Quantum Computers and Quantum Secure Communication
Uploaded by
Antra PramanikQuantum Computers and Quantum Secure Communication
Uploaded by
Antra PramanikQuantum Computers and
Quantum Secure Communications
SoSe20
950488885
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Goal Today‘s Lecture
Presentation and Motivation
Description
Organization of the Course
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 2
Section 1
PRESENTATION AND MOTIVATION
Q(CaSC) / Johanna Sepúlveda 3
Motivation
Johanna Sepúlveda
UNIVERSITY OF SOUTH BRITTANY, France
2013: Post-Doctoral Researcher
UNIVERSITY OF LYON, France
NATIONAL UNIVERSITY OF COLOMBIA, Colombia 2014: Post-Doctoral Researcher
2004: Diploma Summa Cum Laude in Electrical Engineering
(1st place)
TECHNICAL UNIVERSITY OF MUNICH, Germany
2015: Post-Doctoral Researcher
AIRBUS DEFENCE AND SPACE, Germany
AREAS OF RESEARCH: 2019: Senior Scientist Secure Communication
Expert on Post-Quantum Security
• Embedded system design
• System-on-Chip (SoC) design
• Embedded Security
• Multi-processor SoCs (MPSoCs) UNIVERSITY OF SÃO PAULO, Brazil
security (Network-on-Chip) Polytechnic School
2006: M.Sc. in Electrical Engineering (1st place)
• New technologies for MPSoCs
2011: Ph.D. in Electrical Engineering – Microelectronics (1st place)
2012: Post-Doctoral Researcher/ External Professor
• MPSoC security for critical systems
• Post-quantum security
Q(CaSC) / Johanna Sepúlveda 4
Motivation
Johanna Sepúlveda
More than 9 years in the research area of Post-Quantum
security
Build the TUM Post-Quantum security area since 2015
Habilitation at TUM
“Secure and Efficient Post-quantum Security”
TUM lecturer of Quantum Computing and Quantum Secure
Communication
Q(CaSC) / Johanna Sepúlveda 5
Motivation
Post-Quantum together with my TUM Students
(Master thesis Advisor)
Best Paper award VLSISoC
2018
More than 20 publications on PQC
Walter Gademann Prize for the best Mint Prize for the best Master Thesis 2019
Master Thesis at TUM Post-Quantum
(2018) Q(CaSC) / Johanna Sepúlveda 6
Motivation
Post-Quantum together with my TUM Students
Q(CaSC) / Johanna Sepúlveda 7
Section 1
FIRST CONTACT
Q(CaSC) / Johanna Sepúlveda 8
Motivation
Topics of the Lecture: Quantum Computers
Q(CaSC) / Johanna Sepúlveda 9
Motivation
Topics of the Lecture: Security Threat
Today traditional
Factoring large integers
cryptography
Computing discrete logarithms
is considered secure
Traditional Cryptography
(RSA, ECC)
Cryptoanalysis
Side-Channel
Attacker
Quantum
In the future, traditional public key Threat
cryptography
will be broken by large Quantum
computers
Q(CaSC) / Johanna Sepúlveda 10
Motivation
Topics of the Lecture: QKD and PQC
RSA ECC
Key Length qubits Key Length qubits
1024 2048 163 1000
140
2048 4096 224 1300
3072 6144 256 1500
120
4096 8192 383 2300
15360 30720 Superconductors
512 3000
100
Quantum secure communications are required Ion Trap
Spin
PHYSICAL QUBITS
80
1. Quantum Key Distribution (QKD)
Neutral atoms
60
Quantum Channels
40 2. Post-Quantum Cryptography (PQC)
20
Traditional Infrastructure
0
1995 2000 2005 2010 2015 2020 2025
Physical qubits ≠ Operational qubits
Q(CaSC) / Johanna Sepúlveda 11
Topics of the Lecture: PQC
Post-Quantum Cryptography Distribute
Key
Classical Network Encrypt
Decrypt
Sign
Verify
Algorithms used to secure messages on a classical computer and that are
resistant to traditional computer attacks AND quantum computer attacks
Code Hash Multivariate Isogeny Lattice
Q(CaSC) / Johanna Sepúlveda 12
Long Term Security
0 5 10 15 20 25 30
Life cycle of the product
(Years)
Expose industry and
products to attacks
Post-Quantum Security
Integration Process IonQ’s 79 qubits Quantum Processor
Q(CaSC) / Johanna Sepúlveda 13
Motivation
Topics of the Lecture: In a nut shell
Quantum Computer
Quantum Secure Communication
QKD PQC
Quantum-Key Distribution Post-Quantum Cryptography
Q(CaSC) / Johanna Sepúlveda 14
Motivation
Quantum Computers and Quantum Secure
Communications
Standardization efforts
Be part of security history
Interesting topic
Excellent experts are required
Q(CaSC) / Johanna Sepúlveda 15
Motivation
Airbus: More than Aircrafts
Since 2019
Q(CaSC) / Johanna Sepúlveda 16
Section 2
DESCRIPTION
Q(CaSC) / Johanna Sepúlveda 17
Description
Learning Objective
At the end of this module, students
1) are able to understand the basic concepts of quantum computers and quantum
secure communication;
2) are familiar with quantum and post-quantum algorithms;
3) understand the vulnerabilities of post-quantum cryptography implementations;
4) can evaluate the security and performance of Post-Quantum cryptographic
implementations; and
5) can apply security countermeasures to post-quantum cryptographic
implementations.
Theory
Q(CaSC) / Johanna Sepúlveda 18
Description
Ways of working: QCQSC
Lecture
Physical: At TUM (face to face)
Virtual: Lectures will have a video, so you can attend asynchronous
(look the lecture when you want)
Virtual: Some synchronous session for discussion (scheduled)
Videos of the lecture (released weekly)
• Available on the Moodle page of the course; and
• ONLY for the enrolled students (do not share without permission)
Additional material (Moodle)
• Articles, exercises, links
Q(CaSC) / Johanna Sepúlveda 19
Description
Ways of working: QCQSC
Lecture Organization (3 hour lecture)
Forum (Moodle):
The topic of the
week, share news
Presentation
Live streaming
(Google Hangout)
Reading articles
Using quantum emulators and online tools (e.g., IBM)
Answering questions
Watch additional educational material
* Time distribution of the activities may change according to the topic of the lecture
Q(CaSC) / Johanna Sepúlveda 20
Description
Ways of working: QCQSC
Tools for our communication
Lecture is open to
Email list of the course and Contact to me suggestions: Together
we make the course
Interactive questions
Course forum and repository
https://www.menti.com/
Q(CaSC) / Johanna Sepúlveda 21
Description
Contact
Johanna Sepúlveda
Email: [email protected]
(Questions, requests, schedule appointment)
Expected answer from 10 min to 7 hrs
24h/7
Q(CaSC) / Johanna Sepúlveda 22
Description
Grading (5 credits)
Final exam = 100% of the grade
Oral if we are less than 15 students
Otherwise will be written
OPTIONAL:
Bonus points (0.3)
Exercises and Demonstrations
Presentation/Poster (15 min)
• Topic defined by myself
• Student should request the topic until May 20th
Q(CaSC) / Johanna Sepúlveda 23
Description
Requirements
Willing to learn
Basic concepts of security
• Symmetric/asymmetric cryptography, hash
Basic concepts on embedded software programming
Basic knowledge of computer architecture
• Components, memory hierarchy, communication
structure
OPTIONAL
Advance knowledge of computer architecture
Optimization techniques for embedded programming
Secure implementation of cryptographic algorithms
Q(CaSC) / Johanna Sepúlveda 24
Description
References
Online books
• Post-Quantum Cryptography
Daniel Bernstein, Johannes Buchmann, Erik Dahmen
• An Introduction to Mathematical Cryptography
Jeffrey Hoffstein, Jill Pipher, Joseph Silverman, ISBN: 978-1-4939-1711
Online Articles (Springer, IEEE, IACR)
Videos
NIST (https://csrc.nist.gov/projects/post-quantum-cryptography)
ETSI (https://www.etsi.org/technologies/quantum-key-distribution)
Q(CaSC) / Johanna Sepúlveda 25
Further Reading and Assignments
They will be pointed at the end of each lecture
Optional
Mandatory
Goal of the further reading activities / exercises
Prepare for the next lecture; or
Complement the current lecture; or
Practical experience of the learnt during the lecture
Polls and on-line questionnaires (optional but recommended)
Give bonus points
Prepare to the final exam
Q(CaSC) / Johanna Sepúlveda 26
Section 3
ORGANIZATION OF THE COURSE
Q(CaSC) / Johanna Sepúlveda 27
Organization
Blocks of the Course
Quantum Quantum Post-Quantum
Computers Communication Cryptography
(QKD)
Basic Principles Main Principles
Evolution Main Principles Families and Types
Structure Structure Main Components
Types Types Standard
Programming Standard Secure implementation
Security threat Use cases Use cases
Q(CaSC) / Johanna Sepúlveda 28
Section 4
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 29
Feedback of the lecture (Menti)
Q(CaSC) / Johanna Sepúlveda 30
Summary
Today we:
Have our first contact and now we are connected
Discuss the learning objective, ways of working and grading
of the lecture
We are ready to start … Quantum
Computers
Basic Principles
Next lecture: Quantum Computer principles Evolution
Structure
Types
Programming
Security threat
Q(CaSC) / Johanna Sepúlveda 31
Recommended Material (optional)
Basic knowledge to prepare for the next lecture (Recommended for the students
that have no heard before quantum computers)
Introduction to Quantum Computers
Friendly Lecture (Pages 1 -23)
Approx. 30-45 min reading
https://courses.csail.mit.edu/6.857/2019/files/NAE-report-on-quantum-computing.pdf
Q(CaSC) / Johanna Sepúlveda
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
2. Quantum Computer Basics
Q(CaSC) / Johanna Sepúlveda 2
Goal Today‘s Lecture
Summary of Previous Lecture
Quantum Computer Need
Basic Quantum Principles
Quantum Computer Principles
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 3
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 4
Summary
L1: Introduction – Learning Objectives
UNDERSTAND
QKD
Quantum-Key Distribution
FAMILIAR
PQC
Quantum Computer Post-Quantum Cryptography
UNDERSTAND (vulnerabilities)
EVALUATE (security and performance)
APPLY (security countermeasures)
5
Q(CaSC) / Johanna Sepúlveda 5
We are Ready to Start..
Quantum
Computers
Basic Principles
Evolution
Structure
Types
Programming
Security threat
Q(CaSC) / Johanna Sepúlveda 6
Section 2
THE BEGINNING:
QUANTUM COMPUTER NEED
Q(CaSC) / Johanna Sepúlveda 7
Need
Motivation
Computer Organization
Memories
Processor(s)
Communication
Software
structure
(buses, networks)
Interfaces
Peripherals
Machine used to process, store and communicate information
Q(CaSC) / Johanna Sepúlveda 8
Need
Motivation
Computer Organization
Processor
Datapath Memory
Bus
Registers
Address
Control Data
Unit Control
ALU
Interfaces Co-processors
Q(CaSC) / Johanna Sepúlveda 9
Need
Motivation
Integrated Circuits
Transistor Wires
Gates
Q(CaSC) / Johanna Sepúlveda 10
Need
Motivation
Systems-on-Chips (SoCs): Small
Q(CaSC) / Johanna Sepúlveda 11
Need
Motivation
Systems-on-Chips (SoCs): Medium
Q(CaSC) / Johanna Sepúlveda 12
Need
Motivation
Systems-on-Chips (SoCs): Large
Hardware Software
Computation
Structure
Processors,
Memories,
Peripherals
Communication
Structure
Network-on-Chip (NoC) Tile-MX100(www.eetimes.com)
Q(CaSC) / Johanna Sepúlveda 13
Need
Motivation
Moore Law (1965)
The number of transistors in a chip doubles almost every two years
Since 2014, it doubles every three years
Q(CaSC) / Johanna Sepúlveda 14
Need
Motivation
More than Moore
Reshaping New SoC organization
Transistors
FinFet
Silicon nanosheet
New
Technologies
3D SoC Stacking
Carbon Nanotubes
Q(CaSC) / Johanna Sepúlveda 15
Shulaker et al. Monolithic 3D integration: A path from concept to reality. DATE 2015
Need
Motivation
Evolution of Computation
Classical and Quantum computers will co-exist
Sycamore
2019
Q(CaSC) / Johanna Sepúlveda 16
Need
Motivation
Evolution of Computation
Processors
P (e.g. Intel AVX)
Neuromorphic
ASIC ASIC u Microcontroller cPros: Processors
(Application Specific Integrated Circuit) C ▪ Programmable (e.g. Intel Loihi)
(e.g. ARM Cortex-M) ▪ Low unit cost Pros:
Pros: ▪ Manycore
Pros: architecture ▪ Programmable
▪ Fast and easy development ▪ On-memory
▪ Low unit cost ▪ Advance features
▪ High performance (e.g., Out-of-order processing
▪ Wide variety
▪ Low area/power operation) Cons:
Cons: ▪ General purpose
Cons: ▪ Application
▪ Very expensive Cons:
▪ Time-consuming ▪ Specific and limited ISA specific
▪ Specific and fixed
FPGAs Microcontroller for ISA GPUs
(e.g. Nvidia)
(Field Programmable
Gate Array) SAFE
uC
safety
Pros: (e.g. AURIX of Infineon) Pros:
▪ Programmable Pros: ▪ High performance
configuration ▪ Massively parallel
▪ Parallel architecture ▪ Easy development
▪ Predictable behavior operations
▪ Fast development ▪ Safe-oriented design
Cons: ▪ Include security modules (HSM) Cons:
▪ Poor control over
optimization
Cons: ▪ High cost
▪ Specific and fixed ISA ▪ Black-box
▪ High power consumption ▪ Application oriented design
▪ Expensive in large-scale compilers
production ▪ Performance
▪ Volatile depends on nice
Q(CaSC) / Johanna Sepúlveda 17
code behavior flow
Need
Motivation
New Ways of Computation
Q(CaSC) / Johanna Sepúlveda 18
Need
Motivation
MIT Physics and Computation 1981
Q(CaSC) / Johanna Sepúlveda 19
Need
Motivation
Quantum Computer: Milestones
Richard Feynman
Proposed a quantum model to perform computations
❑ Quantum circuits:
✔ Quantum gates can be combined
✔ Able to perform computation
David Deutsch
Q(CaSC) / Johanna Sepúlveda 20
Section 3
QUANTUM PRINCIPLES
Q(CaSC) / Johanna Sepúlveda 21
Principles
Quantum Physics Principles
1. Wave–Particle Duality
2. Uncertainty Principle of Heisenberg
3. Schrödinger Equation
Q(CaSC) / Johanna Sepúlveda 22
Principles
Wave-Particle Duality (1905)
Experiment of Young Photoelectric Effect
(Wave nature of light) (Particle nature of light)
E= h. v E= m. c2
Q(CaSC) / Johanna Sepúlveda 23
Principles
Uncertainty Principle of Heisenberg (1927)
The position and the velocity of an object cannot both be measured exactly,
at the same time
Velocity= V
Velocity=?
Measure has a strong effect on the quantum system
Q(CaSC) / Johanna Sepúlveda 24
Principles
Uncertainty Principle of Heisenberg (1927)
The position and the velocity of an object cannot both be measured exactly,
at the same time
Planck Constant
Position Uncertainty
Momentum Uncertainty
Tunneling effect
(p=m.v)
There is a probability that the electron
traverse the potential barrier and that
escapes (transmitted)
Q(CaSC) / Johanna Sepúlveda 25
Principles
Schrödinger Equation (1925)
Function that describes the state of a quantum-mechanical system
Conditions that the electron must meet in order that a system is stable
: Electronic probability density
Orbitals:
Described by quantum numbers (n l ml ms)
✔ n: Principal quantum number (energy level) n=[1,∞)
✔ l: Azimuthal quantum number (angular momentum) l=[0,n-1]
s, p, d, f..
✔ ml: Magnetic quantum number (orientation) ml=[-l,l]
✔ ms: Spin quantum number (magnetic field orientation) ms=[-1/2,1/2]
Q(CaSC) / Johanna Sepúlveda 26
Principles
Quantum Phenomena
1. Superposition
2. Entanglement
Q(CaSC) / Johanna Sepúlveda 27
Principles
Superposition
The state of a system is determined by the value of a set of
macroscopic variables which characterize the system in a given
moment of time. New temporal values of such variables will
characterize the state in other moment
All possible states simultaneously Collapses to a single state when measured
Q(CaSC) / Johanna Sepúlveda 28
Principles
Superposition
Mental experiment: Cat of Schrödinger
Duck
Q(CaSC) / Johanna Sepúlveda 29
Principles
Entanglement: Classical physical notions
Realism Locality
Universe exist even if we do Each bit of the universe act
not observe it with the immediate
surrounding (no faster than c)
Q(CaSC) / Johanna Sepúlveda 30
Principles
Entanglement: Copenhagen Interpretation
In the universe:
✔ Reality needs an observer
✔ Exists as superposition of
possibilities
✔ Described by of the superposition
✔ Experience when the moment is
measured
Niels Bohr
Q(CaSC) / Johanna Sepúlveda 31
Principles
EPR Paradox (Einstein, Podolsky, Rosen)
"Spooky action at a distance“
Mental experiment to demonstrate the
paradox of the Copenhagen interpretation
of the universe
In the universe:
✔ All is real
✔ All is physical
✔ All is defined (local hidden variables)
Local realism
Q(CaSC) / Johanna Sepúlveda 32
Principles
Entanglement
The entangled particles are somehow connected.
In an entangled pair the measure of the state of a
particle will influence the state of the second
particle
Q(CaSC) / Johanna Sepúlveda 33
Principles
Entanglement- Bell inequalities (1964)
e--
(spin different
direction)
Photon pos
✔ Wavefunction is entangled
✔ Until measure we cannot know which is the spinning direction
✔ Measure of one will let us know the measure of the second one
✔ Measure has an effect (AXIS of measurement)
Q(CaSC) / Johanna Sepúlveda 34
Principles
Entanglement: Bell inequaliteis are violated
Alain Aspect: Entangled photons (polarization)
Correlation of states Correlation of states
Correlation with selected axis No correlation with selected axis
1st
Copenhagen Interpretation Local Realism
Universe is not local
Q(CaSC) / Johanna Sepúlveda 35
Principles
Entanglement
Q(CaSC) / Johanna Sepúlveda 36
Section 4
QUANTUM COMPUTER
Q(CaSC) / Johanna Sepúlveda 37
Q-Computer
New Ways of Computation
Q(CaSC) / Johanna Sepúlveda 38
Q-Computer
Qubit: Quantum bit
Q(CaSC) / Johanna Sepúlveda 39
Q-Computer
Qubit: Quantum bit
✔Quantum superposition
of 0 and 1
✔Whole sphere of states
✔Noisy qubit (somewhere
inside the sphere)
Q(CaSC) / Johanna Sepúlveda 40
Q-Computer
Qubit: Operation
Q(CaSC) / Johanna Sepúlveda 41
Q-Computer
Qubit: Operation
Q(CaSC) / Johanna Sepúlveda 42
Q-Computer
Qubit: Operation
Q(CaSC) / Johanna Sepúlveda 43
Section 5
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 44
Tasks
Summary
• System-on-Chips (SoC) to develop different
applications
• Moore law reached their limit (new ways of
computation)
• Quantum mechanics
– Princpiples: Wave-particle duality, Uncertainty Principle,
Schödinger equation
– Phenomena: Superposition and Entanglement
Q(CaSC) / Johanna Sepúlveda 45
Tasks
Summary
• Bit: Fundamental carrier of information
Possible bit states:
0 or 1
• Qubit: Fundamental carrier of quantum information
Possible qubit states: Any superposition described by
the wavefunction
Complex numbers
Q(CaSC) / Johanna Sepúlveda 46
Next Lecture
Quantum
Computers
Basic Principles
Next lecture: Quantum Computer Evolution
Structure
Types
Programming
Security threat
Q(CaSC) / Johanna Sepúlveda 47
Recommended Material (optional)
1) More videos:
What is the Heisenberg Uncertainty Principle?
https://www.youtube.com/watch?v=TQKELOE9eY4
Quantum Entanglement, Bell Inequality, EPR paradox
https://www.youtube.com/watch?v=v657Ylwh-_k
Qubit video used in the lecture (Spanish)
https://www.youtube.com/watch/?v=ilPfvMEOmCs
2) Familiarization with Qubit notation
Linear Algebra for Quantum Computation (28 pages)
https://link.springer.com/content/pdf/bbm%3A978-1-4614-6336-8%2F1.pdf
Q(CaSC) / Johanna Sepúlveda
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
3. Quantum Computer Operation
Q(CaSC) / Johanna Sepúlveda 1
Goal Today‘s Lecture
Summary of Previous Lecture
Qubits
Quantum Gates and Organization
Quantum Computer Technology
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 2
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 3
Summary
L2: Traditional Computer organization
Processor
Datapath Memory
Bus
Registers
Address
Control Data
Unit Control
Since 2014, it doubles every three years
ALU
Interfaces Co-processors
Q(CaSC) / Johanna Sepúlveda 4
Summary
L2: Quantum Computer Principles
1. Wave–Particle Duality
2. Uncertainty Principle of Heisenberg
Velocity=?
Measure has a strong effect on the quantum system
3. Schrödinger Equation
Function that describes the state of a quantum-mechanical system
Q(CaSC) / Johanna Sepúlveda 5
Summary
L2: Quantum Computer Principles
Superposition Entanglement
Collapses to a single state when measured
Q(CaSC) / Johanna Sepúlveda 6
Summary
L2: Qubits: Quantum bit
Qubit: Fundamental carrier of
quantum information
0
Possible qubit states: Any
superposition described by the
wavefunction
1
Complex numbers
Q(CaSC) / Johanna Sepúlveda 7
We are Ready to Start..
Quantum
Computers
Basic Principles
Evolution
Structure
Types
Programming
Security threat
Q(CaSC) / Johanna Sepúlveda 8
Section 2
QUBITS
Q(CaSC) / Johanna Sepúlveda 9
Qubit
Motivation
Qubit Notation
|𝟎
𝑩𝑲
Dirac Bra-Ket Notation
Row Column
Vector Vector
|𝟏 𝟐 𝟐
Where 𝜶 + 𝜷 =𝟏 Quantum amplitudes
Q(CaSC) / Johanna Sepúlveda 10
Qubit
Motivation
Qubit Notation
The state of a single qubit is a vector in a 2-dimensional
complex vector space
|𝒙 = α How much “0” is there in the state
How much “1” is there in the state
Where 𝜶 𝟐 + 𝜷 𝟐 =𝟏
Probability of 0 Probability of 1
Q(CaSC) / Johanna Sepúlveda 11
Qubit
Motivation
Qubit Notation
Qubit 0 (Ket 0)
1
|𝟎 = 0
Qubit 1 (Ket 1)
0
|𝟏 = 1
Q(CaSC) / Johanna Sepúlveda 12
Qubit
Motivation
Exercise: How does these qubit look like?
1
2
1
2
Q(CaSC) / Johanna Sepúlveda 13
Qubit
Motivation
Exercise: How does these qubit look like?
1
2
1
2
Q(CaSC) / Johanna Sepúlveda 14
Exercise: How does these qubit look like?
1
2
1
2
α Where 𝜶 𝟐
+ 𝜷 𝟐
=𝟏
β
1
α=β =
2
1
Probability of 0 = α2 =
2
Q(CaSC) / Johanna Sepúlveda 15
Compute
Motivation
Exercise: How does these qubit look like?
1
2
1
−
2
3𝑒𝑖𝜋/7
?
5
−4𝑖
5
Q(CaSC) / Johanna Sepúlveda 16
Need
Motivation
Measure a Qubit: State Collapse
Outcome 0
(High Probability)
Superposition
between 0 and 1
(Dominant 0)
Outcome 1
Q(CaSC) / Johanna Sepúlveda (Low Probability) 17
Need
Motivation
How do we Differentiate A from B?
Outcome 0
(50% Probability)
This is true for A and B
How to differentiate A from
B? Outcome 1
(50% Probability)
Q(CaSC) / Johanna Sepúlveda 18
Compute
Motivation
Qubit explosion of States
Explosion of states
Q(CaSC) / Johanna Sepúlveda 19
Compute
Motivation
Qubit explosion of States (Example 3 Qubits)
Register of N qubits will be in superposition of 2N possible strings
Quantum state 0 0 0
𝑪𝟎 |𝟎𝟎𝟎 +
𝑪𝟏 |𝟎𝟎𝟏 + 1 1 1
𝑪𝟐 |𝟎𝟏𝟎 + Superposition of states
(8 states at once)
𝑪𝟑 |𝟎𝟏𝟏 +
𝑪𝟒 |𝟏𝟎𝟎 +
2N
𝑪𝟓 |𝟏𝟎𝟏 +
𝑪𝟔 |𝟏𝟏𝟎 + Complex numbers
(N qubits)
𝑪𝟕 |𝟏𝟏𝟏
𝑪𝟎 𝟐 + 𝑪𝟏 𝟐 𝟐
+ 𝑪Q(CaSC) 𝑪𝟑 𝟐Sepúlveda 𝟐 𝟐 𝟐 𝟐
Where 𝟐 + / Johanna + 𝑪𝟒 + 𝑪𝟓 + 𝑪𝟔 + 𝑪𝟕 =𝟏
20
Compute
Motivation
Qubit explosion of States (Example 3 Qubits)
𝑪𝟎 |𝟎𝟎𝟎 + 0 0 0
𝑪𝟏 |𝟎𝟎𝟏 +
𝑪𝟐 |𝟎𝟏𝟎 +
𝑪𝟑 |𝟎𝟏𝟏 +
𝑪𝟒 |𝟏𝟎𝟎 +
𝑪𝟓 |𝟏𝟎𝟏 +
𝑪𝟔 |𝟏𝟏𝟎 +
1 1 1
𝑪𝟕 |𝟏𝟏𝟏
Superposition Entanglement
The state is described as a The state of entangled Qubits cannot
proportion of the different kets be described independently
Q(CaSC) / Johanna Sepúlveda 21
Compute
Motivation
How does it look like? (Example 3 Qubits)
No
1 0
1
1
High dimensional sphere
Q(CaSC) / Johanna Sepúlveda 22
Compute
Motivation
Main Message of this Section
Superposition and entanglement allow
Quantum Computers to store and manipulate vast
amounts of information
Q(CaSC) / Johanna Sepúlveda 23
Section 3
QUANTUM GATES
Q(CaSC) / Johanna Sepúlveda 24
Gates
Quantum Gates
Transformation of Qubits
|𝟎
Rotation Measure
Interaction among
qubits
|𝟏
Q(CaSC) / Johanna Sepúlveda 25
Need
Motivation
Quantum Gates: Single qubits
Quantum Gate
Pauli Gate Hadamard
Rotation around the different axis Creates superposition
50% |𝟎
X Y Z 50% |𝟏
Q(CaSC) / Johanna Sepúlveda 26
Need
Motivation
Quantum Gates
Quantum Gate
Pauli Gate
Q(CaSC) / Johanna Sepúlveda 27
Gates
Quantum Gates: Rotations
Q(CaSC) / Johanna Sepúlveda 28
Gates
Quantum Gates (Single Qubit)
A gate act acts on a qubit can be described as a 2x2 matrix
Any quantum gate can be written as a matrix of 2𝑛 × 2𝑛 where 𝑛 is the
number of qubits to which it is applied
The inverse of a gate 𝐺 is the conjugated-transposed gate:
Q(CaSC) / Johanna Sepúlveda 29
Gates
Quantum Gates (matrix)
0 1
Pauli X-gate (NOT gate or bit flip): 𝑋 =
1 0
1 0
• 0 = 1 =
0 1
0 1 1 0
• 𝑋0 = = = 1
1 0 0 1
0 −𝑖
Pauli Y-gate (bit flip and phase shift): 𝑌 =
𝑖 0
1 0
Pauli Z-gate (phase shift): Z =
0 −1
−𝑖𝜋
1 0
=𝑒 4 𝑒 0
𝑖𝜋 4
Phase gate: 𝑆 = 𝑖𝜋
0 𝑖 0 𝑒 4
−𝑖𝜋
1 0 𝑖𝜋 𝑒 8 0
π/8 gate: T = 𝑖𝜋 = 𝑒 8
𝑖𝜋
0 𝑒 4 0 𝑒 8
How will the gate s look like?
Q(CaSC) / Johanna Sepúlveda 30
Gates
Quantum Gates: Reversible logic
Quantum systems can reconstruct the past
Control NOT
Q(CaSC) / Johanna Sepúlveda 31
Gates
Quantum Gates: Swap Gate
Q(CaSC) / Johanna Sepúlveda 32
Gates
Quantum Gates (multiple Qubit)
SWAP
Controlled NOT
(CNOT)
Controlled Z
(CZ)
Toffoli
Q(CaSC) / Johanna Sepúlveda 33
Gates
Quantum Gates: Measure
Quantum Gate
Measure
The probability to collapse to 0 is much higher
than the probability to collapse to 1
Q(CaSC) / Johanna Sepúlveda 34
Gates
Quantum Circuits
To implement a global rotation, we decompose it into a
sequence of smaller dimensional rotations
INPUT OUTPUT
Circuits of 1 and 2 qubit gates
Q(CaSC) / Johanna Sepúlveda 35
Gates
Qubits and Quantum Circuits
Q(CaSC) / Johanna Sepúlveda 36
Section 3
QUANTUM ORGANIZATION
Q(CaSC) / Johanna Sepúlveda 37
Organization
Quantum Computer
Q(CaSC) / Johanna Sepúlveda 38
Organization
Sycamore Quantum Chip
Q(CaSC) / Johanna Sepúlveda 39
Organization
Quantum Computer Shield
Q(CaSC) / Johanna Sepúlveda 40
Organization
Coaxial Cable (Communication)
Q(CaSC) / Johanna Sepúlveda 41
Organization
Control Lines
Q(CaSC) / Johanna Sepúlveda 42
Organization
Control Lines
Q(CaSC) / Johanna Sepúlveda 43
Organization
Quantum Computer (Cooling System)
Q(CaSC) / Johanna Sepúlveda 44
Organization
Quantum Computer (Output)
Q(CaSC) / Johanna Sepúlveda 45
Organization
Quantum Computer (Programming Interface)
Q(CaSC) / Johanna Sepúlveda 46
Organization
Quantum Computer Status: NISQ
Noisy Intermediate Scale Quantum
Era
• Noisy imperfect but nontrivially big
quantum computers
• Beyond classical computational
power
• Quantum supremacy: quantum
computers can do something that
classical computers cannot unless
they run for a really long time
Q(CaSC) / Johanna Sepúlveda 47
Need
Motivation
Quantum Computer Supremacy
Q(CaSC) / Johanna Sepúlveda 48
Section 4
QUANTUM TECHNOLOGIES
Q(CaSC) / Johanna Sepúlveda 49
Technology
DiVicenzo Criteria
Q(CaSC) / Johanna Sepúlveda 50
Technology
Technologies
Q(CaSC) / Johanna Sepúlveda 51
Technology
Superconducting Circuits
Qubits are built by microwave
circuit oscillators
Microwave pulses are used to
control the qubits
Q(CaSC) / Johanna Sepúlveda 52
Technology
Trapped Ions
Ions stored in electromagnetic
traps
Ion qubits and interactions
manipulated by laser beams and
interactions between ions
Q(CaSC) / Johanna Sepúlveda 53
Technology
Electrically-Controlled Quantum Dots
Qubits are electron spins trapped
in a semiconductor nanostructure
Microwave pulses are used to
control the qubits
Q(CaSC) / Johanna Sepúlveda 54
Technology
Neutral Atoms
Atoms trapped by optical lattice
formed by lasers
Qubits formed by electronic
states of the atoms, and can be
controlled by lasers
Q(CaSC) / Johanna Sepúlveda 55
Technology
Quantum Computers (SoA)
Q(CaSC) / Johanna Sepúlveda 56
Section 5
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 57
Tasks
Summary
• Quantum Computers offer exponential
computational power due the superposition and
entanglement phenomena
• Quantum gates operate on quantum bits (qubits):
Rotation, Interaction (reversible) and measurement
• Currently we are at the noisy Intermediate Scale
Quantum Era
• Several technologies meet the DiVicenzo Criteria
Q(CaSC) / Johanna Sepúlveda 58
Recommended Material and tasks (optional)
1) Documentation material
A field guide to quantum computing
https://quantum-computing.ibm.com/docs/guide/
2) Solve challenges presented in the lecture and send it back to me by email
3) Build and send the simulation result of a quantum circuit executed in the IBM
computer (Try to predict the output using the matrix notation)
Example: (But you can try any other circuit)
Q(CaSC) / Johanna Sepúlveda
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
4. Quantum Algorithms
Q(CaSC) / Johanna Sepúlveda 1
Goal Today‘s Lecture
Summary of Previous Lecture
Quantum Technology
Quantum Logic Synthesis
Quantum Algorithms
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 2
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 3
Summary
Qubit Notation (1) Complex vector space
|𝟎
Quantum amplitudes
How much “0”
is there in the
α state
|𝒙 = β
How much “1”
|𝟏 is there in the
state
𝟐 𝟐
Where 𝜶 + 𝜷 =𝟏
Probability of 0 Probability of 1
Q(CaSC) / Johanna Sepúlveda 4
Summary
Qubit Notation (2) Univector EXTRA
Imaginary
𝛼= a+ib
α 𝛼= r𝑒 𝑖∅
b r
Φ Real
a
Q(CaSC) / Johanna Sepúlveda 5
Summary
Qubit Notation (2) Univector EXTRA
𝛼= r𝑒 𝑖∅
𝜑 = 𝑟0 𝑒 𝑖∅0 |0 + 𝑟1 𝑒 𝑖∅1 |1
Where 𝒓𝟎 𝟐 + 𝒓𝟏 𝟐 = 𝟏 𝜃
𝜑 =𝑒 𝑖∅0
(𝑟0 |0 + 𝑟1 𝑒 |1 𝑖∅ 𝑟1 = sin
2
Where ∅ = ∅𝟏 − ∅𝟎 𝜃
𝑟0 = cos
Q(CaSC) / Johanna Sepúlveda 2 6
Summary
Qubit explosion of States (Example 3 Qubits)
Register of N qubits will be in superposition of 2N possible strings
Quantum state
0 0 0
𝑪𝟎 |𝟎𝟎𝟎 +
𝑪𝟏 |𝟎𝟎𝟏 +
1 1 1
𝑪𝟐 |𝟎𝟏𝟎 + Superposition of states
(8 states at once)
𝑪𝟑 |𝟎𝟏𝟏 +
𝑪𝟒 |𝟏𝟎𝟎 +
2N
𝑪𝟓 |𝟏𝟎𝟏 +
𝑪𝟔 |𝟏𝟏𝟎 + Complex numbers
(N qubits)
𝑪𝟕 |𝟏𝟏𝟏
𝑪𝟎 𝟐 + 𝑪𝟏 𝟐 𝟐
+ 𝑪Q(CaSC) 𝑪𝟑 𝟐Sepúlveda 𝟐 𝟐 𝟐 𝟐
Where 𝟐 + / Johanna + 𝑪𝟒 + 𝑪𝟓 + 𝑪𝟔 + 𝑪𝟕 =
7 𝟏
Summary
Qubit explosion of States (Example 3 Qubits)
𝑪𝟎 |𝟎𝟎𝟎 + 0 0 0
𝑪𝟏 |𝟎𝟎𝟏 +
𝑪𝟐 |𝟎𝟏𝟎 +
𝑪𝟑 |𝟎𝟏𝟏 +
𝑪𝟒 |𝟏𝟎𝟎 +
𝑪𝟓 |𝟏𝟎𝟏 +
𝑪𝟔 |𝟏𝟏𝟎 +
1 1 1
𝑪𝟕 |𝟏𝟏𝟏
Superposition Entanglement
The state is described as a The state of entangled Qubits cannot
proportion of the different kets be described independently
Q(CaSC) / Johanna Sepúlveda 8
Summary
Quantum Gates
Transformation of Qubits
|𝟎
Rotation Measure
Interaction among
qubits
|𝟏
Q(CaSC) / Johanna Sepúlveda 9
Summary
Quantum Gates (Single Qubit)
A gate act acts on a qubit can be described as a 2x2 matrix
Any quantum gate can be written as a matrix of 2𝑛 × 2𝑛 where 𝑛 is the
number of qubits to which it is applied
Q(CaSC) / Johanna Sepúlveda 10
Summary
Quantum Gates: Reversible logic
Quantum systems can reconstruct the past
Control NOT
Q(CaSC) / Johanna Sepúlveda 11
Qubits and Quantum Circuits
Q(CaSC) / Johanna Sepúlveda 12
Summary
Quantum Computer
Quantum Computer
(Output)
Quantum Computer
Shield
Coaxial Cable
Cooling System (Communication)
Noisy Intermediate Scale Quantum Era (NISQ)
Quantum Computer (Programming
Interface) Quantum Chip
Q(CaSC) / Johanna Sepúlveda 13
Summary
IBM Quantum Computer
Q(CaSC) / Johanna Sepúlveda 14
We are Ready to Start..
Quantum
Computers
Basic Principles
Evolution
Structure
Types
Programming
Security threat
Q(CaSC) / Johanna Sepúlveda 15
Section 2
QUANTUM TECHNOLOGY
Q(CaSC) / Johanna Sepúlveda 16
Technology
DiVicenzo Criteria
Q(CaSC) / Johanna Sepúlveda 17
Technology
Quantum Computer Metrics
Coherence
Wavelength
Frequency
Velocity
Wave C is not coherent and will not cause maximum or minimum
interference
Q(CaSC) / Johanna Sepúlveda 18
Technology
Quantum Computer Metrics
Decoherence
Q(CaSC) / Johanna Sepúlveda 19
Technology
Quantum Computer Metrics
Control
Facility to process and interact with qubits
Q(CaSC) / Johanna Sepúlveda 20
Technology
Quantum Computer Metrics
Scalability Number of operations that can be performed
(circuit depth)
INPUT OUTPUT
Number of qubits that can be operated
(circuit wide)
Q(CaSC) / Johanna Sepúlveda 21
Technology
Evaluation Summary of Quantum Technologies
Q(CaSC) / Johanna Sepúlveda 22
Technology
Superconducting Circuits
Qubits are built by microwave
circuit oscillators
Microwave pulses are used to
control the qubits
Efficient (FAST)
Q(CaSC) / Johanna Sepúlveda 23
Technology
Trapped Ions
Ions stored in electromagnetic
traps
Paul trap
Ion qubits and interactions
manipulated by laser beams and
interactions between ions
Accuracy on qubit operation
Q(CaSC) / Johanna Sepúlveda 24
Technology
Electrically-Controlled Quantum Dots
Qubits are electron spins trapped
in a semiconductor nanostructure
Microwave pulses are used to
control the qubits
Q(CaSC) / Johanna Sepúlveda 25
Technology
Neutral Atoms
Atoms trapped by optical lattice
formed by lasers
Qubits formed by electronic
states of the atoms, and can be
controlled by lasers
Q(CaSC) / Johanna Sepúlveda 26
Technology
Quantum Computers (SoA)
Q(CaSC) / Johanna Sepúlveda 27
Section 3
Quantum Logic Synthesis
Q(CaSC) / Johanna Sepúlveda 28
Q-Synthesis
Interview: Quantum Logic Synthesis
Q(CaSC) / Johanna Sepúlveda 29
Q-Synthesis
Interview: Oracle
Inputs Output
Oracle
Implements a liner function
Programmable
Unitary
Q(CaSC) / Johanna Sepúlveda 30
Section 4
Quantum Algorithm
Q(CaSC) / Johanna Sepúlveda 31
Algorithm
Software
Q(CaSC) / Johanna Sepúlveda 32
Algorithm
Quantum Computers Utilization
Quantum computers are only
better than classical
(existing) computers at
specific computational
tasks
Quantum algorithms have
different degrees of:
• Evidence for/amount of
beyond-classical power
• Implementation Feasibility
Q(CaSC) / Johanna Sepúlveda 33
Algorithm
Main Quantum Applications
Q(CaSC) / Johanna Sepúlveda 34
Algorithm
Optimization in Industry
Find the shortest
circuit between cities
These problems are computationally difficult
(NP-hard)
Find the portfolio with the
minimum risk for a given return
Q(CaSC) / Johanna Sepúlveda 35
Algorithm
Optimization in Industry
Grover Algorithm: Polynomial speedup for these problems
Optimization, Monte Carlo Methods in Finance [Montanaro], Database Search,…
Q(CaSC) / Johanna Sepúlveda 36
Algorithm
Main Quantum Applications
Chemistry simulation
Optimization
Machine learning
Cryptography
Grover Algorithm
Shor Algorithm
Q(CaSC) / Johanna Sepúlveda 37
Algorithm
Implications on Security
QUANTUM
Grover’s algorithm
SKC/PKC in CLASSIC
Halves the security of AES
Better classic algorithm
AES-128 AES-256
~ 1034 Steps
Shor’s algorithm
In a classic computer (THz)
(1 trillion of ops / sec)
~ 107 Steps
In a quantum computer (MHz)
~ 17 Trillions of years (1 million of ops / sec)
~ 10 Seconds
Q(CaSC) / Johanna Sepúlveda 38
Algorithm
Quantum Threat
Type Algorithm Key strength Key strength Quantum attack
Classic Quantum
(bits) (bits)
Better classic algorithm
RSA-2048 112
Asymmetric
(Public Key)
~ 10RSA-3072
34 Steps
ECC-256
128
128
0 Shor’s Algorithm
ECC521 256
In a classic computer (THz)
(1 trillion of ops / sec)
AES-128 128 64
Grover’s
Symmetric
AES-256 256 128 Algorithm
Q(CaSC) / Johanna Sepúlveda 39
Algorithm
Quantum Algorithm Main Steps
Initialize Transform Measure
Q(CaSC) / Johanna Sepúlveda 40
Algorithm
Quantum Algorithm Main Steps
Initialize Transform Measure
Q(CaSC) / Johanna Sepúlveda 41
Section 5
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 42
Tasks
IBM Quantum Programming
https://quantum-computing.ibm.com/
Homework: Open an account and start exploring!
Q(CaSC) / Johanna Sepúlveda 43
Tasks
Grover’s Algorithm
https://quantum-computing.ibm.com/docs/guide/q-algos/grover-s-algorithm
Challenge: Execute the Grover’s Algorithm
(Send the result)
Q(CaSC) / Johanna Sepúlveda 44
Tasks
Grover’s Algorithm
Grover algorithm explanation (TU Delft):
https://www.youtube.com/watch?v=IT-O-KSWlaE
Steps of the Grover’s algorithm (Berkley):
https://www.youtube.com/watch?v=PAVKuYv1HC8
Next week I will explain the algorithm!
Q(CaSC) / Johanna Sepúlveda 45
Tasks
Summary
Several technologies meet the five DiVicenzo Criteria
Three metrics can be used to classify quantum
computers: Coherence, control and scalability
Logic synthesis is challenging and different
optimization guidelines can be used
Quantum algorithms are defined by three steps:
Initialization, transformation and measurement
Q(CaSC) / Johanna Sepúlveda 46
Next Lecture
End of the Quantum Block
Quantum
Shor’s and Grover’s Quantum Algorithms
Computers
Q&A Quantum Technology Basic Principles
Evolution
Structure
Types
Programming
Security threat
Q(CaSC) / Johanna Sepúlveda 47
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
5. Quantum Security Threat:
Grover and Shor Algorithms
Q(CaSC) / Johanna Sepúlveda 2
Goal Today‘s Lecture
Summary of Previous Lecture
Security Basics
Grover’s Algorithm
Shor’s Algorithm
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 3
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 4
Summary
DiVicenzo Criteria
Q(CaSC) / Johanna Sepúlveda 5
Summary
Quantum Computer Metrics
Decoherence
Q(CaSC) / Johanna Sepúlveda 6
Summary
Quantum Computer Metrics
Control
Facility to process and interact with qubits
Q(CaSC) / Johanna Sepúlveda 7
Summary
Quantum Computer Metrics
Scalability Number of operations that can be performed
(circuit depth)
INPUT OUTPUT
Number of qubits that can be operated
(circuit wide)
Q(CaSC) / Johanna Sepúlveda 8
Summary
Evaluation Summary of Quantum Technologies
Q(CaSC) / Johanna Sepúlveda 9
Summary
Quantum Computers (SoA)
Q(CaSC) / Johanna Sepúlveda 10
Summary
Main Quantum Applications
Chemistry simulation
Optimization
Machine learning
Cryptography
Grover Algorithm
Shor Algorithm
Q(CaSC) / Johanna Sepúlveda 11
Summary
Quantum Algorithm Main Steps
Initialize Transform Measure
Q(CaSC) / Johanna Sepúlveda 12
Summary
We are Ready to Start..
Quantum
Computers
Basic Principles
Evolution
Structure
Types
Programming
Security threat
Q(CaSC) / Johanna Sepúlveda 13
Section 2
SECURITY BASICS
Q(CaSC) / Johanna Sepúlveda 14
Security
Symmetric Key Cryptography
Public values
Alice‘s values Alice Bob Bob‘s values
Present Trivium Triple
AES Prince Lucifer DES
Q(CaSC) / Johanna Sepúlveda 15
Security
Public Key Cryptography
Public values
Alice‘s values Alice Bob Bob‘s values
RSA ECC
Q(CaSC) / Johanna Sepúlveda 16
Security
Public Key Cryptography
Q(CaSC) / Johanna Sepúlveda 17
Security
Public Key Cryptography
Application layer
(I) COAP Generate Data Data
Security layer
(II) DTLS Encryption EData
Transport layer
(III) UDP UDP Header UDP
Network layer
(IV) IPv6 IP Header IP
Adaptation layer
(V) 6LoWPAN 6LoWPAN
Physical layer F F
MAC/PHY
(VI) IEEE 802.15.4 H F
Q(CaSC) / Johanna Sepúlveda 18
Security
Public Key Cryptography
Client Server
Application layer
(I) COAP *ClientHello
*ClientHelloVerify
Security layer
(II) DTLS ClientHello
C1
Transport layer ServerHello, Certificate
UDP C2
(III)
Network layer
C3 ClientKeyExchange
ClientKeyExchange
(IV) IPv6 PKC
Adaptation layer Finished
(V) 6LoWPAN C4
ChangeCipherSpec
C5
Physical layer
Finished
(VI) IEEE 802.15.4
C6
SC
Q(CaSC) / Johanna Sepúlveda 19
Security
Traditional Security
Today traditional cryptography
is considered secure
Factoring large integers Computing discrete logarithms
Traditional Cryptography
(RSA, ECC)
Cryptoanalysis
Side-Channel
Attacker
IonQ’s 79 qubits Quantum Processor
In the future, traditional cryptography
will be broken
Q(CaSC) / Johanna Sepúlveda 20
Security
Quantum Computer RSA ECC
Key Length qubits Key Length qubits
1024 2048 163 1000
2048 4096 224 1300
140
3072 6144 256 1500
4096 8192 383 2300
120
15360 30720 512 3000
100
80
QUBITS
60
Superconductors
40 Ion Trap
Spin
20
Neutral atoms
0
1995 2000 2005 2010 2015 2020 2025
Q(CaSC) / Johanna Sepúlveda 21
Security
Implications on Security
QUANTUM
Grover’s algorithm
SKC/PKC in CLASSIC
Halves the security of AES
Better classic algorithm
AES-128 AES-256
~ 1034 Steps
Shor’s algorithm
In a classic computer (THz)
(1 trillion of ops / sec)
~ 107 Steps
In a quantum computer (MHz)
~ 17 Trillions of years (1 million of ops / sec)
~ 10 Seconds
Q(CaSC) / Johanna Sepúlveda 22
Security
Quantum Threat
Type Algorithm Key strength Key strength Quantum attack
Classic Quantum
(bits) (bits)
Better classic algorithm
RSA-2048 112
Asymmetric
(Public Key)
~ 10RSA-3072
34 Steps
ECC-256
128
128
0 Shor’s Algorithm
ECC521 256
In a classic computer (THz)
(1 trillion of ops / sec)
AES-128 128 64
Grover’s
Symmetric
AES-256 256 128 Algorithm
Q(CaSC) / Johanna Sepúlveda 23
Section 3
Grover Algorithm
Q(CaSC) / Johanna Sepúlveda 24
Grover
Optimization in Industry
Find the shortest
circuit between cities
These problems are computationally difficult
(NP-hard)
Find the portfolio with the
minimum risk for a given return
Q(CaSC) / Johanna Sepúlveda 25
Grover
Unstructured Search (N cards)
Worst case: Average case:
N trials N/2 trials
Winner w
Searching large databases is an important problem with
broad applications
Traditional Computer Quantum Computer
O(N) (Grover’s Algorithm)
O( 𝑵)
Q(CaSC) / Johanna Sepúlveda 26
Grover
Grover’s Algorithm: Amplitude amplification trick
Optimal search (exhaustive search) algorithm for a quantum
computers which has further applications as a subroutine for other
quantum algorithms
Four stages:
Initialization Transformation
Oracle
Amplification
Measurement
Q(CaSC) / Johanna Sepúlveda 27
Grover
Grover’s Algorithm: Parameter Definition
We know the value 𝒚 = 𝒇(𝒙) Our goal
N= 8 2𝑛=3 = 8
Initialization Oracle Amplification Measurement
Advantage: All searches are performed in a single iteration
N: number of cards
Repeat O( 𝑁) times
n: Qubits required to express
the N cards
Q(CaSC) / Johanna Sepúlveda 28
Grover
Grover’s Algorithm: Initialization 𝒚 = 𝒇(𝒙)
Bar graph of the amplitudes of the state
Initialization Oracle Amplification Measurement
Repeat O( 𝑁) times
Uniform
Superposition
𝟏
Amplitude
𝑵 mean
X
1 2 3 4 5 6 7 8
|000 |001 |010 |011 |100 |101 |110 |111
Q(CaSC) / Johanna Sepúlveda 29
Grover
Grover’s Algorithm: Oracle (1) 𝒚 = 𝒇(𝒙)
Bar graph of the amplitudes of the state
Initialization Oracle Amplification Measurement
Repeat O( 𝑁) times
Marks the solution(s)
Amplitude
𝟏
(flip sign of that <
𝑵 mean
state’s amplitude)
X
1 2 3 4 5 6 7 8
Identify Flip
Q(CaSC) / Johanna Sepúlveda 30
Grover
Grover’s Algorithm: Oracle (2)
Function that can be encoded through quantum
gates (distinguisher)
Oracle
Described as an unitary matrix
Quantum computer computes f(x) simultaneously
Result:
Amplitude representation: Inversion
Geometrical representation: reflection
Example:
Function that returns ‘0' for all possible input states,
except one input state (x= winner)
Q(CaSC) / Johanna Sepúlveda 31
Grover
Grover’s Algorithm: Amplification 𝒚 = 𝒇(𝒙)
Bar graph of the amplitudes of the state
Initialization Oracle Amplification Measurement
Repeat O( 𝑁) times
Reflection about the 𝟓
mean, thus increasing 𝟒 𝟐
the amplitude of the
Amplitude
marked state <
𝟏
mean
𝟏 𝑵
𝟒 𝟐 X
1 2 3 4 5 6 7 8
Diffusion operator / amplitude purification
(inversion about the mean)
Q(CaSC) / Johanna Sepúlveda 32
Grover
Grover’s Algorithm: Measurement 𝒚 = 𝒇(𝒙)
Note that the number of iterations in Grover's algorithm is critical.
High probability to many
If you make too measure |011
thethe
iterations state of success decreases again
probability
After only one iteration: 77% of getting the correct answer
For M solutions
𝑵
r≈ times
𝑴
After two iterations: 94% of getting the correct answer
Repeat
𝛑 𝑵
r =( ) times
𝟒
N=8, r ≈ 2
Q(CaSC) / Johanna Sepúlveda 33
Grover
Grover’s Algorithm: Geometrical 𝒚 = 𝒇(𝒙)
Define states
Define function
Q(CaSC) / Johanna Sepúlveda 34
Grover
Grover’s Algorithm: Quantum Circuit 𝒚 = 𝒇(𝒙)
Mathematical and Geometrical development:
https://www.quantiki.org/wiki/grovers-search-algorithm
https://www.cs.cmu.edu/~odonnell/quantum15/lecture04.pdf
https://www.nature.com/articles/s41467-017-01904-7
https://arxiv.org/ftp/arxiv/papers/0705/0705.4171.pdf
Simulator
http://davidbkemp.github.io/animated-qubits/grover.html
Q(CaSC) / Johanna Sepúlveda 35
Grover
Advanced Encryption Standard (AES-128)
AddRoundKey S0 S1 S2 S3
State Key
S4 S5 S6 S7
S8 S9 S10 S11
SubBytes
S12 S13 S14 S15
ShiftRows 9 rounds S3 XOR
k3
S‘3
MixColumns
S‘0 S‘1 S‘2 S‘3
AddRoundKey
S‘4 S‘5 S‘6 S‘7 State
S‘8 S‘9 S‘10 S‘11
SubBytes S-Box
S‘12S‘13S‘14 S‘15
ShiftRows
V3
AddRoundKey
https://arxiv.org/pdf/1512.04965.pdf
Q(CaSC) / Johanna Sepúlveda 36
Grover
Grover’s Algorithm: AES-128 case
Goal: To find the key x (winner) from 2128 possible solution that produce the
Ciphertext y
Challenge: To define the oracle for the case of AES.
We may know a set of plaintext/Ciphertext pairs
Message
(plaintext) Encrypted
AES Message
Secret (ciphertext)
N: number of
candidates
key
https://arxiv.org/pdf/1512.04965.pdf
https://eprint.iacr.org/2019/1146.pdf
Q(CaSC) / Johanna Sepúlveda 37
Grover
Grover’s Algorithm: AES-128 Oracle and results
The number of logical qubits required to implement a Grover attack on AES
is relatively low
Exponential development (2016: 3000 to 7000 Qubits; 2019: 864 to 1200)
Large circuit depth due the unrolling the entire Grover iteration
https://arxiv.org/pdf/1512.04965.pdf
https://eprint.iacr.org/2019/1146.pdf
Q(CaSC) / Johanna Sepúlveda 38
Section 4
Shor Algorithm
Q(CaSC) / Johanna Sepúlveda 39
Shor
Public Key Cryptography
Factoring large integers Computing discrete logarithms
Traditional Cryptography
(RSA, ECC)
Cryptoanalysis
Side-Channel
Attacker
N= p.q
Hard problem
Today traditional cryptography
is considered secure
Q(CaSC) / Johanna Sepúlveda 40
Shor
RSA Public-key Cryptography
N integer number has a unique decomposition into a product of primes, finding the
prime factors is believed to be a hard problem
Q(CaSC) / Johanna Sepúlveda 41
Shor
Complexity of Factoring (1)
N with d decimal digits
Brute force
All primes p, (0 < p < 𝑁)
Check if p divides N
Worst case: 𝑵 trials
(Exponential in d)
Quadratic sieve
Search two integers a2- b2 multiple of N
Check a ± b have common factors with N
Worst case: 𝑵 trials
(Exponential in 𝒅)
Q(CaSC) / Johanna Sepúlveda 42
Shor
Complexity of Factoring (2)
Q(CaSC) / Johanna Sepúlveda 43
Shor
Shor Algorithm: Mathematical
Large PrimesBackground
Euclid’s Algorithm: gcd (N,g) =p1 N is known, attacker
obtain p1 from the guess g
Guess N = p1 . p2 and calculates p2
Goal:
Get p2 or g (shares factor with N) so we can 𝑵
𝒑𝟐 =
𝒈𝒄𝒅(𝑵, 𝒈)
g = p1 . c
But if they do not share factors, it was demonstrated
𝒈𝒑 = m. N + 1
𝒑 𝒑
Factors of N (𝑔 +1). (𝑔 −1) = m. N
2 2
𝒑
𝒈2 ±1
Period Finding
p has a repeating property
𝒈𝒑+𝒙 = m2. N + r
Q(CaSC) / Johanna Sepúlveda 44
Shor
Period Finding
Modular math:
Dividend Quotient
𝑎
=𝑞 With remainder r
𝑁
Divisor
𝒓 = 𝑎 𝑚𝑜𝑑 𝑁
𝒈𝒑+𝒙 = m2. N + r
Factoring is easy if we find a period of the modular exponential
Given N and a, find smallest r
𝑎𝒓 − 1 is multiple of N
r: Period of a modulo N
Q(CaSC) / Johanna Sepúlveda 45
Shor
Shor Algorithm
Superposition of all the
possible powers
Quantum
Measure r (superposition of
Guess 𝒈 answers with equal r)
Extract p They are p apart from each
other
Obtain a better guess
𝒑 Initialization
𝑔 ±1
2
Oracle
Quantum Fourier Transformation
Measurement
Q(CaSC) / Johanna Sepúlveda 46
Shor
Shor’s Algorithm (1994)
Key step:
Quantum Fourier Transformation
Initialization Measurement
𝟏
𝒑
Find Period
Transformation
Q(CaSC) / Johanna Sepúlveda 47
Shor
Shor Algorithm (steps) N=21
𝑵
Choose a random g < 𝑵 . If gcd(𝒈, 𝑵) ≠ 1 then 𝑝1 = gcd(𝑥, 𝑁) and 𝑝2 = 𝑝1 ,
otherwise continue
𝒈 = 11
Choose the parameter 𝑴 ∈ 𝑁 2 , 2𝑁 2 ∶ 𝑴 = 2 𝒎
𝑴 = 512
𝒎=9
Apply the quantum algorithm to find the period 𝑟 of 𝑔𝑝 𝑚𝑜𝑑 𝑁
Initialize a register with 𝒏 + 𝒎 + 1 qubits
First 𝑛 + 𝑚 qubits to |0 and the last qubit to |1
𝒏 = 𝑙𝑜𝑔2 𝑵 = 5 and
Q(CaSC) / Johanna Sepúlveda 48
Shor
Shor Algorithm (steps) N=21
Apply Hadamard transformation to the first 𝒎 qubits to get an uniform
superposition of all states up to 𝑀 − 1
Apply the oracle function 𝑼𝒇 = 𝑥 𝑎 𝑚𝑜𝑑 𝑁
Apply the QFT on the first register
Measure the value on the first register
𝑃(𝑐)
𝑴
High probability of getting a multiple of
𝒑
𝑐 = 427
0
𝑀 2𝑀 3𝑀 𝑐
0
𝑝 𝑝 𝑝
Q(CaSC) / Johanna Sepúlveda 49
Shor
Period Finding
𝑴
Assuming we get a multiple of , use the continued fraction
𝒑
expansion algorithm (runs with polynomial time) to get the integer
𝑴
fraction that best approximates the result 𝑐 = 𝑘 , 𝑘 ∈ ℕ
𝒑
𝑐 𝑘 427 𝑘
min − = −
𝑴 𝒑 512 𝒑
p= 𝟔
𝒑
If 𝒑 is even and 𝒈 2 ≡ ±1 𝑚𝑜𝑑 𝑵, continue, otherwise, select a new
6
guess g11 2 ≡ 1331 ≡ 8 ≢ ±1 𝑚𝑜𝑑 21
Compute the factors of 𝑵 as
𝒑 𝒑
𝑝1 = gcd(𝒈 2 − 1, 𝑵) and p2 = gcd(𝒈 2 + 1, 𝑵)
𝑝1 = gcd 7, 21 = 7
𝑝2 = gcd 9, 21 = 3
Q(CaSC) / Johanna Sepúlveda 50
Section 5
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 51
Tasks
Quantum Programming
https://qiskit.org/textbook/ch-algorithms/grover.html
Q(CaSC) / Johanna Sepúlveda 52
Tasks
Summary: Quantum Threat
Two types of cryptography for securing
communication: Symmetric key and Public key
Security of symmetric key cryptography (SC) will be
halved by the Grover algorithm. AES-256 should be
used in the future
Public key cryptography (PKC) will be broken by the
Shor algorithm. Quantum secure communications
should be used
Q(CaSC) / Johanna Sepúlveda 53
Next Lecture
End of the Quantum Block
Quantum Communication Quantum
Communication
(QKD)
Main Principles
Structure
Types
Standard
Use cases
Q(CaSC) / Johanna Sepúlveda 54
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
6. Quantum Key Distribution (QKD)
Q(CaSC) / Johanna Sepúlveda 2
Goal Today‘s Lecture
Summary of Previous Lecture
Key Exchange
Quantum Communication
Important Information Lecture
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 3
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 4
Summary
Symmetric Key Cryptography
Public values
Alice‘s values Alice Bob Bob‘s values
Present Trivium Triple
AES Prince Lucifer DES
Q(CaSC) / Johanna Sepúlveda 5
Summary
Public Key Cryptography
Public values
Alice‘s values Alice Bob Bob‘s values
RSA ECC
Q(CaSC) / Johanna Sepúlveda 6
Summary
Traditional Security
Today traditional cryptography
is considered secure
Factoring large integers Computing discrete logarithms
Traditional Cryptography
(RSA, ECC)
Cryptoanalysis
Side-Channel
Attacker
IonQ’s 79 qubits Quantum Processor
In the future, traditional cryptography
will be broken
Q(CaSC) / Johanna Sepúlveda 7
Summary
Public Key Cryptography
SC
Q(CaSC) / Johanna Sepúlveda 8
Summary
Quantum Threat
Type Algorithm Key strength Key strength Quantum attack
Classic Quantum
(bits) (bits)
Better classic algorithm
RSA-2048 112
Asymmetric
(Public Key)
~ 10RSA-3072
34 Steps
ECC-256
128
128
0 Shor’s Algorithm
ECC521 256
In a classic computer (THz)
(1 trillion of ops / sec)
AES-128 128 64
Grover’s
Symmetric
AES-256 256 128 Algorithm
Q(CaSC) / Johanna Sepúlveda 9
Summary
Unstructured Search (N cards)
Worst case: Average case:
N trials N/2 trials
Winner w
Searching large databases is an important problem with
broad applications
Traditional Computer Quantum Computer
O(N) (Grover’s Algorithm)
O( 𝑵)
Q(CaSC) / Johanna Sepúlveda 10
Summary
Grover’s Algorithm: Amplitude amplification trick
Optimal search (exhaustive search) algorithm for a quantum
computers which has further applications as a subroutine for other
quantum algorithms
Four stages:
Initialization Transformation
Diffusion operator / amplitude purification
Oracle
(inversion about the mean)
Amplification
Measurement
Q(CaSC) / Johanna Sepúlveda 11
Summary
Grover’s Algorithm: AES-128 case
Goal: To find the key x (winner) from 2128 possible solution that produce the
Ciphertext y
Challenge: To define the oracle for the case of AES.
We may know a set of plaintext/Ciphertext pairs
Message
(plaintext) Encrypted
AES Message
Secret (ciphertext)
N: number of
candidates
key
https://arxiv.org/pdf/1512.04965.pdf
https://eprint.iacr.org/2019/1146.pdf
Q(CaSC) / Johanna Sepúlveda 12
Summary
Shor’s Algorithm
Period Finding
Goal: Find prime factors of N If N = 11, a= 5
Euclid’s Algorithm: gcd (N,g) r = 𝑎 𝑥 𝑚𝑜𝑑 𝑁
Equals to:
𝒑
𝒈2 ±1
5, 3, 4, 9, 1, 5, 3, 4, 9, 1, 5, 3, …
𝒈𝒑+𝒙 = m2. N + r
𝒓 = 𝒈𝒑+𝒙 𝑚𝑜𝑑 𝑁𝑚2 Period = 5
Factoring is easy if we find a period of the modular exponential
Given N and a, find smallest r
𝑎𝒓 − 1 is multiple of N
r: Period of a modulo N
Q(CaSC) / Johanna Sepúlveda 13
Summary
Shor’s Algorithm
Superposition of all the
possible p
Quantum
Measure r
Guess 𝒈
They are p apart from each
Extract p other
Obtain a better guess
𝒑 Initialization
𝑔 ±1 2
Oracle
𝑃(𝑐)
Quantum Fourier Transformation
0
Measurement
𝑀 2𝑀 3𝑀 𝑐
0
𝑝 𝑝 𝑝 Q(CaSC) / Johanna Sepúlveda 14
Summary
Shor’s Algorithm (1994)
Key step:
Quantum Fourier Transformation
Initialization Measurement
𝟏
𝒑
Find Period
Transformation
Q(CaSC) / Johanna Sepúlveda 15
Summary
We are Ready to Start..
Quantum
Communication
(QKD)
Main Principles
Structure
Types
Standard
Use cases
Q(CaSC) / Johanna Sepúlveda 16
Section 2
KEY EXCHANGE
Q(CaSC) / Johanna Sepúlveda 17
Key Exchange
Security Services (CIA)
Secrecy of the
Confidentiality information
Data is not
Cryptographic Primitivescorrupted
Integrity
(Use a key)
Authentication Integrity of the
source
Q(CaSC) / Johanna Sepúlveda 18
Key Exchange
Key Exchange
Cryptographic primitive that allows that
keys are exchanged between communication parties
01101010100001010101000010…..
Enforced Negotiated
A single party decides the secret Each party contributes to the
and transfer it establishment of a common secret
Q(CaSC) / Johanna Sepúlveda 19
Key Exchange
Key Exchange: Enforced
Alice Bob
Alice puts secret information
into a lockable briefcase
Alice locks the briefcase
Bob unlocks the briefcase
and retrieves the information
Q(CaSC) / Johanna Sepúlveda 20
Key Exchange
Key Exchange: Enforced (Encapsulation)
Alice Bob
Public key of Bob
Secret key of Bob
Alice selects the
secret
Alice locks the secret with
the public key of Bob
Bob unlocks the secret with
the Bob secret key
Q(CaSC) / Johanna Sepúlveda 21
Key Exchange
Key Exchange: Diffie-Hellman
Alice Bob
modulo p and base g
Selects Selects
g, p, a g, p, A b
Calculate Calculate
A = ga mod p B = gb mod p
B
Calculate Calculate
K = Ba mod p K = Ab mod p
Alice and Bob now share the
same secret K
Q(CaSC) / Johanna Sepúlveda 22
Whitfield Diffie and Martin Hellman.
Key Exchange
Once they share the secret…
Message
(plaintext) Encrypted
Encryption Message
Secret (ciphertext)
(key)
Encrypted
Message
(ciphertext) Message
Decryption (plaintext)
Secret
(key)
Q(CaSC) / Johanna Sepúlveda 23
Key Exchange
One Time Pad
Message
(plaintext) Encrypted
XOR
Encryption Message
Secret (ciphertext)
(key)
Pre-shared key (identical) should have at least the same size as the message
being sent (ABSOLUTELY SECURE)
Encrypted
Message
(ciphertext) Message
XOR
Decryption (plaintext)
Secret
(key)
Q(CaSC) / Johanna Sepúlveda Claude Shannon 24
Key Exchange
Once they share the secret…
Message
(plaintext) Encrypted
AES Message
(ciphertext)
Secret
(key)
Encrypted
Message
(ciphertext) Message
AES (plaintext)
Secret
(key)
Q(CaSC) / Johanna Sepúlveda 25
Key Exchange
Attacks: Man-in-the Middle (MiM)
Alice Bob
Selects Selects
g, p, a g, p, A b
Calculate Calculate
A = ga mod p B = gb mod p
B
Calculate Calculate
K = Ba mod p K = Ab mod p
Q(CaSC) / Johanna Sepúlveda 26
Key Exchange
Attacks: Man-in-the Middle (MiM)
Alice MiM Bob
Selects Selects Selects
g, p, a g, p, A g, p, m g, p, M b
Calculate Calculate Calculate
A = ga mod p M = gm mod p B = gb mod p
Calculate
B
M
Calculate KB = Bm mod p Calculate
KA = Ba mod p KA = Am mod p KB = Mb mod p
Share the same secret KA Share the same secret KB
AUTHENTICATION and TAMPER DETECTION
Q(CaSC) / Johanna Sepúlveda 27
Key Exchange
Attacks: Eavesdropping
Alice Bob
Selects Selects
g, p, a g, p, A b
Calculate Calculate
A = ga mod p B = gb mod p
B
Calculate Calculate
K = Ba mod p K = Ab mod p
Passive (just looking)
Q(CaSC) / Johanna Sepúlveda 28
Section 3
Quantum Communication
Q(CaSC) / Johanna Sepúlveda 29
QComm
Quantum Communication: Use Cases
Teleportation Quantum-key-Distribution
To share quantum states To share confidential random
Q(CaSC) / Johanna Sepúlveda 30
QComm
Teleportation
Goal: To transfer information without the physical
communication but through entanglement
Teleportation does not allow to scan a person and reassembled somewhere
else in the universe. Instead, it involves entangling two particles, separating
them, and keeping them stable enough to remain linked together
Q(CaSC) / Johanna Sepúlveda 31
QComm
QIN: Teleportation
Alice/Rick Bob/ Morty
Photon to be transmitted
Transmitted photon
Interact with other Modify the
particle quantum state
Any interaction with the entangled particle in one side (Rick) will be
immediately perceived by the other side (Morty)
Low Energy
Entangled Photons
High Energy Photon
Q(CaSC) / Johanna Sepúlveda 32
QComm
Practical Teleportation: Entanglement
Q(CaSC) / Johanna Sepúlveda 33
QComm
Teleportation in Practice: Communication
Laying fiber-optic cables is
expensive
Deployment takes a long time
Loss due to long cable
(entanglement will be
destroyed)
Quantum repeaters that
rebroadcast quantum
information could extend a
network's reach (no mature)
Separating entangled photons through fiber-optic cables is problematic
Q(CaSC) / Johanna Sepúlveda 34
QComm
Teleportation in Space: Micius 2017
Ngari transmitter
(Tibetan mountain)
Space-based link
Q(CaSC) / Johanna Sepúlveda 35
QComm
Teleportation in Space: Micius 2017
Over 32 evenings, 911 photon pairs showed teleportation
when the scientists measured their states
Q(CaSC) / Johanna Sepúlveda 36
QComm
Quantum-Key-Distribution (QKD)
Message
Encrypted Message
Message
Classical Network
Quantum Quantum
Quantum Channel
Device Device
Goal: To transfer a secret random key between a pair
It is impossible that an attacker read (measure) without
getting noticed
Q(CaSC) / Johanna Sepúlveda 37
QComm
QKD BB84 Protocol
1. Encode
2. Measure
3. Reconciliation Charles Bennett Giles Brassard
1. Encode Information: Polarized photons (polarization beam splitter)
Two polarization basis (four possible states)
V
L R
Orthogonal polarization
Q(CaSC) / Johanna Sepúlveda 38
QComm
QKD BB84 Protocol
2. Measure polarization: Aligned case
Input Output
(V)
(H)
Measure basis
Q(CaSC) / Johanna Sepúlveda 39
QComm
QKD BB84 Protocol
2. Measure polarization: Misaligned case (50% V, 50% H)
Input Output
(V)
OR
OR
(H)
OR
Measure basis OR
Q(CaSC) / Johanna Sepúlveda 40
QComm
QKD BB84 Protocol
2. Measure polarization: Aligned case
Input Output
Right
(R)
Left (L)
Measure basis
Q(CaSC) / Johanna Sepúlveda 41
QComm
QKD BB84 Protocol
2. Measure polarization: Misaligned case (50% R, 50% L)
Input Output
Right
(R) OR
OR
OR
Left (L)
Measure basis OR
Q(CaSC) / Johanna Sepúlveda 42
QComm
QKD BB84 Protocol
3. Reconciliation
Agree in a common secret key
Identify the mismatch between the selected basis
(transmitter/receiver)
Usually only 50% of the transmitted information is finally used
(after eliminating the errors)
Requires an authenticated channel (otherwise MiM attacks are
possible)
This key can be used for encrypting information
(OTP, AES)
Key Distillation
Error Correcting capabilities
Q(CaSC) / Johanna Sepúlveda 43
QComm
QKD BB84 Protocol
Alice/Rick Bob/ Morty
Public Channel
Quantum Channel
H V
Raw key
L R
Single photon
Encoding basis Encode
Selected basis
Correct measure Measure
Sifted key
Key Distillation
Reconciliation
Secret key
Q(CaSC) / Johanna Sepúlveda 44
QComm
QKD BB84 Protocol
Q(CaSC) / Johanna Sepúlveda 45
QComm
Possible Attacks: Photon Number Splitting
Avalanche photodiodes Trojan
Q(CaSC) / Johanna Sepúlveda 46
QComm
Decoy State QKD
Multiple source intensity levels to mess the
communication statistics
(against possible attacker: Decoy state)
Q(CaSC) / Johanna Sepúlveda 47
QComm
QKD Practical Realization
Q(CaSC) / Johanna Sepúlveda 48
QComm
QKD Today
Q(CaSC) / Johanna Sepúlveda 49
QComm
QKD Applications
Q(CaSC) / Johanna Sepúlveda 50
QComm
QKD Advantages and Disadvantages
Quantum Key Distribution promises unconditional security
. Security is based on the laws of quantum physics
Detect compromised channels (Eavesdropping): Virtually
unhackable
Still, we have many challenges to address
Poorly understood
Only to exchange key (other cryptoprimitives are not supported)
Rely on authenticated channels
Secure Implementation is mandatory
Extend the performance (limits on rate and range)
Cost
Killer application
Q(CaSC) / Johanna Sepúlveda 51
QComm
Quantum Key Distribution
Q(CaSC) / Johanna Sepúlveda 52
Section 5
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 53
Tasks
QKD Standard
https://www.etsi.org/technologies/quantum-key-distribution
Describing complete systems involving point-to-point devices from physical
implementation to output of final secret key
Revision on security proofs, focusing on security definition, device models,
implementation security and quantum key distribution protocols
Implementation security: protection against Trojan horse attacks in one-way QKD
systems
Characterisation of optical output of QKD transmitter modules, on control interface
QKD Application Interface (API) in response to new network developments on
review of network architectures
Revision to Group Report (GR) on QKD vocabulary and definitions
Q(CaSC) / Johanna Sepúlveda 54
Tasks
QKD attacks: Answer Moodle Questions
https://www.etsi.org/images/files/ETSIWhitePapers/etsi_wp27_qkd_imp_sec_FINAL.pdf
Q(CaSC) / Johanna Sepúlveda 55
Tasks
Summary: Quantum Communication
Key distribution is critical for the security
Two types of quantum communication:
Teleportation and Quantum Key Distribution
Quantum communication is very promising and it still
requires a huge research effort
While Quantum communication is secure,
implementation attacks can be performed
Q(CaSC) / Johanna Sepúlveda 56
Tasks
Next Lecture
Third Block
Post Quantum Security Post-Quantum
Cryptography
(PQC)
NIST Competition Main Principles
Families and Types
Principles PQC Main Components
Standard
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 57
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
SoSe20
950488885
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Goal Today‘s Lecture
Summary of Previous Lecture
Cryptography
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 2
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 3
Summary
Security Services (CIA)
Secrecy of the
Confidentiality information
Data is not
Cryptographic Primitivescorrupted
Integrity
(Use a key)
Authentication Integrity of the
source
Q(CaSC) / Johanna Sepúlveda 4
Summary
Key Exchange
Cryptographic primitive that allows that
keys are exchanged between communication parties
01101010100001010101000010…..
Enforced Negotiated
A single party decides the secret Each party contributes to the
and transfer it establishment of a common secret
Q(CaSC) / Johanna Sepúlveda 5
Summary
Once they share the secret…
Message
(plaintext) EncryptedM
Encryption essage
Secret (ciphertext)
(key)
One-Time-Pad (OTP) or AES
EncryptedM
essage
(ciphertext) Message
Decryption (plaintext)
Secret
(key)
Q(CaSC) / Johanna Sepúlveda 6
Summary
Public Key Cryptography
SC
Q(CaSC) / Johanna Sepúlveda 7
Summary
Traditional Security
Today traditional cryptography
is considered secure
Factoring large integers Computing discrete logarithms
Traditional Cryptography
(RSA, ECC)
Cryptoanalysis
Side-Channel
Attacker
IonQ’s 79 qubits Quantum Processor
In the future, traditional cryptography
will be broken
Q(CaSC) / Johanna Sepúlveda 8
Summary
Quantum Threat
Type Algorithm Key strength Key strength Quantum attack
Classic Quantum
(bits) (bits)
Better classic algorithm
RSA-2048 112
34
RSA-3072 128
Asymmetric
(Public Key)
~ 10ECC-256
Steps 128
0 Shor’s Algorithm
ECC521 256
In a classic computer (THz)
(1 trillion of ops / sec)
AES-128 128 64
Grover’s
Symmetric
AES-256 256 128 Algorithm
Q(CaSC) / Johanna Sepúlveda 9
Summary
Quantum Communication: Use Cases
Teleportation Quantum-key-Distribution
To share quantum states To share confidential random
Goal: To transfer Goal: To transfer a secret
information without the random key between a pair
physical communication but It is impossible that an
through entanglement attacker read (measure)
without getting noticed
Q(CaSC) / Johanna Sepúlveda 10
Summary
QIN: Teleportation
Alice/Rick Bob/ Morty
Photon to be transmitted
Transmitted photon
Interact with other Modify the
particle quantum state
Any interaction with the entangled particle in one side (Rick) will be
immediately perceived by the other side (Morty)
Low Energy
Entangled Photons
High Energy Photon
Q(CaSC) / Johanna Sepúlveda 11
Summary
Teleportation in Practice: Communication
Separating entangled photons
through fiber-optic cables is Space-based link
problematic
Q(CaSC) / Johanna Sepúlveda 12
Summary
Teleportation in Space: Micius 2017
Over 32 evenings, 911 photon pairs showed teleportation
when the scientists measured their states
Q(CaSC) / Johanna Sepúlveda 13
Summary
Quantum-Key-Distribution (QKD)
Message
Encrypted Message
Message
Classical Network
Quantum Quantum
Quantum Channel
Device Device
Goal: To transfer a secret random key between a pair
It is impossible that an attacker read (measure) without
getting noticed
Q(CaSC) / Johanna Sepúlveda 14
Summary
QKD BB84 Protocol
1. Encode
2. Measure
3. Reconciliation Charles Bennett Giles Brassard
1. Encode Information: Polarized photons (polarization beam splitter)
Two polarization basis (four possible states)
V
L R
Orthogonal polarization
Q(CaSC) / Johanna Sepúlveda 15
Summary
QKD BB84 Protocol
2. Measure polarization: Aligned case
Input Output
(V)
(H)
Measure basis
Q(CaSC) / Johanna Sepúlveda 16
Summary
QKD BB84 Protocol
2. Measure polarization: Misaligned case (50% V, 50% H)
Input Output
(V)
OR
OR
(H)
OR
Measure basis OR
Q(CaSC) / Johanna Sepúlveda 17
Summary
QKD BB84 Protocol
3. Reconciliation
Agree in a common secret key
Identify the mismatch between the selected basis
(transmitter/receiver)
Usually only 50% of the transmitted information is finally used
(after eliminating the errors)
Requires an authenticated channel (otherwise MiM attacks are
possible)
This key can be used for encrypting information
(OTP, AES)
Key Distillation: Amplification of Privacy
Error Correcting capabilities
Q(CaSC) / Johanna Sepúlveda 18
Summary
QKD BB84 Protocol
Alice/Rick Bob/ Morty
Public Channel
Quantum Channel
H V
Raw key
L R
Single photon
Encoding basis Encode
Selected basis
Correct measure Measure
Sifted key
Key Distillation
Reconciliation
Secret key
Q(CaSC) / Johanna Sepúlveda 19
Summary
Possible Attacks: Photon Number Splitting
Avalanche photodiodes Trojan
Q(CaSC) / Johanna Sepúlveda 20
Summary
QKD Practical Realization
Q(CaSC) / Johanna Sepúlveda 21
Summary
QKD Today
Q(CaSC) / Johanna Sepúlveda 22
Summary
QKD Applications
Q(CaSC) / Johanna Sepúlveda 23
Summary
Quantum Key Distribution
Q(CaSC) / Johanna Sepúlveda 24
Summary
We are Ready to Start..
Post-Quantum
Cryptography
Main Principles
Families and Types
Standard
Main Components
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 25
Section 2
CRYPTOGRAPHY
Q(CaSC) / Johanna Sepúlveda 26
Crypto
Human need: Communication
Defense
Technology
Health
Logistics
Share information Selectively
Q(CaSC) / Johanna Sepúlveda 27
Crypto
Cryptography History
Q(CaSC) / Johanna Sepúlveda 28
Crypto
Once Upon a Time
It’s a warm summer evening in ancient Greece…
Cryptography:
‘Krypto’ meaning hidden
and ‘graphene’ meaning writing.
Q(CaSC) / Johanna Sepúlveda 29
Crypto
Classical Cryptography
Hieroglyph (4500 ago): Only known by Egyptian scribes
Scytale (2500 ago):
Greeks and Spartans for
military incursions
Polybius Square (2500 ago):
Caesar cipher (1900 ago): Romans for military incursions.
Greeks. They use coordinates of the
Replace letters with a letter that is 3 letters beyond.
letters Q(CaSC) / Johanna Sepúlveda 30
Crypto
Middle Age Cryptography
Cryptanalysis: Science that develops a set of techniques to
decipher the cryptography
Q(CaSC) / Johanna Sepúlveda 31
Crypto
Reinassance Cryptography
Vigenère cipher
• Further development of
substitution-based
ciphers (polyalphabetic)
• Based on a series of
interwoven Caesar
ciphers
• Different alternatives
(rows)
Tabula Recta (1600)
Q(CaSC) / Johanna Sepúlveda 32
Crypto
Premodern Cryptography
One Time Pad –OTP (1919): Gilbert Vernam
Enigma Sigaba
Mechanical and Electromechanical cipher machines (1950)
Q(CaSC) / Johanna Sepúlveda 33
Crypto
Modern Cryptography
Note: Algorithms and parameters are PUBLIC and private keys are PRIVATE
Q(CaSC) / Johanna Sepúlveda 34
Crypto
Modern Cryptography
SYMMETRIC-KEY
(SKC)
Q(CaSC) / Johanna Sepúlveda 35
Crypto
Cryptography Definition
Parameters Performance
Length of the key
Modulo Cost
Noise
Seed….
Cryptographic
Pre-defined values Primitive Security
that define the
cryptographic
primitives Other
In order to allow interoperability and to widely deploy secure communication,
standards are needed
Q(CaSC) / Johanna Sepúlveda 36
Crypto
NIST: National Institute of Standards and Technology
Goal: To stablish a measurement system (standards)
Security standards (key distribution, signature, encryption, etc..)
Industry
Researchers
Government Requirements
Get Expertise
Unilateral
Define
Public:
Competition
Q(CaSC) / Johanna Sepúlveda 37
Crypto
NBS: Data Encryption Standard (DES)
Symmetric key encryption (Block ciphers)
• 1973: First request for proposals (No suitable candidate)
• 1974: Second request for proposals (No suitable candidate)
• 1977: Announced U.S. FIPS PUB 46 (15.01.1977)
• 1999: Broken in 22 hours and 15 minutes
• 1999: Triple DES reaffirmed in the U.S. FIPS PUB 46-3
Q(CaSC) / Johanna Sepúlveda 38
Crypto
The Drama
NSA
Lucifer
(128-bits)
National Bureau of
Public and open discussion is mandatory
Standard
The Bureau
DES Backdoor intentionally
introduced?
(56-bits)
Q(CaSC) / Johanna Sepúlveda 39
Crypto
NIST Public Process (Competition)
Q(CaSC) / Johanna Sepúlveda 40
Crypto
NIST: Advanced Encryption Standard (AES)
Symmetric key encryption
• 1997: Announce the AES competition
• 1998: 15 Candidates were submitted
• Public discussion (Advantages and Disadvantages) Joan Daemen and Vincent Rijmen
• Security assessment
• Performance assessment (PCs of various architectures, smart cards, hardware implementations)
• Feasibility in limited environments (smart cards , low gate count ASICS, FPGAs)
• 1999: Narrow to 5 candidates: MARS, RC6, Rijndael, Serpent and Twofish
(Finalists)
• 2000: NIST announced Rijndael as the winner
• 2001: Draft of the AES (U.S. FIPS PUB 197)
• 2002: AES is effective as a Federal government standard. Included in the
ISO/IEC 18033-3, NSA approved (used in top secret messages)
Q(CaSC) / Johanna Sepúlveda 41
Crypto
NIST: Lightweight Standard
SHA-3
Secure Hash Algorithm 3
(2015)
Lightweight
57 Candidates
(2019)
Q(CaSC) / Johanna Sepúlveda 42
Crypto
NIST: Lightweight Standard
Symmetric key encryption
• 2015: Start public discussions on Lighweight Cryptography
• 2018: Announce the lightweight competition
• 2019: 57 candidates were submitted
• Caesar competition (public authenticated encryption)
• Performance assessment (PCs of various architectures, smart cards, hardware
implementations)
• Feasibility in limited environments (smart cards , low gate count ASICS, FPGAs)
• 2020: Secure implementation
• Time-protected
• Side-channel (power, EM, ..)
Q(CaSC) / Johanna Sepúlveda 43
Crypto
Side-Channel: Which activity did they practice
We can get information that is leaked from the normal operation of the
system to retrieve some secret/privilege information
(time to execute a function, power, access, etc..)
Q(CaSC) / Johanna Sepúlveda 44
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
8. Lattice-Based Post-Quantum Cryptography
Principles
Q(CaSC) / Johanna Sepúlveda 2
Goal Today‘s Lecture
Summary of Previous Lecture
Lattices
Lattice-Based Cryptography
LWE / NTRU
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 3
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 4
Summary
Once Upon a Time
It’s a warm summer evening in ancient Greece…
Cryptography:
‘Krypto’ meaning hidden
and ‘graphene’ meaning writing.
Q(CaSC) / Johanna Sepúlveda 5
Summary
Cryptography History
Post-Quantum
Enigma
Cryptographic Algorithm and Cryptographic Implementation
Cryptanalysis
must be secure
Hieroglyph
Scytale
Q(CaSC) / Johanna Sepúlveda 6
Summary
Side-Channel: Which activity did they practice
We can get information that is leaked from the normal operation of the
system to retrieve some secret/privilege information
(time to execute a function, power, access, etc..)
Q(CaSC) / Johanna Sepúlveda 7
Summary
Traditional Security
Today traditional cryptography
is considered secure
Factoring large integers Computing discrete logarithms
Traditional Cryptography
(RSA, ECC)
Cryptoanalysis
Side-Channel
Attacker
IonQ’s 79 qubits Quantum Processor
In the future, traditional cryptography
will be broken
Q(CaSC) / Johanna Sepúlveda 8
Summary
Quantum Threat
Type Algorithm Key strength Key strength Quantum attack
Classic Quantum
(bits) (bits)
Better classic algorithm
RSA-2048 112
Asymmetric
(Public Key)
~ 10RSA-3072
34 Steps
ECC-256
128
128
0 Shor’s Algorithm
ECC521 256
In a classic computer (THz)
(1 trillion of ops / sec)
AES-128 128 64
Grover’s
Symmetric
AES-256 256 128 Algorithm
Q(CaSC) / Johanna Sepúlveda 9
Summary
Quantum Computer Threat
RSA ECC
Key Length qubits Key Length qubits
140 1024 2048 163 1000
2048 4096 224 1300
120 3072 6144 256 1500
4096 8192 383 2300
100
15360 30720 512 3000
80
QUBITS
60
Superconductors
40 Ion Trap
Spin
20
Neutral atoms
0
1995 2000 2005 2010 2015 2020 2025
Q(CaSC) / Johanna Sepúlveda 10
Summary
QKD: Quantum Key Distribution
Message
Encrypted Message
Message
Classical Network
Quantum Quantum
Quantum Channel
Device Device
Used to distribute keys
ONLY
Q(CaSC) / Johanna Sepúlveda 11
Summary
QKD: Quantum Channels
Q(CaSC) / Johanna Sepúlveda 12
Summary
PQC: Post-Quantum Cryptography
Distribute Key
Post-Quantum Cryptography Encrypt
Decrypt
Classical Network Sign
Verify
Code Hash Multivariate Isogeny Lattice
Algorithms used to secure messages on a classical computer and that are resistant to
traditional computers attacks AND quantum computers attacks
Q(CaSC) / Johanna Sepúlveda 13
Summary
PQC: Comparison
Code-based Hash-based Multivariate Isogeny Lattice-based
(e.g. NTRU, LWE, RLWE)
(e.g. McEliece, Rollo) (e.g. Merkle Hash- (e.g. Rainbow, LUOV) (e.g. SIKE)
trees)
Pros: Pros: Pros: Pros: Pros:
Well studied Well studied Multipurpose Elliptic-based Efficient
problem problem Very efficient Smallest key Public
Multipurpose Very efficient signature sizes key, digital
Fast schemes signatures
, FHE, IBE
Cons: Cons: Cons: Cons: Cons:
Very large No encryption Most public key Low efficiency Key sizes
key sizes schemes schemes are Difficult to when
Track of signed broken construct compared to
messages
classical
crypto
Q(CaSC) / Johanna Sepúlveda 14
Summary
PQC Momentum
0 5 10 15 20 25 30
Life cycle of the
product
(Years)
Expose industry and products
to attacks
Post-Quantum Security
Integration Process IonQ’s 79 qubits Quantum Processor
Q(CaSC) / Johanna Sepúlveda 15
Summary
NIST Post-Quantum Candidates
Code Hash Multivariate Isogeny Lattice
First NIST
18 3 9 2 32
Round
Second NIST
7 2 4 1 12
Round
Q(CaSC) / Johanna Sepúlveda 16
Summary
NIST: National Institute of Standards and Technology
• 2012: NIST starts meetings regarding Post-Quantum cryptography
• 2016: NIST announces the Post-Quantum standardization process
April: Report NISTIR 8105 (Report of PQC)
December: Official call for proposal
• 2017: Deadline of submissions of Post-Quantum cryptography (30.11)
• 2018: First conference of standarization of Post-Quantum cryptography
• 2019: Starts the second round of the Standardization process
– January: Report NISTIR 8240 (Report of PQC)
August: Second conference of standarization
• 2020: Freeze of second round and start internal discussion for selection of the
third round candidates
• 2022: Expected arrival of standardization (set of recommended algorithms)
Q(CaSC) / Johanna Sepúlveda 17
Summary
QKD IS NOT PQC
Advantages
- Wider portfolio of mathematical problems in
Advantages
comparison with traditional PKI (e.g., RSA, ECC)
- Security based on physics
- Use already deployed technology
- Virtually unhackable - Able to exchange keys, encrypt, decrypt, sign,
- Detect eavesdropping verify
- Target of intense research
- Standardization process (NIST, IETF, IEEE,
ETSI)
Disadvantages - Scalability
- Prone to implementation attacks - Good performance
- Resilience: Change in photon polarization - Target of intense research
- Lacks many security features (only key
exchange) Disadvantages
- Need of dedicated channel (high cost) - Prone to implementation attacks
- Restricted key throughput (scalability) - Need to build confidence
Q(CaSC) / Johanna Sepúlveda 18
- Larger keys when compared to traditional PKI
Summary
We are Ready to Start..
Post-Quantum
Cryptography
Main Principles
Families and Types
Standard
Main Components
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 19
Section 2
LATTICE
Q(CaSC) / Johanna Sepúlveda 20
Lattice
Lattices: Which concepts arrive to our mind?
Space Dimension
(Euclidean Space)
Basic Element
(Basis)
Order
(Pattern)
Q(CaSC) / Johanna Sepúlveda 21
Lattice
Lattices
𝑏2
𝑏1
Q(CaSC) / Johanna Sepúlveda 22
Lattice
Lattices: Scary and Friendly Basis
An n-dimensional
Dimension lattice ℒ is
• Generated by a basis
ℬ = 𝑏1 , … , 𝑏𝑛 Rank
ℬ𝑚𝑥𝑛
• ℒ = 𝑛𝑖=1(ℤ ⋅ 𝑏𝑖 )
𝑏2 Scary Basis: Long
ℒ(ℬ) = 𝑩𝒙 | 𝒙 ∈ ℤ𝑛 𝑏1 vectors
Friendly Basis: Short
vectors
Basis is not unique:
There are multiple basis that can generate the same lattice
Q(CaSC) / Johanna Sepúlveda 23
Lattice
Lattices
Are these vectors a basis for our lattice?
𝑏1
𝑜
𝑏2
Q(CaSC) / Johanna Sepúlveda 24
Lattice
Lattice Basis: Example
Dimension = 2
Rank = 1
Q(CaSC) / Johanna Sepúlveda 25
Lattice
Lattice Summary
Can be described as a periodic space of points, defined in an
euclidean space n-dimensional
Represented compactly by the basis
• Independent n vectors
• Lattice is represented by the set of all basis lineal combination
• Basis is not unique
Many HARD PROBLEMS can be described with lattices
Problems that are not possible to solve or that require
excessive traditional and quantum computational capabilities
2O(n) Quantum steps required to solve lattice problems
Q(CaSC) / Johanna Sepúlveda 26
Lattice
Shortest Vector Problem (SVP)
Given a lattice L, find the
shortest non-zero vector to the origin
We know
Some variations:
𝑆
α-shortest vector
𝑜
Q(CaSC) / Johanna Sepúlveda 27
Lattice
Closest Vector Problem (CVP)
Given a lattice L and a target t, find the
closest lattice point
We know
𝑜
𝐶
Q(CaSC) / Johanna Sepúlveda 28
Lattice
Learning With Errors (LWE) [Regev 2005]
Given a lattice L and a vector r (visible vector), find
the verctor s hidden by the error e
𝑟
e
𝑜
𝒔
Q(CaSC) / Johanna Sepúlveda 29
Lattice
Interpretation of LWE
Algebraic: Solving noisy random linear
equations
Geometric: Bounded Distance Decoding in lattices
Learning Theory: Learning linear functions over ℤ𝒏𝒒 under
random classification noise
Elements reduced
mod q
Coding Theory: Decoding random linear q-ary codes
Q(CaSC) / Johanna Sepúlveda 30
Lattice
LWE - Algebraic
𝑎11 𝑎12 … 𝑎1𝑛 𝑆11
𝐴𝑚𝑥𝑛 = 𝑎21⋮ 𝑎22… 𝑎2𝑛 𝑆𝑛𝑥1 = 𝑆21⋮
⋮ ⋮
𝑎𝑚1 𝑎𝑚2 … 𝑎𝑚𝑛 𝑆𝑛1
A is a basis of a lattice S is the secret vector
𝑒11 𝑟11
𝑒𝑚𝑥1 = 𝑒21⋮ 𝑟𝑚𝑥1 = 𝑟21⋮
𝑒𝑚1 𝑟𝑚1
Gaussian
e is the error vector r is the visible vector
(small magnitude)
𝑨. 𝒔 + 𝒆 = 𝒓 Decisional LWE
Given
Search LWE Find (and/or)
Q(CaSC) / Johanna Sepúlveda 31
Lattice
LWE – Decisional and Searching Problem
𝑨. 𝒔 + 𝒆 = 𝒓
Public values
n: Size of the secret
m: Number of samples
q: Modulus
Χ: Error distribution
Decisional LWE
Search LWE
(Distinguish)
Q(CaSC) / Johanna Sepúlveda 32
Lattice
More Hard Lattice-based Problems
Short Integer Solution (SIS): Find the short non-zero 𝒛 ∈ ℤ𝑚
Shortest Independent Vector Problem (SIVP): Given a basis for a
lattice find n linear independent vectors constrained by
Covering Radius Problem (CRP): Given a basis for a lattice find
the smallest sphere that when placed at every lattice point it
includes 2 lattice points
Bounded Distance Decoding (BDD): Given a basis, a point x and
a bound β, find the smallest vector v such that 𝑥 − 𝑣 ≤ 𝛽
Q(CaSC) / Johanna Sepúlveda 33
Lattice
Lattice Problems: Hardness
Quantum
[R‘05]
GapSVP, ≤
SIVP
≤ ≤
Search-LWE Decision-LWE Crypto
Decision version of SVP
GapSVP
≤
[BFKL94] [R‘05],
[R‘05], [PW‘08]
[P‘09] [GPV08]
NTRU LWE
Classical
[P‘09]
Q(CaSC) / Johanna Sepúlveda 34
Lattice
Lattice Hard Problems: Summary
V´
𝑠 c c
Shortest Vector Closest Vector Learning
Problem Problem with errors
More configuration parameters (Rich exploration trade-off)
Easy to proof the security
Q(CaSC) / Johanna Sepúlveda 35
Lattice
Taxonomy of Lattices
Lattices
NTRU
LWE
(SVP)
Ring
RLWE ILWE MLWE LWR LWP RLWR
Replaces n dimensional
vectors by polynomials of
degree smaller than n
Q(CaSC) / Johanna Sepúlveda 36
Section 3
LATTICE-BASED CRYPTOGRAPHY
Q(CaSC) / Johanna Sepúlveda 37
Crypto
Public Key Cryptography (PKC)
ga mod p N=p.q
Computing discrete logarithms Public values Factoring large integers
Alice‘s values Alice Bob Bob‘s values
PKC Security
Transformation is based on hard problems which can not be
solved or they require excessive computation/memory resources
Easy to calculate….hard to crack
Q(CaSC) / Johanna Sepúlveda 38
Crypto
Key Exchange: Diffie-Hellman
Alice Bob
modulo p and base g
Selects Selects
g, p, a g, p, A b
Calculate Calculate
A = ga mod p B = gb mod p
B
Calculate Calculate
K = Ba mod p K = Ab mod p
Alice and Bob now share the
same secret K
Q(CaSC) / Johanna Sepúlveda 39
Crypto
Public Key Cryptography (PKC)
𝑎11 𝑎12 … 𝑎1𝑛
𝐴𝑚𝑥𝑛 = 𝑎21⋮ 𝑎22…
⋮
𝑎2𝑛
⋮
𝑎𝑖𝑗 <q and 𝑎𝑖𝑗 ∈ ℤ𝑛
𝑎𝑚1 𝑎𝑚2 … 𝑎𝑚𝑛
A is a basis of a lattice
Public values
Alice‘s values Alice Bob Bob‘s values
A q
Q(CaSC) / Johanna Sepúlveda 40
Crypto
Public Key Cryptography (PKC)
Use of lattices in the design of cryptographic functions
•Complexity of solving lattice problems in
• n-dimentional lattices grows superpolynomially
or exponentially in n
n: Security parameter
Q(CaSC) / Johanna Sepúlveda 41
Section 4A
LWE CRYPTOGRAPHY
Q(CaSC) / Johanna Sepúlveda 42
LWE
PKC Using Lattices: LWE Primal
Matrix A and integer q
Alice Bob
Private vector Public key
Selects e <<
𝒒 𝒗= A. 𝒙 Selects
𝟒
Multiplying by the inverse 𝑨−𝟏
𝒔 𝒆 𝒗 𝑨−𝟏 −𝟏 𝒙 Binary/ternary vector
𝒃𝟏= A. 𝒙 𝑨
Calculate (Key Generation) 𝒗 𝑨−𝟏 = 𝒙 Calculate (Encryption)
𝒃𝟏 = A. 𝒔 + 𝒆 -----
𝒃𝟏 . 𝒙 = (A. 𝒔 + 𝒆) . 𝒙 𝒗 = A. 𝒙
𝒗= (A.𝒃𝟐
𝒔)𝒗 𝑨−𝟏 + 𝒆 . 𝒙 𝒒
Calculate (Decryption) 𝒃𝟐 = 𝒃𝟏 𝒙 + 𝒎
= 𝒔.
Ciphertext
𝒗 + 𝒆 . 𝒙 𝟐
𝒒 Payload
𝒃𝟐 − 𝒔. 𝒗 = 𝒆.𝒙 + 𝒎 (Preamble)
𝟐
Bit message M to be sent
Q(CaSC) / Johanna Sepúlveda
0 or 1 43
LWE
PKC Using Lattices: LWE Primal
Matrix A and integer q
Alice Bob
𝒒
Private vector e <<
𝟒
𝒒
𝒃𝟐 − 𝒔. 𝒗 = 𝒆.𝒙 + 𝒎
𝟐
Small Small only if m= 0
To transfer M, b2 has to be calculated and sent
multiple times
Q(CaSC) / Johanna Sepúlveda 44
LWE
Security Proof
Matrix A and integer q
Alice Bob
𝒒
e <<
𝟒
Public key
𝒃𝟏 = A. 𝒔 + 𝒆 𝒒
𝒃𝟐 = 𝒃𝟏 𝒙 +𝒎
𝒃𝟏 : 𝐑𝐚𝐧𝐝𝐨𝐦 𝐮𝐧𝐢𝐟𝐨𝐫𝐦 𝐯𝐚𝐥𝐮𝐞 𝟐
• Very hard problem LWE- decision Problem
A 𝒃𝟏
• If 𝒃𝟏 is random, it can be proven that Bob and Alice encrypt as
OTP case (Ciphertext Uniform random)
𝒃𝟐 𝒗
Q(CaSC) / Johanna Sepúlveda 45
Crypto
PKC Using Lattices: Dual
Matrix A and integer q
Alice Bob
Binary/ternary vector 𝒒
e <<
Public key 𝟒
Selects (uniform m> n log q) Selects
𝒙 𝒗= A. 𝒙
𝒗
Multiplying by the inverse 𝑨−𝟏
𝒔 𝒆𝟏 𝒆𝟐
Calculate (Key Generation) 𝒗 𝑨−𝟏 = A. 𝒙 𝑨−𝟏 Calculate (Encryption)
𝒗 𝑨−𝟏 = 𝒙
𝒗 = A. 𝒙 𝒃𝟏 = A. 𝒔 + 𝒆𝟏
𝒃𝟏 ----- 𝒃𝟐
𝒃𝟏 . 𝒙 = (A. 𝒔 + 𝒆𝟏 ) . 𝒙 𝒒
Calculate (Decryption)
−𝟏
𝒃 𝟐 =𝒔. 𝒗 + 𝒆 𝟐 + 𝒎
= (A. 𝒔)𝒗 𝑨 + 𝒆 . 𝒙 𝟐
𝒃𝟏 . 𝒙 = 𝒔. 𝒗 + 𝒆𝟏 .𝒙 Ciphertext Payload 𝟏
= 𝒔. 𝒗 + 𝒆𝟏 . 𝒙
(Preamble)
𝒒
𝒃𝟐 − 𝒃𝟏 . 𝒙 = 𝒆𝟐 − 𝒆𝟏 .𝒙 + 𝒎
𝟐
Bit message M to be sent
Q(CaSC) / Johanna Sepúlveda
0 or 1 46
LWE
PKC Using Lattices: Dual
Matrix A and integer q
Alice Bob
𝒒
e <<
𝟒
𝒒
𝒃𝟐 − 𝒃𝟏 .𝒙 = 𝒆𝟐 − 𝒆𝟏 .𝒙 + 𝒎
𝟐
Small Small only if m= 0
To transfer M, b2 has to be calculated and sent
multiple times
Q(CaSC) / Johanna Sepúlveda 47
LWE
Comparison LWE Primal and LWE Dual
Primal Dual
Keys Keys
Public key Public key
𝑨. 𝒃𝟏 = A. 𝒔 + 𝒆 𝑨 . 𝒗 = A. 𝒙
Pseudorandom Random
Secret key Secret key
𝒔 𝒙
Unique Many
Ciphertext Ciphertext
𝒗 = A. 𝒙 𝒃𝟏 = A. 𝒔 + 𝒆𝟏
𝒒 𝒒
𝒃𝟐 = 𝒃𝟏 𝒙 + 𝒎 𝒃𝟐 =𝒔. 𝒗 + 𝒆𝟐 + 𝒎
𝟐 𝟐
Fresh LWE sample with many encryption Many LWE samples with unique
coins 𝒙 encryption coins 𝒔
Q(CaSC) / Johanna Sepúlveda 48
LWE
Traditional Digital Signatures
Verify the integrity of the communication
Public values
Alice‘s values Alice Bob Bob‘s values
Message Signature Sign
Signature
calculated
Verification: Compare the received signature (green) with
the calculated signature (they should match)
Q(CaSC) / Johanna Sepúlveda 49
LWE
Traditional Digital Signatures
Verify the integrity of the communication
Public values
Alice‘s values Alice Bob Bob‘s values
Message Signature Sign
Signature
calculated
Verification: Compare the received signature (green) with
the calculated signature (they should match)
Q(CaSC) / Johanna Sepúlveda 50
LWE
Tool: Trapdoor Function
Public function f
Secret trapdoor function f -1
f
D
R
x y
f -1
Hash and Sign
𝒚 = 𝑯 𝒎𝒆𝒔𝒔𝒂𝒈𝒆
𝑺𝒊𝒈𝒏𝒂𝒕𝒖𝒓𝒆 = 𝒇−𝟏 (𝒚)
Q(CaSC) / Johanna Sepúlveda 51
LWE
Lattice-based Signatures
Generate uniform 𝑣𝑘 = 𝐴 with a secret trapdoor 𝑠𝑘 = 𝑇
Calculate 𝑆𝑖𝑔𝑛(𝑇, 𝜇):
Use T to sample a short 𝑧 ∈ ℤ𝑛 s.t. 𝐴𝑧 = 𝐻 𝜇 ∈ ℤ𝑛𝑞
Verify (A, μ, z)
Check 𝐴𝑧 = 𝐻 𝜇 and that 𝑧 is short 𝐴𝑧 = 𝐻 𝜇 ∈ ℤ𝑛𝑞
Security: To forge a signature for a new message μ*, requires to find a
short 𝑧 ∗ s.t. 𝐴𝑧 ∗ = 𝐻 𝜇∗ ∈ ℤ𝑛𝑞 (SIS hard)
f
D
R
x y
f -1
Q(CaSC) / Johanna Sepúlveda 52
Section 4B
NTRU CRYPTOGRAPHY
Q(CaSC) / Johanna Sepúlveda 53
NTRU
PKC Using Lattices: NTRU
Public-key-based cryptography
– Shortest Vector Problem (SVP)
– Polynomials in the Ring
(Nth Degree Truncated Polynomial Ring Units)
𝑅𝑁 , 𝑞 = 𝕫𝑞 𝑥
𝑥𝑁 − 1
Operations: Addition and Multiplication
NTRU steps
1. Key generation
2. Encryption
IEEE 1363.1 (2009)
3. Decryption NTRUEncrypt lattice-based
public-key encryption algorithm
Q(CaSC) / Johanna Sepúlveda 54
NTRU
NTRU: Key Generation
Public values
Alice‘s values Alice N, p, q Bob
1. Generate two random polynomials
𝑟
𝑓 𝑅𝑁
𝑟
𝑔 𝑅𝑁
2. Compute inverses
𝑓𝑝 ∗ 𝑓 = 1 𝑚𝑜𝑑 𝑝
𝑓𝑞 ∗ 𝑓 = 1 𝑚𝑜𝑑 𝑞
3. Store private key (sk)
𝒔𝒌 = 𝒇, 𝒇𝒑
4. Compute the public key (h)
𝒉 = 𝒑 𝒇𝒒 ∗ 𝒈 𝒎𝒐𝒅 𝒒
Q(CaSC) / Johanna Sepúlveda 55
NTRU
NTRU: Encryption
Public values
Alice‘s values Alice N, p, q Bob
𝒉 = 𝒑 𝒇𝒒 ∗ 𝒈 𝒎𝒐𝒅 𝒒
𝒔𝒌 = 𝒇, 𝒇𝒑
m
1. Format m as a ternary
polynomial
m(x)
2. Generate a random polynomial
r and mask (Padding scheme)
3. Compute the ciphertext e(x)
e(x)= 𝒓 ∗ 𝒉 + 𝒎′ 𝒎𝒐𝒅 𝒒
Q(CaSC) / Johanna Sepúlveda 56
NTRU
NTRU: Decryption
Public values
Alice‘s values Alice N, p, q Bob
𝒉 = 𝒑 𝒇𝒒 ∗ 𝒈 𝒎𝒐𝒅 𝒒
𝒔𝒌 = 𝒇, 𝒇𝒑
e(x)= 𝒓 ∗ 𝒉 + 𝒎′ 𝒎𝒐𝒅 𝒒
1. Compute
𝑎 = 𝑓 ∗ 𝒆 𝒙 𝑚𝑜𝑑 𝑞
𝑏 = 𝑎 𝑚𝑜𝑑 𝑝
2. Remove the mask
m’ = 𝑓𝑝 ∗ 𝑏 𝑚𝑜𝑑 𝑝
m = 𝑚′ − 𝑴𝑮𝑭 𝑒 − 𝑚′
3. Verify there are not CCA
Q(CaSC) / Johanna Sepúlveda 57
Section 6
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 58
Tasks
Summary: Lattice-based Cryptography
Lattice-based cryptography is based on different hard
lattice problems (SVP, CVP, LWE)
Different constructions are possible (PKC, Signatures)
LWE presents a richer cryptographic space (many
parameters can be configured). However, the
security proofs become more complex than when
compared to the SVP
Q(CaSC) / Johanna Sepúlveda 59
Tasks
Lattice PQC: Answer Moodle Questions
https://eprint.iacr.org/2017/634.pdf
Answer Quiz
Optional-Optional: Look the implementation at this site (try to compile the program)
https://newhopecrypto.org/
Send print screen of the compilation result60
Q(CaSC) / Johanna Sepúlveda
Tasks
Additional Links
Matrix calculator
https://www.mathsisfun.com/algebra/matrix-calculator.html
Formal Lattice
https://ocw.mit.edu/courses/mathematics/18-409-topics-in-theoretical-
computer-science-an-algorithmists-toolkit-fall-2009/lecture-
notes/MIT18_409F09_scribe18.pdf
https://cseweb.ucsd.edu/classes/sp07/cse206a/lec2.pdf
Example Lattice Implementation (video)
https://asecuritysite.com/encryption/lwe_output
https://www.youtube.com/watch?v=hN5TQiz2gWs
Q(CaSC) / Johanna Sepúlveda 61
Tasks
Next Lecture
Third Block
Post Quantum Security Post-Quantum
Cryptography
(PQC)
LATTICE-BASED CRYPTOGRAPHY Main Principles
Families and Types
(Implementation) Main Components
Standard
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 62
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Goal Today‘s Lecture
Summary of Previous Lecture
Lattice Example and RLWE
Operations: Addition and
Multiplication, Random Sampler
Real Life Lattices
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 2
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 3
Summary
PQC: Comparison
Code-based Hash-based Multivariate Isogeny Lattice-based
(e.g. Merkle (e.g. NTRU, LWE, RLWE)
(e.g. McEliece, Rollo) (e.g. Rainbow, LUOV) (e.g. SIKE)
Hash-trees)
Pros: Pros: Pros: Pros: Pros:
▪ Well studied ▪ Well studied ▪ Multipurpose ▪ Elliptic-based ▪ Efficient
problem problem ▪ Very efficient ▪ Smallest key ▪ Public
▪ Multipurpose ▪ Very efficient signature sizes key, digital
▪ Fast schemes signatures
, FHE, IBE
Cons: Cons: Cons: Cons: Cons:
▪ Very large ▪ No encryption ▪ Most public key ▪ Low efficiency ▪ Key sizes
schemes
key sizes schemes are ▪ Difficult to when
▪ Track of signed broken construct compared to
messages
classical
crypto
Q(CaSC) / Johanna Sepúlveda 4
Summary
Lattice Hard Problems
V´
c c
Shortest Vector Closest Vector Learning
Problem Problem with errors
More configuration parameters (Rich exploration trade-off)
Easy to proof the security
Q(CaSC) / Johanna Sepúlveda 5
Summary
Taxonomy of Lattices
Lattices
NTRU
LWE
(SVP)
Ring
RLWE ILWE MLWE LWR LWP RLWR
Replaces n dimensional
vectors by polynomials of
degree smaller than n
Q(CaSC) / Johanna Sepúlveda 6
Summary
We are Ready to Start..
Post-Quantum
Cryptography
Main Principles
Families and Types
Standard
Main Components
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 7
Section 2
LATTICE EXAMPLE AND
Ring-LWE
Q(CaSC) / Johanna Sepúlveda 8
Ex & RLWE
LWE
q = 13
m=7
n=4
Random - Lattice Secret Result
4 1 11 10 6 4
5 5 9 5 9 8
3 9 0 10 11 1
x =
1 3 3 2 11 10
12 7 3 4 4
6 5 11 4 12
3 3 5 0 9
This expression can be solved easily…For example using Gaussian
elimination
Q(CaSC) / Johanna Sepúlveda 9
Ex & RLWE
LWE
q = 13
m=7
n=4
Random - Lattice Secret Small Noise Result
4 1 11 10 6 0 4
5 5 9 5 9 -1 7
3 9 0 10 11 1 2
x + =
1 3 3 2 11 1 11
12 7 3 4 1 5
6 5 11 4 0 12
3 3 5 0 1 8
This expression become difficult to solve if we do not know secret and
the noise… Q(CaSC) / Johanna Sepúlveda 10
Ex & RLWE
LWE
q = 13
m=7
n=4
Random - Lattice Secret Small Noise Result
4 1 11 10 4
5 5 9 5 8
3 9 0 10 1
x + =
1 3 3 2 10
12 7 3 4 4
6 5 11 4 12
3 3 5 0 9
LWE Search Problem: Given A and r find the secret and/or the error…
Q(CaSC) / Johanna Sepúlveda 11
Ex & RLWE
LWE
q = 13
m=7
n=4
Random - Lattice Secret Small Noise Result
4 1 11 n 10 4
u9 tio
5 5
t r ib 5
on
8
s i
Di u t
3 m 9 0 10
trib+ 1
for
i
x
Di
s =
Un 1 3 3 2 n 10
s sia
12 7 3 4 a u 4
G n
o
6 5 11 4 uti 12
3 3 5 0 is trib 9
n D
s ia
u s
G a
LWE Decision Problem: Given A distinguish r from a random …
Q(CaSC) / Johanna Sepúlveda 12
Ex & RLWE
LWE: Real Dimensions
q = 13 q = 4093
m=7 m = 640
n=4 n = 256
Random - Lattice
4 1 11 10
5 5 9 5
3 9 0 10
1 3 3 2
12 7 3 4
6 5 11 4
3 3 5 0
Q(CaSC) / Johanna Sepúlveda 13
Ex & RLWE
Ring-LWE
3= -10 mod 13
Random - Lattice
❑ Implements ring structure
4 1 11 10 ❑ Each row is the cyclic shift of
3 4 1 11 the row above
2 3 4 1
❑ Special wrapping rule:
12 2 3 4
x wraps to
9 12 2 3 Xnew= –x mod q
10 9 12 2
11 10 9 12
We do not need to store all the matrix A…just the generator vector
Q(CaSC) / Johanna Sepúlveda 14
Section 3a
OPERATIONS:
ADDITION AND MULTIPLICATION
Q(CaSC) / Johanna Sepúlveda 15
Operation:+*
Identify Operations
Polynomial Multiplication
Polynomial Addition Schoolbook NTT
Karatsuba Toom–Cook
Next
Lecture
Discrete Sampling
Uniform Binomial Modular Reduction
Gaussian Trinary Next
Lecture
Q(CaSC) / Johanna Sepúlveda 16
Operation:+*
Polynomial Addition
Parameters: q=5
n=4
= (4,2,0,1) 4 2 0 1
= (2,1,4,0)
2 1 4 0
6 3 4 1
1 3 4 1
Q(CaSC) / Johanna Sepúlveda 17
Operation:+*
Polynomial Multiplication - Schoolbook
Parameters: q=5
n=4
= (4,2,0,1)
= (2,1,4,0)
a) Multiply then reduce b) Multiply and reduce
0 0 0 0 0 0 0 0
16 8 0 4 8 0 4 -16
4 2 0 1 0 1 -4 -2
8 4 0 2 2 -8 -4 0
8 8 18 10 1 4 0 10 -7 -4 -18
-8 -8 -18
0 3 1 2 0 3 1 2
Q(CaSC) / Johanna Sepúlveda 18
Operation:+*
Polynomial Multiplication - Karatsuba
Parameters: q=5
n=4
❑ Divide and Conquer (Anatoly Karatsuba in 1960)
The multiplication is
Where the value of the coefficients is
Q(CaSC) / Johanna Sepúlveda 19
Operation:+*
Polynomial Multiplication - Karatsuba
Parameters: q=5
n=4
= (4,2,0,1) a = 42
b = 01
= (2,1,4,0) c = 21
d = 40
The multiplication is
Where the value of the coefficients is
Q(CaSC) / Johanna Sepúlveda 20
Challenge: How do we reduce? So we can get the answer we got with the schoolbook?
Operation:+*
Polynomial Multiplication - NTT
Q(CaSC) / Johanna Sepúlveda Butterfly 21
Operation:+*
Polynomial Multiplication - NTT
❑ Forward Transformation: ❑ Backward Transformation:
Pre-Processi Twiddle Coefficient Post-Processi
ng Factors s ng
Butterfly
Operation
Update Twiddle
Factor
Q(CaSC) / Johanna Sepúlveda 22
Operation:+*
Polynomial Multiplication - NTT
Q(CaSC) / Johanna Sepúlveda 23
Operation:+*
Polynomial Multiplication – NTT (NIST 2nd round)
Algorithm with Usage of NTT Dimension Modulo
(n) (q)
CRYSTALS-Dilithium 256 8380417
Signatures FALCON 512/1024 12289
qTESLA 1024/2048 8058881/12681217
CRYSTALS-KYBER 256 7681
Ding Key Exchange 512/1024 120833
NewHope 512/1024 12289
Key
Exchange/ R.EMBLEM 512 12289
Encryption Hila5 / Round5 1024 12289
KCL 256 7681
Titanium 1024/2048 86017/1198081
NIST 2nd round announcement
NIST 3rd round
Q(CaSC) / Johanna Sepúlveda 24
Operation:+*
Polynomial Multiplication: Recommendation
NTT friendly
Multiplication should be implemented in a secure way to avoid the leak of information
Q(CaSC)(sensitive
/ Johanna Sepúlveda
factors) 25
Section 3b
OPERATIONS:
RANDOM SAMPLERS
Q(CaSC) / Johanna Sepúlveda 26
Random
Random Samplers
❑ Different PQC schemes have different parameters of the random
distributions (shape/type, size, mean, fixed/changing, statistical
distance)
▪ Type: Uniform, Gaussian, Binomial
▪ Operations (floating point)
▪ Memory
▪ Execution time
Challenges
❑ Efficiency: Time consuming operation (Gaussian sampling takes up
50% of the running time for some schemes)
❑ Flexibility: Should be supported by constrained devices
▪ Memory (kilobytes instead of gigabytes)
▪ Not all devices have floating point arithmetic capability
Samplers should be implemented in a/secure
Q(CaSC) way to avoid the leak of information
Johanna Sepúlveda 27
Random
Random Samplers: Statistical Distance
center c
The statistical distance between the desired
distribution and the output distribution should be negligible
Q(CaSC) / Johanna Sepúlveda 28
Random
Random Samplers: Types
Knuth-Yao
Rejection
Sampling
SHAKE
Inversion
SHA-3
Cumulative
Keccak Distribution
Ziggurat Table (CDT)
Leftover Hash Lemma (LHL)
Hash functions are good Bernoulli
randomness extractors Sampler
Suggested Reading: https://www.sav.sk/journals/uploads/0212094402follat.pdf
Q(CaSC) / Johanna Sepúlveda 29
Random
Sponge as Random Sources
KECCAK Discrete Samplers
SHA-3 SHAKE
Sponge construction is based on a wide random function and allows
absorbing any amount of data and squeezing any amount of data.
It acts as a pseudorandom function with regard to all previous inputs.
Bit Rate
Capacity
Leftover Hash Lemma (LHL)
Hash functions are good
Q(CaSC) / Johanna Sepúlveda 30
randomness extractors https://nvlpubs.nist.gov/nistpubs/FIPS/NIST.FIPS.202.pdf
Random
Rejection Sampling
A large proportion of samples will be rejected
We must evaluate f(x) for each candidate point (computationally expensive)
Q(CaSC) / Johanna Sepúlveda 31
Random
Discrete Ziggurat
Q(CaSC) / Johanna Sepúlveda 32
Random
Knuth-Yao: DDG Tree
Discrete Distribution Generating tree
It traverses the tree uniformly randomly (root - terminal)
Internal node The algorithm outputs the label of the node
(2 children: 0/1)
Terminal node
(Integer label)
Q(CaSC) / Johanna Sepúlveda 33
Section 5
LATTICE-BASED IN REAL LIFE
Q(CaSC) / Johanna Sepúlveda 34
Real Life
Hybrid approaches for TLS
X25519 (ECDH) +PQC (KEMs)
Support: X25519 (ECDH), PQC Shared secret 1: f(X25519)
Keys (ephemeral: ECDH, PQC) Shared secret 2: KEM
encapsulation
Shared secret 1: f(X25519)
Shared secret 2: KEM decapsulation
PK (X25519)
KEM Encapsulation
Master secret: Shared secret 1|| Shared secret 2
Q(CaSC) / Johanna Sepúlveda 35
Real Life
Google Experiment (2016)
❑ Enhance of BoringSSL with Post-Quantum algorithms
2016: NewHope (RLWE) CECPQ1
2019: NTRUHRSS (NTRU) CECPQ2
❑ Layered approach
❑ X25519: 128-bits security Elliptic curve (ECDH)
❑ PQ algorithm
“[. . . ] we did not find any unexpected
impediment to deploying something like
NewHope. There were no reported
problems caused by enabling it.”
“…median connection latency only
increased by 1 ms, the latency for the
slowest 5% increased by 20ms and, for
the slowest 1%, by 150ms...”
Q(CaSC) / Johanna Sepúlveda 36
Real Life
Google Experiment (2019)
❑ Enhance TLS 1.3 with Post-Quantum algorithms (53 days)
SIKE
Small key size
NTRU-HRSS Large computational
Larger key size cost
Small computational
cost
Q(CaSC) / Johanna Sepúlveda 37
Real Life
Google Experiment (2019)
Q(CaSC) / Johanna Sepúlveda 38
Real Life
Google Experiment (2019)
The small key size of CECPQ2b does not make up for
its large computational cost
… But not true for slowest connections (Windows computers, Android
mobile devices). One possible explanation for this is packet
fragmentation and packet loss
Q(CaSC) / Johanna Sepúlveda 39
Section 6
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 40
Tasks
Summary: Lattice Implementation
Lattice-based cryptography is based on four main
operations: addition, multiplication, random sampling
and modular reduction
Operations that deal with the secret information
should be implemented in a secure way
(First line: Constant-time/ independent of the secret)
Many alternatives: Different trade-offs
Q(CaSC) / Johanna Sepúlveda 41
Tasks
NTT Implementation: Answer Moodle Questions
https://ieeexplore.ieee.org/document/8741027
Answer Quiz
Q(CaSC) / Johanna Sepúlveda 42
Tasks
Additional Links
Samplers
https://www.sav.sk/journals/uploads/0212094402follat.pdf
https://www.youtube.com/watch?v=DOC8aamM57M
Q(CaSC) / Johanna Sepúlveda 43
Tasks
Next Lecture
Third Block
✔ Post Quantum Security Post-Quantum
Cryptography
(PQC)
CODE-BASED CRYPTOGRAPHY Main Principles
Families and Types
Main Components
Standard
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 44
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
10. Code-Based Post-Quantum Cryptography
Q(CaSC) / Johanna Sepúlveda 2
Goal Today‘s Lecture
Summary of Previous Lecture
Error Correcting Codes
Linear Codes
Code-based Cryptography
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 3
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 4
Summary
Traditional Security
Today traditional cryptography
is considered secure
Factoring large integers Computing discrete logarithms
Traditional Cryptography
(RSA, ECC)
Cryptoanalysis
Side-Channel
Attacker
IonQ’s 79 qubits Quantum Processor
In the future, traditional cryptography
will be broken
Q(CaSC) / Johanna Sepúlveda 5
Summary
Post-Quantum Cryptography
Distribute Key
Post-Quantum Cryptography Encrypt
Decrypt
Classical Network Sign
Verify
Code Hash Multivariate Isogeny Lattice
Algorithms used to secure messages on a classical computer and that are resistant to
traditional computers attacks AND quantum computers attacks
Q(CaSC) / Johanna Sepúlveda 6
Summary
PQC: Comparison
Code-based Hash-based Multivariate Isogeny Lattice-based
(e.g. NTRU, LWE, RLWE)
(e.g. McEliece, Rollo) (e.g. Merkle Hash- (e.g. Rainbow, LUOV) (e.g. SIKE)
trees)
Pros: Pros: Pros: Pros: Pros:
Well studied Well studied Multipurpose Elliptic-based Efficient
problem problem Very efficient Smallest key Public
Multipurpose Very efficient signature sizes key, digital
Fast schemes signatures
, FHE, IBE
Cons: Cons: Cons: Cons: Cons:
Very large No encryption Most public key Low efficiency Key sizes
key sizes schemes schemes are Difficult to when
Track of signed broken construct compared to
messages
classical
crypto
Q(CaSC) / Johanna Sepúlveda 7
Summary
NIST Post-Quantum Candidates
Code Hash Multivariate Isogeny Lattice
First NIST
18 3 9 2 32
Round
Second NIST
7 2 4 1 12
Round
Q(CaSC) / Johanna Sepúlveda 8
Summary
Lattices: Scary and Friendly Basis
An n-dimensional
Dimension lattice ℒ is
• Generated by a basis
ℬ = 𝑏1 , … , 𝑏𝑛 Rank
ℬ𝑚𝑥𝑛
• ℒ = 𝑛𝑖=1(ℤ ⋅ 𝑏𝑖 )
𝑏2 Scary Basis: Long
ℒ(ℬ) = 𝑩𝒙 | 𝒙 ∈ ℤ𝑛 𝑏1 vectors
Friendly Basis: Short
vectors
Basis is not unique:
There are multiple basis that can generate the same lattice
Q(CaSC) / Johanna Sepúlveda 9
Summary
Lattice Hard Problems
V´
𝑠 c c
Shortest Vector Closest Vector Learning
Problem Problem with errors
More configuration parameters (Rich exploration trade-off)
Easy to proof the security
Q(CaSC) / Johanna Sepúlveda 10
Summary
LWE
𝑨. 𝒔 + 𝒆 = 𝒓 q = 13
m=7
n=4
Random - Lattice Secret Small Noise Result
7𝑥4 4𝑥1 7𝑥1 7𝑥1
ℤ13 ℤ13 ℤ13 ℤ13
4 1 11 10 6 0 4
5 5 9 5 9 -1 7
3 9 0 10 11 1 2
x + =
1 3 3 2 11 1 11
12 7 3 4 1 5
6 5 11 4 0 12
3 3 5 0 1 8
This expression become difficult to solve if we do not know secret and
the noise… Q(CaSC) / Johanna Sepúlveda 11
Summary
LWE: Real Dimensions
q = 13 q = 4093
m=7 m = 640
Random - Lattice 𝑨. 𝒔 + 𝒆 = 𝒓 n=4 n = 256
7𝑥4
ℤ13
4 1 11 10
5 5 9 5
3 9 0 10
1 3 3 2
12 7 3 4
6 5 11 4
3 3 5 0
Security Parameters: * Dimension n (~256 - 1024)
* Modulo q (~211 – 232)
* Gaussian 𝝈
Q(CaSC) / Johanna Sepúlveda 12
Summary
Ring-LWE
3= -10 mod 13
Random - Lattice
7𝑥4
ℤ13
Implements ring structure
4 1 11 10 Each row is the cyclic shift of
3 4 1 11 the row above
2 3 4 1
Special wrapping rule:
12 2 3 4
x wraps to
9 12 2 3 Xnew= –x mod q
10 9 12 2
11 10 9 12
We do not need to store all the matrix A…just the generator vector
Q(CaSC) / Johanna Sepúlveda 13
Summary
Identify Operations
Polynomial Multiplication
Polynomial Addition Schoolbook NTT
Karatsuba Toom–Cook
Discrete Sampling
Uniform Binomial Modular Reduction
Gaussian Trinary
Q(CaSC) / Johanna Sepúlveda 14
Summary
Polynomial Addition
𝒗+𝒌=𝒄
Polynomial Addition
ℤ𝑞 𝑥
Parameters: q=5 𝑅= 𝑛
𝑥 +1
n=4
𝒗 = 4𝑥 3 + 2𝑥 2 + 1 = (4,2,0,1) 4 2 0 1
𝒌 = 2𝑥 3 + 𝑥 2 + 4𝑥 = (2,1,4,0)
2 1 4 0
6 3 4 1
𝑚𝑜𝑑 5
𝒄=𝒗+𝒌 1 3 4 1
𝒄 = 𝑥 3 + 3𝑥 2 + 4𝑥 + 1 = (𝟏, 𝟑, 𝟒, 𝟏)
Q(CaSC) / Johanna Sepúlveda 15
Summary
Polynomial Multiplication - Schoolbook
Polynomial
Parameters: q=5 𝒗. 𝒌 = 𝒍 Multiplication
n=4 ℤ𝑞 𝑥
𝑅= 𝑛
𝑥 +1
𝒗 = 4𝑥 3 + 2𝑥 2 + 1 = (4,2,0,1)
𝒌 = 2𝑥 3 + 𝑥 2 + 4𝑥 = (2,1,4,0)
a) Multiply then reduce b) Multiply and reduce
0 0 0 0 0 0 0 0
16 8 0 4 8 0 4 -16
4 2 0 1 0 1 -4 -2
8 4 0 2 2 -8 -4 0
8 8 18 10 1 4 0 10 -7 -4 -18
𝑚𝑜𝑑 5
-8 -8 -18
𝑚𝑜𝑑 𝑥 𝑛 + 1
0 3 1 2 0 3 1 2
l = 3𝑥 2 + 𝑥 +Q(CaSC)
𝟐 =/ Johanna
(𝟎, 𝟑, 𝟏, 𝟐)
Sepúlveda 16
Summary
Polynomial Multiplication - Karatsuba
Polynomial
Parameters: q=5 𝒗. 𝒌 = 𝒍 Multiplication
n=4 ℤ𝑞 𝑥
𝑅= 𝑛
𝑥 +1
𝒗 = 4𝑥 3 + 2𝑥 2 + 1 = (4,2,0,1) a = 42
𝒌 = 2𝑥 3 + 𝑥 2 + 4𝑥 = (2,1,4,0)
b = 01
c = 21
d = 40
𝑣 = 𝑎𝐵𝑚 + 𝑏 = 42 (100) +01
𝑘 = 𝑐𝐵𝑚 + 𝑑 = 21 (100) +40 𝒪 (𝑛𝑙𝑜𝑔23 )
The multiplication is 𝑚𝑜𝑑 5
𝑚𝑜𝑑 𝑥 𝑛 + 1
𝑥𝑦 = 𝑎𝐵𝑚 + 𝑏 𝑐𝐵𝑚 + 𝑑
R𝑒𝑠𝑢𝑙𝑡 = 𝑧2 𝐵2𝑚 +𝑧1 𝐵𝑚 + 𝑧0 = 882. 1002 + 1701.100 +40
= 8990140
Where the value of the coefficients is
𝑧2 = 𝑎𝑐 = 42.21 = 882 𝐵𝑚 ≤ 𝑅𝑒𝑠𝑢𝑙𝑡 ≤ 2𝐵𝑚
𝑧1 = 𝑎𝑑 + 𝑏𝑐 𝑧1 = 𝑎 + 𝑏 𝑐 + 𝑑 − 𝑧2 − 𝑧0
𝑧0 = 𝑏𝑑 = 1.40 = 40 = 42 + 1 21 + 40 − 882 −40
= 1701
Q(CaSC) / Johanna Sepúlveda 17
Polynomial Multiplication - NTT 𝒗. 𝒌 = 𝒍 Summary
Polynomial
Multiplication
Number Theoretic Transform (NTT) ℤ𝑞 𝑥
𝑅= 𝑛
𝑥 +1
Discrete Fourier Transform (DFT) defined over a
finite field or ring
NTT exists if q is a prime, n a power of two and if
𝒒 = 1 𝑚𝑜𝑑 2𝒏
𝒄 = 𝑁𝑇𝑇 −1 𝑁𝑇𝑇 𝒂 𝑁𝑇𝑇 𝒔
denotes coefficient-wise multiplication
Q(CaSC) / Johanna Sepúlveda Butterfly 18
Summary
Polynomial Multiplication - NTT
Bit Reversal
Cooley-Tukey
Q(CaSC) / Johanna Sepúlveda 19
Summary
Polynomial Multiplication – NTT
𝑛−1
𝑖𝑗
𝑎𝑖 = 𝜔𝑛 ∙ 𝑎𝑗
𝑗=0 Butterfly
Operation
Update Twiddle Factor
Q(CaSC) / Johanna Sepúlveda 20
Summary
Polynomial Multiplication – NTT
NTT and NTT-1 hardware
Q(CaSC) / Johanna Sepúlveda 21
Summary
Polynomial Multiplication – NTT (NIST 2nd round)
Algorithm with Usage of NTT Dimension Modulo
(n) (q)
CRYSTALS-Dilithium 256 8380417
Signatures FALCON 512/1024 12289
qTESLA 1024/2048 8058881/12681217
CRYSTALS-KYBER 256 7681
Ding Key Exchange 512/1024 120833
Key NewHope 512/1024 12289
Exchange/ R.EMBLEM 512 12289
Encryption Hila5 / Round5 1024 12289
KCL 256 7681
Titanium 1024/2048 86017/1198081
NIST 2nd round announcement
Q(CaSC) / Johanna Sepúlveda 22
Summary
Random Samplers: Types
Knuth-Yao
Rejection
Sampling
SHAKE
Inversion
SHA-3
Cumulative
Keccak Distribution
Ziggurat Table (CDT)
Leftover Hash Lemma (LHL)
Hash functions are good Bernoulli
randomness extractors Sampler
Suggested Reading: https://www.sav.sk/journals/uploads/0212094402follat.pdf
Q(CaSC) / Johanna Sepúlveda 23
Summary
Hybrid approaches for TLS
X25519 (ECDH) +PQC (KEMs)
Support: X25519 (ECDH), PQC Shared secret 1: f(X25519)
Keys (ephemeral: ECDH, PQC) Shared secret 2: KEM
encapsulation
Shared secret 1: f(X25519)
Shared secret 2: KEM decapsulation
PK (X25519)
KEM Encapsulation
Master secret: Shared secret 1|| Shared secret 2
Q(CaSC) / Johanna Sepúlveda 24
Summary
Google Experiment (2016)
Enhance of BoringSSL with Post-Quantum algorithms
2016: NewHope (RLWE) CECPQ1
2019: NTRUHRSS (NTRU) CECPQ2
Layered approach
X25519: 128-bits security Elliptic curve (ECDH)
PQ algorithm
“[. . . ] we did not find any unexpected
impediment to deploying something like
NewHope. There were no reported
problems caused by enabling it.”
“…median connection latency only
increased by 1 ms, the latency for the
slowest 5% increased by 20ms and, for
the slowest 1%, by 150ms...”
Q(CaSC) / Johanna Sepúlveda 25
Summary
Google Experiment (2019)
Q(CaSC) / Johanna Sepúlveda 26
Summary
Google Experiment (2019)
The small key size of CECPQ2b does not make up for
its large computational cost
… But not true for slowest connections (Windows computers, Android
mobile devices). One possible explanation for this is packet
fragmentation and packet loss
Q(CaSC) / Johanna Sepúlveda 27
Summary
NIST Lattice-Based Schemes
KEM/PKC
Crystals-Kyber Saber
LAC
QTesla
Frodo Round5
ThreeBears Crystals-
Dilithium
NewHope NTRU Signatures
NTRU Prime Falcon
Q(CaSC) / Johanna Sepúlveda 28
Summary
We are Ready to Start..
Post-Quantum
Cryptography
Main Principles
Families and Types
Standard
Main Components
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 29
Section 2
ERROR CORRECTING CODES
Q(CaSC) / Johanna Sepúlveda 30
ECC
Noisy Communication Channel
“The fundamental problem of communication is
that reproducing at one point either exactly or
approximately a message selected at another
point”
Claude Shannon
(Information Theory)
I love We are
over!
you!
Communication Channels are imperfect!
Q(CaSC) / Johanna Sepúlveda 31
ECC
Error Correcting Codes
Interference
Repeat Costly
Weather (Electrical, other)
Unreliable/Noisy Channel
Repeat Imperfect
materials
Inefficient
Use unmistakable
words
NATO phonetic alphabet
Error Correcting Codes (ECC)
Binary encoding scheme that allows the message recovering even if some
bits are erroneously flipped
Error Detection Error Correction
Q(CaSC) / Johanna Sepúlveda 32
ECC
Error Correcting Codes
Errors
Message Encoding Scheme
Codeword
(Sourcecode)
Generator
(Transformation) High dimensional state
Valid
codeword
Decoding Scheme
Message
Correction rules
Invalid
codeword
Codespace
Q(CaSC) / Johanna Sepúlveda 33
ECC
Error Correcting Codes: Example
Codespace
Spatial view: 16 messages are mapped on 16 valid codewords
4-Dimensional Space 7-Dimensional Space
Message Codeword Valid
codeword
Generator
(Transformation)
Q(CaSC) / Johanna Sepúlveda 34
ECC
Repetition Codes
Repeat Repeat Repeat
Can Bob correct the received message if there is
moreUnreliable/Noisy
than 1 error? Channel
Generator: Alice will
Corrector: Bob uses
send 3 times the error majority vote
information
Bob’s Received Decoded
Alice’s Information Codeword Information Information
0 000 000 0
001 0
010 0
100 0
Detect: 11error 111 111 1
011 1
101 1
Correct: 1 error Q(CaSC) / Johanna Sepúlveda 110 1 35
ECC
Repetition Codes
Repeat Repeat Repeat
Unreliable/Noisy Channel
1-Dimensional Space 3-Dimensional Space
Message Codeword
Generator
00101
(Transformation)
000 000 111 000 111
0 000 Valid
codeword
1 111
error
Q(CaSC) / Johanna Sepúlveda 36
ECC
Parity Codes
Parity bit: Makes even
Repeat
the number of 1s
Repeat
Unreliable/Noisy Channel
Odd number of 1s elucidates an error
Generator: Alice will
Corrector: Bob uses
send 2 times the error the number of ones
information
Bob’s Received Decoded
Alice’s Information Codeword Information Information
0 00 00 0
01 X
10 X
Detect: 1 error
1 11 11 1
01 X
10 X
Correct: No
Q(CaSC) / Johanna Sepúlveda 37
ECC
Hamming Codes
Analog to NATO approach
In general:
Unreliable/Noisy Channel
Detect: k-errors when HD ≥ k+1
Correct: k-errors when HD ≥ 2k+1
Hamming Distance (HD): Number of bits that are different between
two bit string
String 1: 1101010
HD= 2
String 2: 1111000
Example:
Detect: 1A error B C D
000 011 101 110
HD= 2 HD= 2 HD= 2
Correct: Depends on HD
(010): A, B or D Q(CaSC) / Johanna Sepúlveda 38
ECC
Hamming Codes
Hamming Codeword: All bit positions that are power of 2 are
parity (P) bits (1,2,4,8,…), others are data (D) bits
Example: Use the Hamming code to encode the following bit string
10011010
1. Identify the parity and the data positions
P P D P D D D P D D D D
1 2 3 4 5 6 7 8 9 10 11 12
1 0 0 1 1 0 1 0
2. Locate the data bits in the data positions
Q(CaSC) / Johanna Sepúlveda 39
ECC
Hamming Codes
3. Calculate the parity bit. Position gives the check/skip positions
Parity 1: Check 1 bit, skip 1 bit, check 1 bit,.. (1,3,5,7,9,11 = ?,1,0,1,1,1= 0)
Parity 2: Check 2 bits, skip 2 bits, check 2 bits,.. (2,3,6,7,10,11 = ?,1,0,1,0,1= 1)
Parity 4: Check 4 bits, skip 4 bits, check 4 bits,.. (4,5,6,7,12 = ?,0,0,1,0= 1)
Parity 8: Check 8 bits, skip 8 bits, check 8 bits,.. (8,9,10,11,12 = ?,1,0,1,0= 0)
0 Number of 1s is even
1 Number of 1s is odd
P P D P D D D P D D D D
1 2 3 4 5 6 7 8 9 10 11 12
0 1 1 1 0 0 1 0 1 0 1 0
Q(CaSC) / Johanna Sepúlveda 40
ECC
Hamming Codes: Decoding and Correction
error
P P D P D D D P D D D D
1 2 3 4 5 6 7 8 9 10 11 12
0 1 1 1 0 0 1 0 1 1 1 0
1. Calculate the parity bit and compare with the embedded value in the codeword
Calculate Parity 1: 1,3,5,7,9,11 = ?,1,0,1,1,1= 0 = Parity 1 in the codeword
Calculate Parity 2: 2,3,6,7,10,11 = ?,1,0,1,1,1= 0; Parity 2 in the codeword = 1
Calculate Parity 4: 4,5,6,7,12 = ?,0,0,1,0= 1 = Parity 4 in the codeword
Calculate Parity 8: 8,9,10,11,12 = ?,1,1,1,0= 1; Parity 8 in the codeword = 0
2. Calculate the position of the error bit
P2, P8 are wrong
Error position= 2+8 = 10
Q(CaSC) / Johanna Sepúlveda 41
Section 3
LINEAR CODES
Q(CaSC) / Johanna Sepúlveda 42
Linear Codes
Linear Codes: Generator Matrix G
G is a matrix that express the transformation from a
message to a valid codeword
Generator Matrix
𝑤 = 𝑚𝐺
Codeword
(high dimension)
Message vector
(low dimension)
Gkxn is the generator matrix of the linear code [n,k,d]q
n: Length of the codeword
k: Information bits
d: Minimum distance of the code
q: Size of the finite field (binary q=2)
Q(CaSC) / Johanna Sepúlveda 43
Linear Codes
Linear Codes: Parity Matrix H
H is a matrix that describes the linear relations that the
components of a codeword must satisfy in order to be
valid (useful to correct errors)
Parity Matrix
0 = 𝑤𝑇 𝐻
Syndrome
0: Valid codeword
Otherwise: invalid codeword
Codeword transponse
(column)
H has columns with every possible combination of 1s
(except column full of 0s)
Q(CaSC) / Johanna Sepúlveda 44
Linear Codes
Linear Codes: Parity Matrix H (Example)
Hamming code (4 to 7)
Message abcd
Parity bits xyz
P P D P D D D
1 2 3 4 5 6 7
x y a z b c d
𝑥 = 𝒂⨁𝒃⨁𝒅
𝑦 = 𝒂⨁𝒄⨁𝒅
𝑧 = 𝒃⨁𝒄⨁𝒅
Q(CaSC) / Johanna Sepúlveda 45
Linear Codes
Linear Codes: Generator Matrix G (Example)
𝑤 = 𝑚𝐺 𝐺𝑘𝑥𝑛
k: Information bits
n: length of the codeword
Matrix G
x y z
1 0 0 0 1 1 0
𝑤 = 𝒂𝒃𝒄𝒅 0 1 0 0 1 0 1
0 0 1 0 0 1 1
0 0 0 1 1 1 1
𝑥 = 𝒂⨁𝒃⨁𝒅
Identity Matrix 𝑦 = 𝒂⨁𝒄⨁𝒅
𝑧 = 𝒃⨁𝒄⨁𝒅
Q(CaSC) / Johanna Sepúlveda 46
Linear Codes
Linear Codes: Parity Matrix H (Example)
Message abcd
Hamming code (4 to 7) Parity bits xyz
Number of columns =
Codeword size (7)
x y a z b c d 𝑥 = 𝒂⨁𝒃⨁𝒅
𝑦 = 𝒂⨁𝒄⨁𝒅
1 0 1 0 1 0 1 𝑧 = 𝒃⨁𝒄⨁𝒅
𝐻= 0 1 1 0 0 1 1
0 0 0 1 1 1 1
Number of rows = a b c d x y z
Number parity bits (3)
1 1 0 1 1 0 0
Systematic 1 0 1 1 0 1 0
representation 0 1 1 1 0 0 1
[I] Q(CaSC) / Johanna Sepúlveda 47
Linear Codes
Linear Codes: Syndrome
𝐻𝑤𝑇=0
𝐻(𝑤 + 𝑒)𝑇= 𝐻𝑤𝑇 + 𝐻𝑒𝑇 = 0+ 𝐻𝑒𝑇
𝒂
a b c d x y z 𝒃
1 1 0 1 1 0 0 𝒄 0
1 0 1 1 0 1 0 . 𝒅 = 0
0 1 1 1 0 0 1 𝒙 0
𝒚
𝒛
Syndrome vector ONLY depends on the error, never on
the codeword
Q(CaSC) / Johanna Sepúlveda 48
Linear Codes
Linear Codes: Syndrome
𝐻𝑤𝑇=0
𝐻(𝑤 + 𝑒)𝑇= 𝐻𝑤𝑇 + 𝐻𝑒𝑇 = 0+ 𝐻𝑒𝑇
𝒂
a b c d x y z 𝒃
1 1 0 1 1 0 0 𝒄 0
1 0 1 1 0 1 0 . 𝒅 = 0
0 1 1 1 0 0 1 𝒙 0
𝒚
𝒛
Syndrome vector ONLY depends on the error, never on
the codeword (identifier of the bit with an error)
Q(CaSC) / Johanna Sepúlveda 49
Linear Codes
Linear Codes: Challenge
Using the following H matrix
1 1 1 0 1 0 0
0 1 1 1 0 1 0
1 1 0 1 0 0 1
a) Encode 1011
b) Decode 1100010
c) Decode 1110010
d) Decode 1011101
Q(CaSC) / Johanna Sepúlveda 50
Section 4
CODE HARD PROBLEMS AND
CRYPTOGRAPHY
Q(CaSC) / Johanna Sepúlveda 51
Crypto
Hard Problems in Coding Theory
General Decoding (GD) Problem: Given an [n, k] code C over Fq,
an integer t0 and a vector c ∈ 𝐹𝑞𝑛
Find a valid codeword w with d(w, c) ≤ t0
From a black point to the closest blue point (valid codeword)
Syndrome Decoding (SD) Problem: Given a matrix H and a vector
s, both over Fq, and a nonnegative integer t0;
Find a vector x ∈ 𝐹𝑞𝑛 with Hamming weight wt(x) = t0 such that
HxT = sT
NP-complete
Hamming Weight= Number of ones in a string (Search Problem)
Goppa Parameterized Syndrome Decoding (GPSD): Given a
matrix H of size 2m×r and a syndrome s, decide whether there
exists a codeword x of weight r/m such that
HxT = sT
NP-complete (Decisional Problem)
Q(CaSC) / Johanna Sepúlveda 52
Crypto
Hard Problems in Coding Theory
Goppa Code Distinguishing (GD): Given an H of size r×n, decide
whether H is the parity check matrix of a Goppa code
(Decisional Problem)
In 2013, it was showed that “high rate” binary Goppa codes can be
distinguished from random linear codes. However it does not work at
• 8 errors for n = 1024 (where McEliece used 50 errors)
• 20 errors for n = 8192 (a variant of classic McEliece).
Q(CaSC) / Johanna Sepúlveda 53
Crypto
Code-Based Cryptosystems
Codes
McEliece Niederreiter
• G of the Goppa Code is hidden 𝑮 • m: Random, small weighted error vector
(scrambling, permuting)
• Public key: H of Generalized Reed
• Public key: 𝑮 Salomon Codes (GRSC)
• Encrypt: Encode m with 𝑮 and add an • Encrypt: Encode m with Public key
error • Decrypt: Decoding the Ciphertext
• Decrypt: Decoding the Ciphertext
(Signatures, Fast)
Quasi-cyclic
Quasi-cyclic (QC)
Low Desity Parity Check
codes
(QC-LDPC) Q(CaSC) / Johanna Sepúlveda 54
Crypto
Code-Based Cryptosystems
Classic McEliece
Goppa BIKE
Short Hamming (MDPC)
NTS-KEM HQC
Goppa Short Hamming
LEDACrypt
Short Hamming (LDPC)
Rollo
LowLow Rank
Rank (LRPC)
RQC
Low Rank
Q(CaSC) / Johanna Sepúlveda 55
Crypto
Code-Based Cryptosystems: Performance
Q(CaSC) / Johanna Sepúlveda 56
Crypto
Identify Operations
Polynomial Multiplication
Matrix Addition Schoolbook
Polynomial square/
square root
Discrete Sampling
Uniform Binomial Modular Reduction
Gaussian Trinary
Q(CaSC) / Johanna Sepúlveda 57
Section 6
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 58
Tasks
Summary: Code-Based Cryptography
Code-based cryptosystems are very robust. Their
problems has been widely studied (1978)
Code-based cryptography is based on different hard
code problems (GDP, SDP, GPSD, GD)
Different constructions are possible (PKC, Signatures),
but PKC/KEM are more popular
Code-based used a huge public key. Structured codes
can decrease this size, but may influence the security.
Q(CaSC) / Johanna Sepúlveda 59
Tasks
McEliece Encryption
https://arxiv.org/pdf/1907.12754.pdf
Answer Quiz
ONLY Section 5 (pages 26-32) with the description of the protocol and the example.
Q(CaSC) / Johanna Sepúlveda 60
Tasks
Additional Links
Matrix calculator
https://www.mathsisfun.com/algebra/matrix-calculator.html
Implementation Codes
Code-based Cryptography: Implementing the McEliece Scheme on
Recongurable Hardware (Pages 37 - 48)
https://www.emsec.ruhr-uni-
bochum.de/media/crypto/attachments/files/2010/04/da_heyse.pdf
Code-based Implementation (video)
https://www.youtube.com/watch?v=FZBnox3MsUQ
Q(CaSC) / Johanna Sepúlveda 61
Tasks
Next Lecture
Third Block
Post Quantum Security Post-Quantum
Cryptography
(PQC)
HASH-BASED CRYPTOGRAPHY Main Principles
Families and Types
Main Components
Standard
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 62
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
Quantum Computers and
Quantum Secure Communications
Johanna Sepúlveda, Ph.D.
[email protected]
Q(CaSC) / Johanna Sepúlveda 1
Goal Today‘s Lecture
Summary of Previous Lecture
Hash
Hash-based Construction
NIST: Third Round
Summary and Tasks
Q(CaSC) / Johanna Sepúlveda 2
Section 1
SUMMARY OF PREVIOUS LECTURE
Q(CaSC) / Johanna Sepúlveda 3
Summary
PQC: Comparison
Code-based Hash-based Multivariate Isogeny Lattice-based
(e.g. Merkle (e.g. NTRU, LWE, RLWE)
(e.g. McEliece, Rollo) (e.g. Rainbow, LUOV) (e.g. SIKE)
Hash-trees)
Pros: Pros: Pros: Pros: Pros:
▪ Well studied ▪ Well studied ▪ Multipurpose ▪ Elliptic-based ▪ Efficient
problem problem ▪ Very efficient ▪ Smallest key ▪ Public
▪ Multipurpose ▪ Very efficient signature sizes key, digital
▪ Fast schemes signatures
, FHE, IBE
Cons: Cons: Cons: Cons: Cons:
▪ Very large ▪ No encryption ▪ Most public key ▪ Low efficiency ▪ Key sizes
schemes
key sizes schemes are ▪ Difficult to when
▪ Track of signed broken construct compared to
messages
classical
crypto
Q(CaSC) / Johanna Sepúlveda 4
Summary
Error Correcting Codes
Errors
Message Encoding Scheme
Codeword
(Sourcecode)
Generator
(Transformation) High dimensional state
Valid
codeword
Decoding Scheme
Message
Correction rules
Invalid
codeword
Codespace
Q(CaSC) / Johanna Sepúlveda 5
Summary
Linear Codes: Generator Matrix G
❑ G is a matrix that express the transformation from a
message to a valid codeword
Generator Matrix
Codeword
(high dimension)
Message vector
(low dimension)
❑ Gkxn is the generator matrix of the linear code [n,k,d]q
▪ n: Length of the codeword
▪ k: Information bits
▪ d: Minimum distance of the code
▪ q: Size of the finite field (binary q=2)
Q(CaSC) / Johanna Sepúlveda 6
Summary
Linear Codes: Parity Matrix H
❑ H is a matrix that describes the linear relations that the
components of a codeword must satisfy in order to be
valid (useful to correct errors)
Parity Matrix
Syndrome
0: Valid codeword
Otherwise: invalid codeword
Codeword transponse
(column)
❑ H has columns with every possible combination of 1s
(except column full of 0s)
Q(CaSC) / Johanna Sepúlveda 7
Summary
Hard Problems in Coding Theory
Q(CaSC) / Johanna Sepúlveda 8
Summary
Hard Problems in Coding Theory
❑ Goppa Code Distinguishing (GD): Given an H of size r×n, decide
whether H is the parity check matrix of a Goppa code
(Decisional Problem)
In 2013, it was showed that “high rate” binary Goppa codes can be
distinguished from random linear codes. However it does not work at
• 8 errors for n = 1024 (where McEliece used 50 errors)
• 20 errors for n = 8192 (a variant of classic McEliece).
Q(CaSC) / Johanna Sepúlveda 9
Summary
Code-Based Cryptosystems
Codes
McEliece Niederreiter
• m: Random, small weighted error vector
• Public key: H of Generalized Reed
Salomon Codes (GRSC)
• Encrypt: Encode m with Public key
• Decrypt: Decoding the Ciphertext
(Signatures, Fast)
Quasi-cyclic
Quasi-cyclic (QC)
Low Desity Parity Check
codes
(QC-LDPC) Q(CaSC) / Johanna Sepúlveda 10
Summary
Code-Based Cryptosystems
Classic McEliece
Goppa BIKE
Short Hamming (MDPC)
NTS-KEM HQC
Goppa Short Hamming
LEDACrypt
Short Hamming (LDPC)
Rollo
LowLow Rank
Rank (LRPC)
RQC
Low Rank
Q(CaSC) / Johanna Sepúlveda 11
Summary
Identify Operations
Polynomial Multiplication
Matrix Addition Schoolbook
Polynomial square/
square root
Discrete Sampling
Uniform Binomial Modular Reduction
Gaussian Trinary
Q(CaSC) / Johanna Sepúlveda 12
Summary
We are Ready to Start..
Post-Quantum
Cryptography
Main Principles
Families and Types
Standard
Main Components
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 13
Section 2
HASH
Q(CaSC) / Johanna Sepúlveda 14
Hash
Hash Function
Message Hash Hash value/
Digest
(Arbitrary size) Function (Fix size)
1 1 0 1 1 1 0
1 0 0 1 1 0 1 1 0 0
1 0 1 1 0 1 1 0 0
1 0 1 1 1 1 1 1 1
Q(CaSC) / Johanna Sepúlveda 15
Hash
Hash Function Properties
A function that maps a bit string of arbitrary length to a
fixed length bit string
Properties/Security bounds:
Collision resistance: It is computationally infeasible to find
any two distinct inputs that map to the same output
Preimage resistance/One way: It is infeasible to generate a
message that yields a given hash value
Second preimage resistance/weak collision resistance:
Given a message x, it is difficult to find a message z such
that hash(x) = hash(z)
Q(CaSC)
Does not rely on hard problems, / Johanna
instead Sepúlveda
they should meet the security bounds 16
Hash
Hash Function Properties
f Collision Resistance
D f
R
y
x
f
z
f -1
Preimage Resistance f -1
Second
Preimage Resistance
Q(CaSC) / Johanna Sepúlveda 17
Hash
Hash Function Properties
Easier to
Stronger break
Collision Resistance
Assumptions
Security
Second
Preimage Resistance
Preimage Resistance Pseudorandom
Weaker Harder
to break
Q(CaSC) / Johanna Sepúlveda 18
Section 3
HASH-BASED CONSTRUCTION
Q(CaSC) / Johanna Sepúlveda 19
Hash PQC
Traditional Signatures
Intractability Cryptographic Hash
Assumption Function
Factoring, discrete
logarithmic, SVP, ..
Digital Signature
Intractability assumption (hard problem): assume the existence of some
puzzle, whose hard instances are easy to generate but no efficient algorithm
can solve them
Q(CaSC) / Johanna Sepúlveda 20
Hash PQC
One-Time Signature (OTS)
Message
(plaintext)
Hashed
Function
Encryption Message
Secret
(key)
Signature scheme that uses a key pair for
signing a single message… Otherwise, an attacker is
able to reveal more parts of the private key and spoof
signatures
Q(CaSC) / Johanna Sepúlveda 21
Hash PQC
Leslie Lamport and Whitfield Diffie (1979)
If 0 publish x
If 1 publish y
Sign message 001
x y
h(x) h(y)
Q(CaSC) / Johanna Sepúlveda 22
Hash PQC
Winternitz OTS
Private key x
h(x)
h(h(x)) Published to sign 2
Public key
h(h(h(x)))
Many key pairs are needed to sign lots of messages, and therefore some kind of key
management is required. A solution to cope with this situation was proposed by
Ralph Merkle in 1979
Q(CaSC) / Johanna Sepúlveda 23
Hash PQC
Merkle Tree Root
Parent
Child Siblings
Tree structure in which leaf node is a hash of a block of data and
each non-leaf node is a hash of its children
Security is based entirely on assumptions about the security of
the underlying hash function
Q(CaSC) / Johanna Sepúlveda 24
Hash PQC
Merkle Tree Properties
Properties of a perfect Merkle binary tree
2n Number of leaves (L)
Each node has 0 or 2 children
All sibling leaves are on the same level
Number of leaves
Number of nodes Number of levels
Q(CaSC) / Johanna Sepúlveda 25
Hash PQC
Merkle Tree Properties
Level 1
Level 1
L=1 Level 2
N=1
V=1 L=2
N=3
V=2
Level 1
Level 2
L=?
Level 3
N=?
V=?
If you have a L= 16, whichQ(CaSC)
is the /value ofSepúlveda
Johanna N and of V? 26
Hash PQC
Merkle Signature Scheme (MSS)
In the MSS verification keys are hashed and build the leaves of
the tree
Neighbouring children are concatenated and then hashed to build
the parent node
The root of the tree represents the public key of the scheme
A signature is generated by applying the private key of an OTS
scheme
Q(CaSC) / Johanna Sepúlveda 27
Hash PQC
Merkle Signature Scheme (MSS)
Merkle Tree
Root node M03=
H(M01+M23)
M01= M23=
Hash function H(M0+M1)
Nodes H(M2+M3)
Leaf Nodes M0= H(m0) M1= H(m1) M2= H(m2) M3= H(m3)
Data m0 m1 m2 m3
Q(CaSC) / Johanna Sepúlveda 28
Hash PQC
Advantage of Merkle Tree
When we use a chain of hashes…
Root
Hash M0 M1 M2 M3 M4 M5 M6 M7
Data m0 m1 m2 m3 m4 m5 m6 m7
My data is not
tampered!
Alice Bob
Q(CaSC) / Johanna Sepúlveda 29
Hash PQC
Advantage of Merkle Tree
Alice wants to proof that m6 is not tampered (using a chain of hashes)…
Alice Bob
Sends m2, hashes and root
Hash m2
Appends other hashes (7)
Calculate root (calculated)
Compare root (calculated, received)
If roots match, the information was not tampered
Q(CaSC) / Johanna Sepúlveda 30
Hash PQC
Advantage of Merkle Tree
Alice wants to proof that m6 is not tampered (using a Merkle)…
M07
M03 M47
M01 M23 M45 M67
M0 M1 M2 M3 M4 M5 M6 M7
m0 m1 m2 m3 m4 m5 m6 m7
Q(CaSC) / Johanna Sepúlveda 31
Hash PQC
Advantage of Merkle Tree
•Verification key
•Path to the root
Alice wants to proof that m6 is not tampered (using a chain of hashes)…
Alice Bob
Key Generation:
Generate N private, public key pairs (Xi, Yi)
Calculate hash hi= H(Yi )
Build the tree with the hi
Private key: (Xi, Yi)
Public key: Root of the tree
Q(CaSC) / Johanna Sepúlveda 32
Hash PQC
Advantage of Merkle Tree
• Verification key
• Path to the root
Alice wants to proof that m6 is not tampered (using a chain of hashes)…
Alice Bob
Sends m2, parent hashes and root
Hash m2
Appends other hashes
Calculate root (calculated)
Compare root (calculated, received)
Merkle tree provides integrity and validity using a small
amount
Q(CaSC) ofSepúlveda
/ Johanna data 33
Hash PQC
Advantage of Merkle Tree
• Verification key
• Path to the root
Alice wants to proof that m6 is not tampered (using a chain of hashes)…
Alice Bob
Verification of the signature: Bob build the leave and
reach the root of the tree using the authentication path.
(root is already known to the verifier as the public key)
p air
y re
ke atu
w
Ne r sig
n Stateful hash-based Stateless hash-based
pe signature Q(CaSC) / Johanna Sepúlveda signature 34
Hash PQC
MSS Considerations
For each new signature generated the next key pair must be used
MSS is a stateful scheme: the index of the used key pair (respectively
leaf) of the MSS tree must be stored
Slow runtimes and large key and signature sizes
• It was not used since RSA were easier to implement
• Post-quantum nature turns hash-based crypto interesting again
Improvements in the implementation:
Signature generation is much faster thanks to the use of
pseudo-random number generators
Stateful hash-based
The construction of large trees is easier signature
Extended Merkle Signature Scheme (XMSS) PQC NIST candidate
Q(CaSC) / Johanna Sepúlveda 35
Hash PQC
XMSS
Winternitz OTS
Q(CaSC) / Johanna Sepúlveda 36
Hash PQC
XMSS Implementation
Simple implementation, but requires large memories
Q(CaSC) / Johanna Sepúlveda 37
Hash PQC
Sphincs+: Hyper-Tree Construction
Stateless hash-based
signature
Uses few-time signature
scheme
(smaller parameters)
Based on publicly
verifiable index selection
Steps: Few-Time Signature
Select FTS
Sign message with FTS
Build parent tree
Use tree to sign the pk
Q(CaSC) / Johanna Sepúlveda 38
Hash PQC
Sphincs+: FORS - Forest of Random Subsets
100 010 011
001 110 111
FORS signature with k = 6 and L = 3, for the
message 100 010 011 001 110 111
Q(CaSC) / Johanna Sepúlveda 39
Section 4
NIST: THIRD ROUND
Q(CaSC) / Johanna Sepúlveda 40
NIST
NIST Post-Quantum Candidates
Code Hash Multivariate Isogeny Lattice
First NIST
18 3 9 2 32
Round
Second NIST
7 2 4 1 12
Round
Third NIST
Round 1 - 1 - 5
(Finalists)
Third NIST
Round 2 1 2 1 2
(Alternate)
NIST Competition
Q(CaSC) / Johanna Sepúlveda 41
NIST
PQC Taxonomy
Stateless
hash-based
signature
Q(CaSC) / Johanna Sepúlveda 42
Section 5
SUMMARY AND TASKS
Q(CaSC) / Johanna Sepúlveda 43
Tasks
Summary: Hash-Based Cryptography
Hash-based cryptosystems can be used for quantum
safe signatures. Their problems has been widely
studied (1979)
Hash-based cryptography is based on different
security boundaries (CR, PIR, SPIR)
We only can sign! KEMs are not possible
Q(CaSC) / Johanna Sepúlveda 44
Tasks
Next Lecture
Third Block
Post-Quantum
Cryptography
Summary / Q&A (PQC)
Main Principles
Families and Types
Main Components
Standard
Secure implementation
Use cases
Q(CaSC) / Johanna Sepúlveda 45
Thank you.
See you soon!
Q(CaSC) / Johanna Sepúlveda
You might also like
- Slides Dan+tanja 20220825 Pqcrypto 16x9No ratings yetSlides Dan+tanja 20220825 Pqcrypto 16x957 pages
- Introduction To Classical and Quantum Computing 1e4pNo ratings yetIntroduction To Classical and Quantum Computing 1e4p400 pages
- Introduction To Quantum Computing For BusinessNo ratings yetIntroduction To Quantum Computing For Business178 pages
- Post-Quantum Security Opportunities and ChallengesNo ratings yetPost-Quantum Security Opportunities and Challenges26 pages
- Gokul Sai R:-1GA17EC041 Harshitha H R: - 1GA17EC049 Gowrav S: - 1GA17EC042 Amogh R: - 1GA17EC013 Bhoomika S S:-1GA17EC019100% (1)Gokul Sai R:-1GA17EC041 Harshitha H R: - 1GA17EC049 Gowrav S: - 1GA17EC042 Amogh R: - 1GA17EC013 Bhoomika S S:-1GA17EC01914 pages
- Foundations of Quantum Technologies 2025No ratings yetFoundations of Quantum Technologies 20251 page
- A Quantum Review - Cyber Security and Emerging TechnologiesNo ratings yetA Quantum Review - Cyber Security and Emerging Technologies4 pages
- Quantum Computing An Applied Approach by Jack D. Hidary Ebook All Chapters PDFNo ratings yetQuantum Computing An Applied Approach by Jack D. Hidary Ebook All Chapters PDF24 pages
- Cryptography and Network Security-Ppt-1 (Autosaved) .PPTMNo ratings yetCryptography and Network Security-Ppt-1 (Autosaved) .PPTM28 pages
- Quantum Machine Learning and Threats To Existing Cryptography100% (1)Quantum Machine Learning and Threats To Existing Cryptography20 pages
- Lectures Notes On Quantum Computing and Quantum InformationNo ratings yetLectures Notes On Quantum Computing and Quantum Information187 pages
- GDC EMEA Whitepaper Quantum-Computing ENNo ratings yetGDC EMEA Whitepaper Quantum-Computing EN17 pages
- Authentication and Key Agreement Based On Anonymous Identity For Peer-To-Peer Cloud0% (1)Authentication and Key Agreement Based On Anonymous Identity For Peer-To-Peer Cloud7 pages
- Switching Theory Architecture and Performance in Broadband ATM Networks100% (1)Switching Theory Architecture and Performance in Broadband ATM Networks424 pages
- 32 New Seminar Topics On Future Technology - 2019-2055No ratings yet32 New Seminar Topics On Future Technology - 2019-205551 pages
- Classical and Quantum Computing With C++ and Java Simulations100% (1)Classical and Quantum Computing With C++ and Java Simulations606 pages
- (IMF Working Papers) Quantum Computing and The Financial System - Spooky Action at A DistanceNo ratings yet(IMF Working Papers) Quantum Computing and The Financial System - Spooky Action at A Distance14 pages
- Amira Abbas: Quantum Computing Lecture at Wits University PDFNo ratings yetAmira Abbas: Quantum Computing Lecture at Wits University PDF102 pages
- Quantum Computing Cyber Security 1689695783No ratings yetQuantum Computing Cyber Security 168969578312 pages
- Quantum Computing Lecture Notes Another SetNo ratings yetQuantum Computing Lecture Notes Another Set105 pages
- Quantum Key Distribution: A Revolutionary Security TechnologyNo ratings yetQuantum Key Distribution: A Revolutionary Security Technology8 pages
- Interoperability in Internet of Things: Dr. Sudip MisraNo ratings yetInteroperability in Internet of Things: Dr. Sudip Misra25 pages
- Superconducting Quantum Circuits, Qubits and ComputingNo ratings yetSuperconducting Quantum Circuits, Qubits and Computing60 pages
- Ethics of Quantum Computing: An OutlineNo ratings yetEthics of Quantum Computing: An Outline22 pages
- Opportunities For Explainable Artificial Intelligence in Aerospace Predictive MaintenanceNo ratings yetOpportunities For Explainable Artificial Intelligence in Aerospace Predictive Maintenance12 pages
- Digital Electronic Circuits: Goutam SahaNo ratings yetDigital Electronic Circuits: Goutam Saha13 pages
- A Review of Quantum Cybersecurity Threats Risks and OpportunitiesNo ratings yetA Review of Quantum Cybersecurity Threats Risks and Opportunities8 pages
- Quantum Cryptography and Quantum Computation: Network Security Course Project ReportNo ratings yetQuantum Cryptography and Quantum Computation: Network Security Course Project Report16 pages
