In theory, quantum key distribution (QKD) offers information-theoretic secure communications. In practice, however, it does not because most security proofs of QKD use assumptions that are not satisfied by the implementations, and thus their security is seriously threatened by quantum hacking. This includes, for instance, the assumption that the legitimate users of the system have a precise mathematical model to describe the behaviour of their apparatuses, that the systems are located in perfectly shielded labs which prevent any unwanted information to be leaked to the eavesdropper, and that all quantum devices and classical post-processing units are honest and follow the prescriptions of the protocol. Here, we review recent results that significantly relax these unrealistic and hardly feasible requirements, and pave the way towards achieving practical QKD with implementation security and enhanced performance.
|