Quantum Technologies Impact on Cybersecurity (And AI)
Quantum Technologies Impact on Cybersecurity (and AI)
Cisco - Polytechnique Symposium April 9-10 2018
Romain Alléaume
Télécom ParisTech & Paris Center for Quantum Computing
Binary unit of quantum information: Qubit
Super conducting Josephson Junction
Trapped Ions / Atoms
1 0 ↵ or = ↵ 0 + 1 0 1 | i | i ✓ ◆ ✓ ◆ ✓ ◆ 2 Photonic (↵, ) C states 2 ↵ 2 + 2 =1 | | | |
2
2 Quantum Computer: Working with n qubits
2-qubit state:
Binary relabelling
n-qubit quantum computer = 1 …
n State of dimension 2n Orthogonal Basis:
n Quantum parameter space 2 exponentially larger than classical 0, 1 n C { }
3 Probing the exponential: supercomputer simulations
Sunway TaihuLight JUQUEEN 1.3 PetaBytes 0.5 PetaBytes 93 Petaflops /15 MW 6 Petaflops / 3 MW
46 46 World Record: 46 qubits (2 x 2 complex amplitudes) December 2017: Jülich (Germany) and Wuhan (China)
2 Adding one qubit => factor 2 increase => 50 qubits without classical reach
4 Probing the exponential: supercomputer simulations
30 - 40 qubits simulator
Sunway TaihuLight JUQUEEN46 46 World Record:1.3 46 PetaBytes qubits simulationLaunched => (2 xby 2 ATOS) real in july parameters 2017 0.5 PetaBytes 93 Petaflops /15 MW Juqueen ( Jülich Germany) & Sunway TaihuLightEntry6 Petaflopsprice ( Wuhan about / 3China) 100 MW k €
46 46 World Record: 46 qubits (2 x 2 complex amplitudes) December 2017: Jülich (Germany) and Wuhan (China)
2 Adding one qubit => factor 2 increase => 50 qubits without classical reach
5 Race for « Quantum superiority / supremacy » has begun
Challenge : Solve a computational problem that cannot be solved classically, on a quantum machine
Remarks: Quantum Supremacy typically target on computational problems that are not useful (sampling) Requires about 50 qubits (with low noise) (Boixo et al, 2016)
6 Main challenge for quantum computing (QC) Noise scaling with qubit number
Google Quantum AI Lab’s intended 2018 progress for quantum computers 2015
Two regimes:
• Near-term (5-10 years): Applications of QC without error correction • Long term (> 15 years ? ): Large QC (fault-tolerant approach needed)
7 Near-term applications of quantum computing
1) Simulating (large) quantum systems
« Nature isn’t classical dammit . . . and if you want to make a simulation of Nature, you’d better make it quantum mechanical » Richard P. Feynman, Simulating Physics with Computers, 1981
Exponentially difficult problem (with system dimension) on classical machines
Quantum chemistry Condensed-matter physics: High-energy physics Drug Design High Tc superconductivity Nuclear fusion
2) Solving optimization problems Applica ons Minimizing complex (nonlinear) functions by “simultaneously • Logis cs sampling” entire space through quantum superposition • Opera onal Research Example: quadratic optimization • VLSI design N N • Finance E(x1,...,xn)= hi xi + Jij xixj K Is it really possible with
i=1 i 3) Searching in a database: Grover Algorithm 1997 Find marked element in an unstructured database of N elements O(√N) vs O(N) 4) Machine Learning / AI J Exponential speed-up for many core ML algorithms - Support vector machine HHL Algorithm (Harrow, Hassimi, Loyd 2009) - Principal component analysis - Topological features of Data K How to efficiently load big data on quantum computer ? 9 Applications of large quantum computer (2) Cryptanalysis 4) Shor algorithm (1994): Polynomial-time factoring algorithm on a quantum computer Shor Algorithms essentially breaks all existing Public-key Cryptography RSA, Elliptic Curve, Discrete Log (Diffie-Hellman) 5) Need for Post-Quantum Cryptography Crypto primitives based on NIST Call for Quantum Resistant Algorithms: problems not solved by Shor November 30 2017 Code-based crypto Lattice-based crypo è New Public-key Crypto for « quantum world » Hash-based crypto è French Project RISQ (BPI) Multivariate –quadratic equations 10 Leveraging multimode photonic technologies to build near-term quantum computers Recent results in our group Universal single unitary synthesis with 4 modes > 2-time increase spectral efficiency Perpectives J Universal components for Technological route to 20-40 qubits optical frequency mode logic processors (103 to 106 modes) J Optimized by Telecom Industry Reference: J. M. Lukens, P. Lougovski, Optica 4, 8-16 (2017) 11 Quantum key distribution: QKD Eve Alice (eavesdropper) Bob Quantum channel Classical channel Secret Key Ks Ø Ø Secret Key Ks Ø Information-theoretic security Ø Secure even against computationally unbounded adversary Ø Everlasting security => Enables long-term data encryption security 12 Telecom Paristech: active player of QKD development in Europe First European QKD Network (Vienna, 2008) CV-QKD Technology: Record distance : 100 km (2012) Collaborative projects with key actors First Commercial System (SeQureNet) (2008-2018) Network and Cryptography (FREQUENCY) Implementation Security, Q hacking (Q-CERT, ETSI) Optical multiplexing (Quantum WDM) Quantum Communications (QCALL) Quantum Technology Flagship 13 Develop QKD at the classical / quantum frontier Integration with optical networks Rupesh Kumar, Hao Qin, Romain Alléaume, Coexistence of continuous variable QKD with intense DWDM classical channels. New Journal of Physics, 17(4), 043027. (2015). Implementation security Hao Qin, Rupesh Kumar, and Romain Alléaume Quantum hacking: Saturation attack on practical continuous-variable quantum key distribution, Phys. Rev. A 94, 012325. (2016) Smaller and Cheaper systems Courtesy - Photonic integration Eleni Diamanti Adrien Marie and Romain Alléaume - Convergence Self-coherent phase reference sharing for continuous-variable quantum key distribution Phys. Rev. A 95, 012316, (2017) 14 Global quantum networks and classical-quantum convergence Satellite Q Com demonstrated in 2016 & 2000 km Beijing-Shanghai QKD network Challenges / Limitations Ø Fundamental dist limit (~250 km) Ø Dedicated HW => high cost Ø Weak cryptographic integration Hybrid Quantum Cryptography Combine Quantum Cryptography with Computational Cryptography to boost the performance of Qcrypto Patent(s): Communications with everlasting security from short- Ø Extend cryptographic functionality // everlasting security term-secure encrypted communication, RA, 2016 Ø Extend max distance (~500 km within reach) Ø Cost reduction by upgraded compatibility with standard telecom hardware 15 Thanks for your attention ! / Questions ?