Getting to grips with quantum computing
Updated: May 18, 2019
One of the MIT’s best-known physicists, Seth Lloyd, has used a musical analogy to explain quantum computers. “A classical computation is like a solo voice — one line of pure tones succeeding each other,” he writes. “A quantum computation is like a symphony — many lines of tones interfering with one another.” Quantum computers are certainly music to the ears of scientists who predict that they will eventually be able to solve incredibly complex computational problems much faster than any technology we have today.
Quantum computing is a relatively new technology. The American physicists Paul Benioff and Richard Feynman first put forward the concept of quantum computing in the early 1980s. British scientist David Deutsch only described the first universal quantum computer in 1985. A lot has been achieved in the subsequent years, but there is still a long way to go before we have quantum computers under our desks.
“The reality of quantum computing is probably 10 to 15 years away, yet it merits our attention now,” says Dr. Seungyun Lee of the joint committee on information technology (JTC1) set up by the global standards bodies IEC and ISO.
“The excitement in the industry for this new paradigm of computer hardware is understandable, given the promise of far greater computational power with whole new multidimensional capabilities.”
Two flavours of quantum computers
The computers we have today store data using bits, which have two states — either on or off — represented as a 1 or a 0. Quantum computing replaces these binary bits with qubits that have more states which are changing continuously. Qubits can be on, off or somewhere in between all at the same time. This state is called superposition and enables qubit-based computers to carry out far more calculations much faster. When qubits become entangled they share all the possible combinations of the quantum states of the individual qubits, substantially boosting computational power in the process. Quantum computers come in two flavours.
Gate-based quantum computing more or less works in the same way as traditional computing. A transistor performs a Boolean function: a sort of binary logic, commonly seen in advanced search engines, that works with modifiers such as ‘AND’ or ‘NOT’. The transistor receives two incoming signals and depending on what it encounters, sends out a new electric signal. In the quantum model, qubits replace the transistors.
The main challenge is increasing the small number of qubits possible today to industrial scale, which is difficult because it is a struggle to keep qubits in their quantum state. Qubits only function “coherently” when they are cooled down to mere thousandths of a degree above absolute zero, which also protects them from the destabilizing effects of radiation, light, sound, vibrations and magnetic fields. All of this limits the size and complexity of problems that gate-based quantum computers are currently able to tackle.
Computers based on quantum annealing take a radically different approach. Quantum annealers run adiabatic quantum computing algorithms. Instead of allowing the entanglement of all qubits, they create an environment where only restricted, local connections are possible. When they attain superposition, they can be used to mediate and control longer range coherences. This makes them suitable for a much narrower range of tasks, such as solving optimization problems — i.e. choosing the best solution from all feasible solutions.
Quantum annealing already works
Quantum annealers have already been used to solve such problems in the domains of finance and the aerospace industry, among others, with potential users limited only by the upwards of 10 million dollars cost of a quantum annealer device. As with gate-based quantum computing, decoherence is a major challenge for quantum annealers and they too require massive refrigeration units. The limited number of tasks that quantum annealers can perform means, for example, that they are unable to run Shor’s “decryption” algorithm. There will be more about this last point later.
It would be wrong to think of gate-based quantum computers and quantum annealers as competing technologies. They are useful for solving different problems. Quantum annealers are sometimes dismissed as not being proper computers, but unlike gate-based quantum devices, they are already delivering significant results. Both technologies face an array of similar challenges, including the need for a radically new software stack. As discussed, they share a common decoherence problem, as qubits cannot inherently reject noise. Because qubits are susceptible to perturbations, errors are hard to eliminate. And quantum algorithms are very difficult to design.
Quantum computing looks set to bring massive benefits, such as accelerating medical research, making advances in artificial intelligence and perhaps even finding answers to climate change. These can be broken down into three areas. Firstly, scientists have combined quantum computing with machine learning for the processing of images and the calculation of probabilities. Secondly, marrying quantum computers with Shor’s algorithm is having a major impact on the world of cyber security and traditional encryption methods. Lastly, quantum simulators facilitate the study of quantum systems — such as quantum chemistry or quantum field theory — that are difficult to study in the laboratory. They will have a massive impact on the critical area of cyber security.
Protecting critical infrastructure
Mobile phone calls, messaging and online banking all rely on complex mathematical algorithms to scramble information in order to protect it from malicious hackers, spies and cyber criminals. It is no exaggeration to say that there would be no confidentiality or security online without encryption and that many of the operations we take for granted today would no longer be feasible. Faced with increasing cyber attacks against critical infrastructure — including but not limited to power utilities, transport networks, factories and the health care industry — encryption is evolving to meet the threat.
The most prevalent system nowadays is public key encryption. It works by giving users two keys: a public key, shared with everyone, as well as a private key. The keys are large numbers that form part of an intricate mathematical algorithm that scrambles a user’s messages. The sender encrypts a message by using the receiver’s public key in order that only the intended recipient can unlock it with her or his private key. Even though the public key is freely available, the numbers involved are sufficiently large to make it very difficult to reverse the encryption process with only the public key.
The power of quantum cryptography
As computers become more powerful, however, and in the face of rogue states with the technology resources to pose a more serious threat, cryptographers are turning away from mathematics and looking to physics — specifically the laws of quantum mechanics — to achieve greater security. Wikipedia defines quantum cryptography as “the science of exploiting quantum mechanical properties to perform cryptographic tasks.”
That is because quantum cryptography is based on the behaviour of quantum particles, which are smaller units than molecules. For example, an encryption system called quantum key distribution (QKD) encodes messages using the properties of light particles. The only way for hackers to unlock the key is to measure the particles, but the very act of measuring changes the behaviour of the particles, causing errors that trigger security alerts. In this way, the system makes it impossible for hackers to hide the fact that they have seen the data.
The threat is so great that scientists are urging organizations to start looking at and adopting quantum encryption systems. Universal gate-based quantum computers may not be available for another decade, but quantum cryptography has already been available for a few years.