In Ian Fleming's first novel, James Bond inspects his hotel room for signs of intrusion after returning from the Royale-les-Eaux casino. He confirms that the hair on his desk has not been moved. He checks that the cupboard handle has no fingerprints on it. He confirms that the water level in the lavatory water cistern hasn't changed. He settles down to think about his bigger goal.
Bond wouldn't find it easy to be sure of his privacy. His secrets would be kept on computers. He would rely on complicated devices that use quantum physics when he needed to share his secrets. These devices look like Jenga puzzles. How could Bond be sure that they were safe to use?
Quantum devices are difficult to measure. You don't want to depend on the details.
Three decades ago, there were hints that the vulnerabilities might not be a problem. Theorists proved that all the users had to do was to have the devices play a game. Winning with a high enough score would prove that no one else could hear.
The process of device-independent quantum key distribution has been demonstrated by two different experiments. Many of the necessary requirements were demonstrated in a third experiment. Three groups of researchers had to engineer complete systems out of quantum components.
Roger Colbeck of the University of York said he was excited and happy to see them.
Their experiments lift the perfect secrecy inherent in the quantum realm into the everyday world. The technology is not practical. Even the most paranoid now know that it is possible to communicate with perfect privacy.
When Claude Shannon proved that perfect secrecy was possible, the new work was born.
Suppose you want to protect the information with a sequence of bits. You can add another sequence of bits called a key. The combined sequence is meaningless to anyone who doesn't know the key.
The question was how to securely distribute the key to the people who need it.
How do you ensure the safety and speed of the person you are sending the key to? Cryptographers created nonrandom keys that could be distributed quickly and securely. The new kinds of keys could be hacked with enough computing power. cryptographers had to learn how to play a nonlocal game.
Physicists invented these games to see if the laws of quantum mechanics were real. Artur Ekert, a physicist at the University of Oxford, realized that they could be the source of secret.
Two people are isolated from each other in a game that Ekert studied. Each player is asked two yes-or-no questions at random. In order to win the round, their answers must correspond in certain ways. They have to give the same answer if they get the first question. They have to give the same answer if they get different questions. They have to give different answers if they get the second question.
How should Alice and Bob play the game? The chance that they will both be asked the same question is 25%. The best strategy they could agree on in advance would be to always give the same answer, no matter what.
Physicists found that Alice and Bob could do better using quantum physics. They would need a set of four quantum objects to separate. We can think of these objects as coins if we interact with them.
Alice and Bob have to put the coins into a special relationship. When Alice flips her first coin, Bob's first coin is more likely to land on the other side. They are more likely to land the same when they flip different coins. When Alice and Bob flip their coins, they are more likely to land on different sides.
The win conditions for the game correspond to this. Alice and Bob flip their first or second coin in response to the first or second question, saying yes for heads and no for tails. Alice and Bob could win 85% if they followed the commands of the quantum coins.
Colbeck said it follows from the laws of quantum mechanics. There is no good reason why that number is that number.
Ekert showed that the game provided a basis for distributing keys. For the case where Alice and Bob used coins manufactured by their enemies, theorists would spend 30 more years turning the game into a protocol, or detailed process.
Alice and Bob have to make a public announcement to turn their game into a process for sharing keys.
After a round is over, they broadcast some of their questions and answers. They can check their answers against their questions and figure out their winning percentage. It establishes a set of extraordinary facts about the process that make perfect security possible if they find a win rate of 85%.
The coin flips are random. How do they know?
An eavesdropper named Eve tries to rig the coins to make them nonrandom, for example, altering Alice and Bob's first coins so that they always land heads. It doesn't seem like it would increase the win rate. If you analyze the probabilities, you will find that Eve can't rig the coins so that Alice and Bob win more than 75% of the time. Victory over 75% means that nothing could have been rigged.
Colbeck said something random is happening.
Alice and Bob can create a key with their random flips. Alice and Bob mark 1 for heads and 0 for tails for each coin flip that they did not publicly disclose. This sequence is a secret to them. They have a good idea of the secret sequence the other person marks down because their coins are correlated. They don't know it perfectly because their correlations don't reach all the way to 100%, but they can find it out by making a slight modification to the game.
Bob and Alice have given each other a random key. There is a problematic possibility. Is it a secret? Eve would have entangled two of her own coins with Alice and Bob. Wouldn't it be possible for her to create her own version of the same key?
Ekert realized that victory at 85% ruled out the possibility of Eve being a spy. It must be spread thinner if it is spread between more than two parties. If Eve's coins were entangled with Alice and Bob, the correlation levels would go down and the winning rate would go down. Eve can't possibly be listening in if they measure an 85% winning rate.
The things that Eve could be listening in on are very small if we get something close to the maximum.
In principle, winning the game at 85% is all Alice and Bob need to do to distribute keys with perfect secrecy. Things are not easy for cryptographers. They know that winning the game with a score of 85% is impossible in the real world. The coins are quantum objects and must be manipulated with complex machines. They are vulnerable to many errors and hacking strategies.
If Alice and Bob won more than 75% of the time, perfect security could still be guaranteed. They succeeded only in cases where the devices were sufficiently trustworthy, manufactured by a friend with assurances that they worked in certain ways. By 2007, researchers were able to show that security was possible for devices with fewer restrictions.
This research proved that security could be achieved with winning percentages that seemed low enough that experimentalists might be able to reach them, even if Eve made the devices herself. As long as the devices allowed Alice and Bob to win at a certain rate, it was clear that they had obtained secret information.
The three groups of researchers decided two years ago that a demonstration of device-independent key distribution was possible. They had to do it.
Different ways of playing the game have already been developed. The Oxford experiment used ionized atoms instead of coins. They flipped them by shooting the atoms with lasers and measuring whether they glowed or stayed dark, just like heads or tails. They were able to reduce the complexity of their setup by using one single coin, instead of two, which they could flip in two different ways. The researchers in the Munich experiment used the same setup as the researchers in the Shanghai experiment, but they used coins instead of atoms. flipping an entangled photon and detecting it were the same thing for them.
All Rights Reserved © 2024
