Quantum Communication

Transcription

1 Quantum Communication Nicolas Gisin, Hugo Zbinden, Mikael Afzelius Group of Applied Physics Geneva University, Switzerland Nonlocal Secret Randomness Quantum Key Distribution Quantum Memories and Repeaters ECOC 11 1

2 Nonlocal Secret Randomness - Alice V + H Quantum Theory: If Alice and Bob make the same measurement (both H/V or both +/-), then the result is random and secret. - Bob V + H This is Nonlocal Secret Randomness! H or V? The result is random, but it is the same at Alice and at Bob + or -? The result is random, but it is the same at Alice and at Bob Quantum Magic, Entanglement 2

3 Nonlocal Secret Randomness Quantum Magic, Entanglement The results at Alice and Bob are two manifestations of the same event: one random event can manifests itself at two locations. Why should one believe this?!? What can one do with it? 3

4 Why should one believe in Nonlocal Secret Randomness? The deep answer: - Alice V + H By rotating the measurements on one side, one can violate Bell s inequality. This proves beyond any doubt that the results are not predetermined: each result is created on the spot and no one could know it in advance. Thus the result is random and secret - Bob V + H 4

5 Why should one believe in Nonlocal Secret Randomness? The easier answer: - Alice V + H If: 1. quantum theory 2. one measures only polarization 3. same measurement same result - Bob V + H then: the quantum state is totally uncorrelated from any third party, thus the result is random and secret 5

6 Let s exploit this gift of nature Secret randomness is nothing but a cryptographic key: QKD. This can be used in many ways: For encryption: guarantee that no third party can learn anything about a message while it is transmitted from Alice to Bob. For authentification: guarantee that what Bob receives is really coming from Alice. 6

7 QKD for encryption One-time pad or AES Plain text Secret key Cipher text Secret key Plain text Use the key to encode and decode the plain text. Photons Quantum key distribution (QKD) Slide taken from the Japanese UQCC: Updating Quantum Cryptography and Communication 7

8 QKD for authentification W.-C. Message authentication Scheme to ensure that data are genuine and have not been altered. Text Text Tag Tag Verification Use the key to choose a hash function to generate the tag. Photons Quantum key distribution (QKD) QKD can be used for any application that requires secret keys8

9 Quantum Random Number Generator CERTIFIED BY THE SWISS FEDERAL OFFICE OF METROLOGY Quantum Random Number Generators High bit rate 4 or 16 Mbits/s On line monitoring Main OS s supported 9

10 Certification of Encryption Solution 1/3 QKD Quantum Key Industry Specification Group First draft: end of 2010 Data Encryption 10 10

11 Certification of Encryption Solution 2/3 Public Key Agreement Key Data Encryption 11 11

12 Certification of Encryption Solution 3/3 Dual Key Agreement QKD Quantum Key Public Key Agreement Conventional Key EAL4+ Industry Specification Group First draft: end of 2010 Data Encryption Key Resulting Key 12 12

13 Complete Solution 13 13

14 Commercial QKD System Key Management Management Raw Key Postprocessing Optical Layer Reliability Key Rate Functionality Security 14

15 Spin-off, University of Geneva km Quantummultiplexed Cryptography Towards quantum below Lake Geneva channel for commercial users. Used daily by some commercial customers 15

16 QKD secures Elections in Geneva Downtown Geneva First Deployment: September 2007 October 2007, with election day on October 21 st Installation time: 30 minutes Continuous operation during more than 7 weeks Encryption of a Gigabit Ethernet link Geneva Government Data Center 4 km Central Vote Counting Station Ballots Mail Votes Subsequent deployments: all future elections are secured using the Cerberis solution Cerberis Solution 16

17 Siemens IT Services and Solutions B.V. xwdm Q Data Center (The Hague) Quantum Channel Dark Fiber Data Center (Zoetermeer) 17

18 Reliability: Swiss Quantum Network Run continuously during 21 months Monitored by the University of Applied Science 18

19 19

20 Quantum Hacking: very useful! 1. The principle of QKD will never be attacked, only the implementation. 2. The adversary can t copy the quantum data and process them later: he has to physically access the line during the key exchange. In contrast, in classical crypto both the principle and the implementation can be attacked. Implementation failures can be attacked remotely. If the principle of classical crypto gets broken, then - All electronic money looses all value - All past communications can be read 20

21 Long distance QKD: World records 150 km of installed fibers, Optics Express 17, (2009) Using Superconducting Single-photon detectors Lausanne 250 km in the lab. in collab. Corning, NJP 11, (2009) 21

22 Secret Key Rate 10MHz 1MHz 100kHz 10kHz 1kHz P2P + WDM Distance limitation +4 Years Today Lab Today Commercial 1Hz Distance [km] There is a hard wall around 400 km! With the best optical fibers, perfect noise-free detectors and ideal 10 GHz single-photon sources, it would take centuries to send 1 qubit over 1000 km! 22

23 Beating the hard wall: Teleportation of entanglement Q teleportation Entanglement Entanglement Entanglement over twice the distance Entanglement between photons that never interacted PRA 71, 05302,

24 Quantum memory Goal: controlled and reversible mapping of a photonic quantum state onto a long lived atomic ensemble. photon in crystal doped with millions of ions photon out at desired time in same Q state The quantum state of the photon is now coded in a huge entangled states of billions of «atoms» 24 Nature 456, 773, 2008

25 Controlling the Dephasing! Atomic Frequency Comb N j= 1 e ikr iδ t 1... e j... continuous Dephasing j e j g Im( P j ) g N P j = e iδ t j Absorption Re( P j ) Frequency periodic Rephasing Im( P j ) Absorption Re( P j ) Frequency δ = m k, = 0, ± 1, ± 2,... k m k Rephasing after T =1 25

26 Quantum memory - dream and reality Property Desired performance Efficiency State-of-the-art (quantum & classical state storage) Fidelity* Multi-mode storage capacity high Pulse duration ns ~100 ps Storage time > sec >2 sec Universal entanglementpreserving Complexity simple 64 (1000) modes atomic vapor & RE-crystal Different storage media and protocols (for qubits) * post selected Hedges et al, Nature (2010); Hosseini et al, Nature Phys. (2011), Usmani et al, Nature Comm (2010), Saglamyurek, WT et al, Nature (2011), Longdell et al, Phys Rev Lett. (2005), Jin et al, quant-ph (2010), Clausen et al, Nature (2011), Rempe et al., Nature (2011), Lvovsky, Sanders, WT, Nature Phot. (2010) 26

27 Demonstration of entanglement between a telecom photon and an excitation stored in a crystal Nd 3+ :YSO crystal Clausen, Usmani et al, Nature 469, ,

28 Photon-Crystal 2-qubit interference Clausen, Usmani et al, Nature 469, , 2011 V raw 80% Photon-Crystal entanglement with a violation of the CHSH-Bell inequality: S=2.64 > 2 28

29 QM A A QM S A (m,m) Φ AB B QM m Quantum Repeaters for Long Distance Fibre-Based Quantum Communication n QM C European Program Photonics S B (n,n ) Φ CD QM B D QM : 2 M The goal of QuReP is to develop a Quantum Repeater - the elementary building block required to overcome current distance limitations for long-distance quantum communication. Switzerland: - Université de Genève - ID Quantique SA Sweden: - Lunds Universitet France: - Laboratoire Aimé Cotton - Laboratoire de Chimie de la Matière Condensée de Paris - Université Pierre et Marie Curie Germany: - Universität Paderborn, DE Source 1 PPLN WDM Filters Integrated Photon pair sources QM Multi-Mode Q Memories Rare-Earth Ion Doped Solids Fidelity High Fidelity BSMs Source 2 Multiple Systems 29

30 Conclusions Nature offers perfect cryptographic keys QKD exists and has found niche markets. Quantum Hacking is very useful, but will never attack the principle of QKD. QKD offers FORWARD SECURITY: the key in future remains always as secure as at the time of creation. The AFC protocol is very promising for a solid-state multimode Q memory. 30

31 Organizer