Researchers at Chalmers University of Technology and Tianjin University have developed a theoretical method that could prepare certain quantum states roughly 1,000 times faster than a slower approach used for comparison. The result comes from calculations and simulations, with a hardware demonstration still ahead.
The peer-reviewed study was published in Physical Review Letters on August 3. Chalmers highlighted it in a September 10 announcement.
The proposed gain concerns operations for preparing and controlling quantum information. It does not mean that a complete quantum computer, or every program running on one, becomes 1,000 times faster.
Why preparation time matters
Before a quantum computer can perform a calculation, its information must be put into an appropriate starting state. Keeping that information intact throughout the calculation is another challenge.
As the US National Institute of Standards and Technology explains, quantum bits, or qubits, store information in quantum states that can combine the possibilities represented by 0 and 1. Carefully designed algorithms manipulate those states to tackle certain difficult problems.
Those states are fragile. Disturbances can corrupt the information, so useful quantum computing requires both accurate control and ways to deal with errors. A shorter operation leaves less time for the environment to interfere, provided the faster control remains accurate.
The work focuses on bosonic codes, which encode information in the states of an oscillator, such as a microwave field confined in a superconducting circuit. An oscillator supports many possible states, giving researchers room to arrange information in ways that help protect it against particular errors.
According to Chalmers’ explanation, creating and controlling these encoded states has been a bottleneck.
A shorter sequence of control signals
The method uses quantum lattice gates, a set of building blocks for manipulating quantum states. Periodic control signals drive the system through a repeating cycle.
The earlier approach changes the system gradually through thousands of cycles. The new protocol constructs the desired transformation within one period. The comparison concerns that specific gradual method, rather than every competing technique for controlling a quantum computer.
In the paper’s numerical benchmarks, optimized control produced selected elementary logical gates with errors around 0.01%, on a timescale of a few microseconds. A microsecond is one-millionth of a second.
A logical gate changes encoded quantum information. The simulated error measures assess how closely the operation matches its target; they are not a measured failure rate for a commercially available machine.
The next step is a laboratory test
Chalmers says the approach is suited to existing superconducting-circuit platforms. These use materials that conduct electricity without resistance at very low temperatures.
Co-author Tangyou Huang said the team was discussing possible experimental implementations with colleagues. The announcement did not report a completed hardware test or give a commercial deployment date.
A laboratory experiment would need to show that real control equipment can deliver the required signals accurately enough. Faster preparation would then need to work as part of a larger system that also measures, corrects and processes quantum information.
The simulations give researchers a proposal to test. They do not settle those engineering questions.
What businesses can take from the result
Our earlier article on corporate spending on quantum computing described companies investing in skills, applications and integration while commercial returns remain uncertain.
This research addresses part of the technical problem behind that uncertainty: preparing and manipulating information accurately enough for longer calculations. It supplies no evidence of a faster drug-discovery project, cheaper logistics plan or improved financial model.
For a company assessing quantum services, a useful test remains the performance of a complete task against the best available conventional method. Speeding up one operation could help future systems meet that test, but the benefit will depend on how much time and error that operation contributes to the whole calculation.
