A further boost for quantum cryptography
Research developed at the Universidad Politécnica de Madrid's Facultad de Informática proposes efficient information reconciliation for quantum key distribution in order to further the development of quantum cryptography. Quantum cryptography enables secure information transmission governed by the laws of physics in contrast to conventional methods which are usually based on the comptuational difficulty of problems. This is achieved by integrating quantum physics and cryptography. One of the key aspects for protecting information transmission is secret key distribution. Quantum key distribution (QKD) is capable of creating secure keys between two parties sharing a quantum channel (using optic fibre, for example), as quantum mechanics provides different ways of computing or transmitting information than conventional security systems. By means of a complex protocol, emitter and receiver exchange a series of qubits (smallest unit of quantum information) encoded as photons. In this way, they can agree on a secret key, as, according to the principles of quantum physics, the receiver will detect any attempt at observing a qubit, which assures a fully secure information exchange. New key distillation methods Advances in modern cryptography on secret key agreement are driving the development of new key distillation methods and techniques. Quantum key distribution is a direct beneficiary of most of these developments, focused on information reconciliation (error correction) and privacy amplification (compression). In this context, highly interactive protocols, that is, protocols requiring a lot of communication, have been used for information reconciliation in the past. Research developed at the Facultad de Informática shows how modern encoding techniques can improve the performance of these information reconciliation methods in QKD and proposes the use of low-density parity-check (LDPC) codes. LDPC codes have good efficiency and error correction rates. But LDPC code efficiency comes at a price, i.e. efficiency is only good for very long codes within a limited error range in the quantum channel. This obliges the use of several codes where the error rate varies significantly for different channel uses, a common situation in QKD. To tackle these problems, the research analyses several techniques for adapting LDPC codes in order to reduce the number of codes required to cover the target error range. These techniques are also used to improve the mean efficiency of short LDPC codes in a encoding scheme with feedback (return message). Short codes are interesting because they can be used for high-performance hardware implementations. The research also introduces a new protocol that bypasses initial error rate estimation required in other proposals. This protocol also has interesting implications for finite key analysis. The research was developed as a PhD thesis by Jesús Martínez Mateo, assistant professor of the Department Applied Matematics, and was supervised by Vicente Martín Ayuso, principal investigator of the Quantum Information and Computation Research Group.
Countries
Spain