10th May 2000 - Cambridge, UK
Potential applications include secure optical communications
The privacy of information sent on optical communication networks could be greatly improved, if researchers at Toshiba Research Europe Ltd. (TREL), on the Cambridge Science Park, have their way.
They have developed a new type of very sensitive photo-detector, which is capable of detecting the faintest possible optical signals. The device, which will be reported at the prestigious Conference on Lasers and Electro-Optics in San Francisco on 10 May 2000, responds to individual photons, the smallest indivisible units, or quanta, of light.
At the heart of the new device lies a layer of quantum dots, tiny disks of a semiconductor, each measuring just tens of nanometers in diameter and a few nanometers in height. Restricted to such short length scales, the electrons inside the dots display ‘quantum’ properties not seen in larger pieces of semiconductor. The TREL paper is the first report of quantum dots being used to detect individual visible, or near infrared photons.
A photon incident upon the device has the effect of liberating an electron trapped within one of the dots. Until now it has seemed very difficult to detect the tiny charge associated with this single electron. However, the TREL researchers have shown that this can be achieved by integrating the quantum dots inside a transistor structure, which was prepared in collaboration with the University of Cambridge.
According to Dr. Andrew Shields, the project leader: “Advances in semiconductor nano-technology are creating devices with exciting new functionalities. The fact that a transistor containing quantum dots can detect individual photons is significant, because such a structure is likely to have several important advantages over conventional detectors.”
The TREL team plans to apply this new photon detector to the emerging field of quantum communication whereby data sent along a conventional optical fibre is encoded at the single photon level. By using single photons two parties can achieve the seemingly impossible feat of being able to detect if their communication has been intercepted or altered en route by a hacker or eavesdropper. The security of the technique is guaranteed by fundamental laws of quantum mechanics, which require that a quantum state (i.e. a single photon) cannot be measured without altering its properties in a detectable way.
One important application of quantum communication is to allow two users of an open network to form a secret cryptographic key (whose confidentiality is guaranteed) that can subsequently be used to encrypt data sent between them. The technique could find a use in a wide range of applications where sensitive information is sent on an open network, for instance for e-commerce, banking and financial services, government and healthcare.
According to Professor Michael Pepper, the managing director of TREL's Cambridge Research Laboratory: “It is thought that computer security is a question of software. This shows that fundamental science may in fact offer the solution”.
Technical background: Single Photon Detection
A photon is the smallest indivisible unit, or quantum, of light. One photon corresponds to an extraordinarily ‘dim’ optical signal, as can be gauged from the fact that an ordinary 100W light bulb will emit approximately 10,000,000,000,000,000,000 (10 billion billion) photons every second.
A signal travelling along an optical fibre is typically made up of several million photons. Such intensities are required to generate enough signal at the other end of the fibre when using ordinary optical detectors. In quantum cryptography, the intensity of the light source is turned down so that the information is encoded upon just a single photon. By using a very sensitive optical detector that can measure single photons it is possible to determine if the signal has been intercepted or altered en route.
The quantum dot single photon detector, proposed and demonstrated by TREL, is based upon a transistor structure in which the conducting channel is closely spaced from a layer of quantum dots. TREL has shown that if the separation of the quantum dots and the channel is just several nanometers, the resistance of the FET is sensitive to a change in the occupancy of a single quantum dot by just a single electron. This attribute allows the device to act as a detector of single photons, since absorption of a photon creates carriers in the semiconductor, which after capture by a dot, produce a detectable change in the resistance of the channel of the FET.
TREL expects the quantum dot detector to have several advantages over conventional types of single photon detector based on avalanche processes, such as the photomultiplier tube or the avalanche photodiode. Photomultiplier tubes are vacuum tube devices that, while excellent for many applications, are fragile, bulky, expensive and have a relatively low efficiency. On the other hand, avalanche photodiodes are very prone to dark count noise, especially when operated at the high frequencies used in optical communications. This issue is especially important for quantum communication, since the noise levels of avalanche photodiodes are recognised as limiting the transmission distance and bit rate. Several other features of conventional detectors, including the requirement for high bias voltages and cryogenic cooling and extreme sensitivity to temperature changes and excess bias also make them inconvenient to use.
The quantum dot single photon detector is a semiconductor detector, which works on an entirely different principle to conventional devices and can, therefore, overcome many of their problems. In particular, by avoiding the avalanche process and its associated problems, it should be less prone to noise. Although it is yet to be verified experimentally, it is anticipated to have a fast time response, since it is based upon a transistor, the building block of today's high-speed electronic circuitry. Another advantage is that the quantum dot detector works at normal operating voltages (<5V) and is more robust.
TREL regards quantum cryptography as the most promising application for the new device. However, single photon detectors have many other applications in science and technology. In medical imaging, for instance, single photons are detected in PET and CT scanners and, more recently, for laser optical imaging. Lifetime fluorescence measurements using single photon counting is also used in the diagnosis of some medical conditions. It is also widely used in analytical chemistry for determining the chemical recipe of samples. Another application is in laser ranging, tracking and imaging and for industrial scanning and process control. Finally, single photon detection is also widely used in scientific research in the fields of particle physics, astrophysics and materials science.
Technical background: Quantum Cryptography
Cryptography is best known as a way of keeping the contents of a message secret. We usually think of the main practitioners of cryptography as being spies and diplomats who wish to communicate secretly. In the future, however, many people predict the widespread use of cryptography as one of the enabling technologies of e-commerce and the network society in general. Confidentiality of network communications, for example, is of great importance for e-commerce and other network applications. However, the applications of cryptography in e-commerce go far beyond simple confidentiality. In particular, cryptography allows the network business and customer to verify the authenticity and integrity of their transactions. If the trend to a global electronic marketplace continues, better cryptographic techniques will have to be developed to protect business transactions.
Sensitive information sent over an open network may be scrambled into a form that cannot be understood by a hacker or eavesdropper. This is done using a mathematical formula, known as an encryption algorithm, which transforms the bits of the message into an unintelligible form. The intended recipient has a decryption algorithm for extracting the original message. There are many examples of information on open networks, which need to be protected in this way, for instance, bank account details, credit card transactions, or confidential health or tax records.
In order to allow different users to use the same algorithm, the algorithm is used in conjunction with a secret key, a long sequence of binary numbers, which is known only by the legitimate users. Only users sharing the same key will be able to decrypt each other's encrypted messages. Since the key allows access to the encrypted information, it is of paramount importance that it is kept secret.
Before two parties can send information securely, they must first exchange a secret key. This however presents a dilemma, sometimes called the Catch 22 of Cryptography — how can the two parties exchange a key secretly before they can communicate in secret? Even if the sender and receiver found a channel that they believed to be secure, they still could not be sure, with total certainty, that the key was transmitted securely.
Quantum cryptography is a technique for two parties to form a key on an open optical network. Such keys can subsequently be used for the encryption of data sent on the network between the two parties, or alternatively, be used for an ‘uncrackable’ type of encryption called the Vernam cipher. An attraction of quantum cryptography is that its security is guaranteed by fundamental laws of nature (Quantum Mechanics). It allows the detection of unauthorised eavesdropping, as well as providing a guarantee of security when there is no eavesdropper present. This is not possible using any other form of key distribution, which relies either upon the difficulty of factorising large numbers, or the assumed privacy of the network.
In optical quantum cryptography the bits used to form the key are carried by single photons travelling either, along an optical fibre, or in an optical free space link. Information can be encoded on the photons in a variety of ways, such as by their polarisation or phase. Because the information is carried by a single photon, it is not possible for a hacker to tap in and remove part of the signal. Since single photons don't split, if the hacker measures the photons on the fibre, they will not be received at the other end, alerting the intended recipient to the presence of the hacker. Furthermore, the technique is also secure from a slightly more sophisticated type of eavesdropping where the hacker first measures the photons and then retransmits them. This is because the laws of quantum mechanics tells us quantum bits (or qubits) of information, such as encoded single photons, have the peculiar characteristic that they are disturbed by measurement. This fact allows the legitimate receiver of the message to test whether it has been intercepted or altered by a hacker on the channel.
Crucial for the realisation of a practical quantum communication system is the development of devices for the generation, manipulation and detection of single photons. The performance of real quantum key distribution systems is limited by the dark count noise in the single photon detector. As its name implies, dark counts are a detector signal measured in the absence of an incident photon. They give ‘wrong’ results, producing errors in the key shared by the users. The noise is problematical because there is no way of knowing if an error in the key is due to dark count noise or due to an eavesdropper. If the error rate is just a few percent, the users can still form a secure key by applying error correction, followed by a technique known as privacy amplification. However, if the error rate rises to tens of percent, it becomes impossible to distinguish the dark count from the presence of an eavesdropper and quantum key distribution fails.
Unfortunately conventional detectors are rather noisy, especially when operated at the high frequencies usually used in optical communications. In practice, the detector noise has the effect of limiting the transmission distance and the bit rate. However, the single photon detector device developed at TREL promises to have less noise and operate at high frequencies. It will, therefore, significantly improve the performance of quantum key distribution systems.
Terms and Conditions
