Squeezed-vacuum Bosonic Codes Achieve Robustness Against Noise Using Evenly Spaced Photon-number Support

Bosonic codes represent a promising avenue for building robust quantum computers, and researchers are continually seeking ways to improve their performance against the inevitable errors that plague quantum systems. Nir Gutman, Eliya Blumenthal, and Shay Hacohen-Gourgy, alongside Ariel Orda and Ido Kaminer, have now introduced a new family of these codes, constructed from carefully arranged squeezed vacuum states. This innovative approach yields codes that demonstrate significant resilience against both photon loss and dephasing, two major sources of error in quantum information processing. The team’s designs utilise relatively simple preparation methods and, crucially, offer a compelling balance between error correction capabilities and practical implementation within existing quantum computing architectures, establishing squeezed-vacuum codes as a viable and hardware-ready option for future quantum technologies.

Squeezed States Enable Bosonic Code Analysis

Scientists have achieved a breakthrough in quantum error correction by developing a novel approach to constructing bosonic codes from squeezed vacuum states, offering a promising pathway to protect quantum information. This research provides a detailed theoretical foundation for these codes, demonstrating their connection to established concepts and providing analytical tools for performance evaluation. The team rigorously derived the mathematical form of the logical basis states, revealing how squeezed vacuum states are combined and filtered to create robust codewords. The analysis demonstrates that these codes exhibit a unique structure, supporting discrete quantum states separated by a specific number of photons, crucial for creating distinct error subspaces and enabling effective error correction.

Researchers employed the Knill-Laflamme method to assess the codes’ resilience against common sources of noise, including photon loss and dephasing, and benchmarked their performance against established cat codes. The results indicate that increasing the number of squeezed vacuum states within a code improves its tolerance to signal loss, though this comes at the cost of increased sensitivity to dephasing. This work establishes a clear connection between the newly developed squeezed-vacuum codes and established concepts like rotation-symmetric codes and ideal number-phase codes, demonstrating that the squeezed-vacuum codes approach the ideal limit as the squeezing parameter increases. The team highlights that their approach requires fewer fundamental building blocks than traditional rotation codes, potentially simplifying implementation and reducing resource requirements.

Squeezed Vacuum Codes Resist Quantum Noise

Researchers have developed a new family of bosonic quantum codes, constructed from superpositions of squeezed vacuum states, that offer protection against both photon loss and dephasing noise, common challenges in quantum computing. These “squeezed-vacuum codes” achieve robustness by arranging the squeezed states at evenly spaced angles and photon numbers, enabling a straightforward preparation process using sequences of conditional rotations. Performance evaluations demonstrate that increasing the number of squeezed states enhances tolerance to signal loss, although this comes at the cost of increased sensitivity to dephasing. This work establishes squeezed-vacuum codes as a viable addition to the existing family of bosonic codes, complementing established approaches like cat codes and binomial codes. The team analysed potential implementations on current experimental platforms, including circuit QED and trapped-ion systems, where the necessary high-fidelity operations are either available or under development. The researchers acknowledge a trade-off between loss and dephasing tolerance, noting that the optimal code design depends on the specific noise characteristics of the quantum computing platform.

Squeezed-Vacuum Codes Protect Against Photon Loss

Scientists have introduced a new family of bosonic codes, constructed from superpositions of squeezed vacuum states, offering protection against both photon loss and dephasing noise. These “squeezed-vacuum codes” achieve robustness by arranging states at evenly spaced angles in phase-space and simultaneously supporting evenly spaced photon numbers. The team developed preparation circuits, including a two-legged code utilizing a Hadamard-conditional-squeezing-Hadamard sequence on an ancilla qubit, and generalized sequences of conditional rotations for creating “m-legged” codewords. Experiments demonstrate that as the number of squeezed-vacuum states within a code increases, loss tolerance improves, though at the cost of increased sensitivity to dephasing.

Performance was evaluated against loss and dephasing noises using the Knill-Laflamme violation function, and benchmarked against cat codes. The two-legged code achieves a code distance of 2, with an angular separation of π/2 between primitive squeezed states, offering protection comparable to a four-legged cat code. Researchers found that a controlled-squeezing operation, conditioning the squeezing of a bosonic mode on a qubit’s state, can generate the two-legged state in a single step. The team outlines implementations in both circuit QED and trapped-ion platforms, where high-fidelity Gaussian operations and conditional controls are either currently available or under active development. Measurements confirm that the codes exhibit improved loss tolerance as the number of squeezed-vacuum states increases, though this comes at the expense of heightened dephasing sensitivity. This work establishes squeezed-vacuum codes as practical, hardware-ready members of the bosonic codes class, offering a promising pathway for robust quantum information processing.

Squeezed Vacuum Codes Resist Loss and Dephasing

Scientists engineered a novel approach to quantum error correction by constructing bosonic codes from superpositioned squeezed vacuum states, offering resilience against both photon loss and dephasing noise. These “squeezed-vacuum codes” achieve robustness by arranging states at evenly spaced angles in phase-space and simultaneously supporting evenly spaced photon numbers. The team developed preparation circuits, beginning with a two-legged code utilizing a Hadamard-conditional-squeezing-Hadamard sequence on an ancilla qubit, and extending this to general “m-legged” codewords through sequences of conditional rotations. To evaluate performance, researchers employed the Knill-Laflamme violation function, benchmarking the codes against established cat codes to assess their tolerance to loss and dephasing.

Experiments demonstrate that as the number of squeezed-vacuum states increases within a code, loss tolerance improves, though at the cost of increased sensitivity to dephasing. The team outlines implementations for both circuit QED and trapped-ion platforms, leveraging the availability of high-fidelity Gaussian operations and conditional controls within these systems. Researchers pioneered a probabilistic method for generating two-legged codewords, employing a Hadamard-conditional-squeezing-Hadamard sequence on an ancilla qubit coupled to a bosonic mode initially in the vacuum state. This method relies on a final measurement, collapsing the mode state into a superposition of squeezed states, and the probability of obtaining either logical code state depends on the squeezing strength.