Decomposition for Imperfect Beamsplitters

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing decomposition algorithms for linear optics unitaries assume perfect implementation of transformations on interferometers, which is not feasible due to imperfections in beamsplitters, leading to degraded performance.

Innovation Solution

A method called 'device compensated decomposition' that accounts for imperfect splitting ratios on beamsplitters by decomposing the desired unitary into matrices that can be implemented using imperfect optical elements, and a calibration process to identify these imperfections, allowing for high-fidelity implementation of unitaries on interferometers with imperfect beamsplitters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If perfect decomposition algorithms are used assuming ideal beamsplitters, then the theoretical accuracy is high, but the actual implementation accuracy degrades due to imperfect beamsplitter ratios

Engineering Contradiction:
Improveunitary transformation accuracyVSAvoidimplementation fidelity
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent performs preliminary calibration of the interferometer to measure actual beamsplitter ratios before implementing the unitary transformation. This preliminary characterization allows the decomposition algorithm to account for real device imperfections rather than assuming ideal components, thereby maintaining high accuracy despite physical limitations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the decomposition approach by incorporating actual measured beamsplitter ratios as parameters into the unitary matrix construction. Instead of using fixed ideal values, the algorithm adapts to the actual device parameters through measured reflectivity and transmissivity values, enabling accurate implementation on imperfect hardware

Inventive Principle:
Principle #35Parameter changes

2Productivity

If decomposition algorithms assume perfect beamsplitters, then the algorithm complexity is low, but the performance degrades when beamsplitters are imperfect

Engineering Contradiction:
Improvecomputational efficiencyVSAvoidbeam splitter ratio precision
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent implements feedback by measuring the actual beamsplitter ratios and using these measurements to adjust the decomposition algorithm. The measured values feed back into the unitary matrix construction, allowing the system to compensate for manufacturing imperfections and achieve high precision despite physical deviations from ideal behavior

Inventive Principle:
Principle #23Feedback

3Measurement precision

If ideal decomposition is used, then the theoretical unitary transformation is achieved perfectly, but actual device imperfections cause degradation

Engineering Contradiction:
Improvetransformation accuracyVSAvoidbeamsplitter imperfections
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent converts the harmful effect of beamsplitter imperfections into a beneficial approach by measuring and characterizing these imperfections. The actual deviant behavior of the beamsplitters is used as input data to construct a corrected decomposition algorithm, transforming the problem of imperfection into an opportunity for more accurate, device-specific optimization

Inventive Principle:
Principle #22Blessing in disguise (Convert harm into benefit)

Data Source

PatentUS11747132B2Methods and apparatus for decomposition to account for imperfect beamsplitters
Publication Date: 2023.09.05 XANADU QUANTUM TECHNOLOGIES HOLDINGS ULC
  • US11747132B2 patent drawing
  • US11747132B2 patent drawing
  • US11747132B2 patent drawing

AI summary

A method includes receiving a representation of an N-mode interferometer and a representation of at least one imperfection associated with the N-mode interferometer at a processor, N being a positive integer value. The processor identifies multiple two-mode interferometers and multiple phases based on the representation of the N-mode interferometer and the representation of the at least one imperfection. The multiple two-mode interferometers and the multiple phases are configured to apply a unitary transformation to an input signal. The method also includes sending a signal to cause at least one of storage or display of a representation of the multiple two-mode interferometers and a representation of the multiple phases.