y. Yu, J. Zhang, and H. Yin, ‘‘Image classification IMRAN LATIF received the master’s degree in
based on quantum machine learning,’’ in Proc. 5th Int. Conf. Intell. engineering from The City University of New
Control, Meas. Signal Process. (ICMSP), 2023, pp. 891–895. York. He is currently the Chief Operations Officer
[101] H. M. Gray and K. Terashi, ‘‘Quantum computing applications in future of the U.S. Department of Energy, Office of
colliders,’’ Frontiers Phys., vol. 10, May 2022, Art. no. 864823. Science, Brookhaven National Laboratory, where
[102] K. A. Tychola, T. Kalampokas, and G. A. Papakostas, ‘‘Quantum machine
he leads the infrastructure operations of high-
learning—An overview,’’ Electronics, vol. 12, no. 11, p. 2379, 2023.
[103] Y.-J. Chang, M.-F. Sie, S.-W. Liao, and C.-R. Chang, ‘‘The prospects performance computing centers supporting global
of quantum computing for quantitative finance and beyond,’’ IEEE cutting-edge research collaborations in physics,
Nanotechnol. Mag., vol. 17, no. 2, pp. 31–37, Apr. 2023. life sciences, quantum and computational sci-
ences, and artificial intelligence. He has over
20 years of leadership experience delivering large-scale engineering,
MUHAMMAD ALI SHAFIQUE received the construction, and sustainability projects with great success for top com-
Bachelor of Science and master’s degrees from panies, including IBM, Microsoft, Google, and Wall Street District. He is
the University of Engineering and Technol- also an Invited Speaker and has delivered expert talks at various tech
ogy, Lahore, in 2015 and 2017, respectively. and entrepreneur conferences worldwide, including DCD, BICSI, Super
He is currently pursuing the Ph.D. degree with Computing, HEPIX, and CHEP. He holds professional engineering licenses
the Department of Computer Science, Kansas in the states of New York and Texas along with several professional
State University. He also holds the position of a certifications, including CEM, PMP, DCEP, CDCDP, and LEED. He is also a
Research Assistant with the ISCAAS Laboratory. Mentor of the U.S. Federal Government-funded internship programs, where
His research interest includes efficient machine he works with students from various universities on leading-edge scientific
learning in resource-constrained systems. He is research projects. He is also spearheading global research initiatives with
committed to conducting research in this area with the aim of advancing the leading academia and big tech companies on data center energy conservation.
current knowledge and understanding of these fields.

22314 VOLUME 12, 2024