Recommended Free Tools
Choose a tomography method by starting with the result your experiment must deliver. If you need a complete density-matrix estimate, use an informationally complete measurement design and a suitable estimator. If the state is credibly low rank, compressed sensing may reduce the number of measurement settings. If you need only selected observables, fidelities, or other properties, classical shadows may be a better fit. In all three cases, check measurement conditioning, detector calibration, finite-shot noise, and drift before treating the estimate as reliable.
First decide what you need to learn about the state
The central choice is between reconstructing the state itself and estimating a defined set of properties. Those are different deliverables, and a method that is efficient for one may not answer the other.
As an Amazon Associate I earn from qualifying purchases.
Choose a full state estimate when downstream work needs the density matrix
Informationally complete tomography aims to estimate the density matrix without relying on a restrictive model of the state. For a Hilbert space of dimension d, unrestricted informational completeness spans an operator space with d2 degrees of freedom. The American Physical Society’s 2025 review, Practical Introduction to Benchmarking and Characterization of Quantum Computers, describes the need for d2 independent measurement effects. A full estimate can support analyses beyond a short, preselected list of observables, but it imposes a correspondingly broad measurement and inference task.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Choose property estimation when the questions are already known
If the experiment only needs a defined collection of observables, fidelities, or other state properties, reconstructing every element of a density matrix may be unnecessary. Classical shadows are designed for estimating selected properties from measurement data; they do not automatically provide a complete density matrix at the same cost. This distinction is especially relevant for n qubits, where d = 2n: unrestricted full reconstruction scales exponentially with qubit count.
#1 Best Overall
Compare the methods against your experiment
| Method | Best fit | Key condition or limitation |
|---|---|---|
| Informationally complete tomography, often with a physical estimator such as maximum likelihood | A full density-matrix estimate is the required output, and the apparatus can implement a sufficiently informative measurement set. | Unrestricted reconstruction has d2 operator degrees of freedom. Finite data and measurement-set conditioning affect uncertainty. Enforcing physical-state constraints does not make an incomplete measurement set informative without additional assumptions. (American Physical Society review, 2025; Joint Quantum-State and Measurement Tomography with Incomplete Measurements) |
| Compressed sensing or other low-rank reconstruction | The state is plausibly low rank or approximately pure, and the available measurement design is compatible with the recovery method. | The reduced scaling relies on low-rank structure and recovery assumptions; assess sensitivity to rank mismatch and noise. (Gross et al., “Quantum state tomography via compressed sensing,” arXiv preprint first posted 2009) |
| Classical shadows | The deliverable is a selected set of observables, fidelities, or other properties rather than a complete state. | Performance depends on the measurements and the target properties. Results from one platform or property family do not establish the same advantage for another. (Struchalin et al., “Experimental Estimation of Quantum State Properties from Classical Shadows,” PRX Quantum, 2021) |
| Joint state-and-measurement estimation | The detector effects are too uncertain for ordinary state-tomography assumptions to be credible. | Requires suitable trusted state preparations or control operations and a joint inference model; it does not remove the need to characterize the experiment. (Joint Quantum-State and Measurement Tomography with Incomplete Measurements) |
Use compressed sensing only when its structure assumptions are defensible
Compressed sensing exploits a state that is low rank or close to low rank, rather than treating every possible density matrix as equally plausible. Gross, Liu, Flammia, Becker, and Eisert report reconstruction using O(rd log2 d) measurement settings for dimension d and rank r, compared with d2 settings for standard methods. This is a result under the paper’s assumptions, not a guaranteed count for a particular laboratory setup.
Before choosing this route, ask whether independent physical or theoretical evidence supports the rank model, whether your apparatus implements a measurement design that meets the recovery conditions, and how the reconstruction behaves if the true state is noisier or higher rank than expected. A low-rank assumption can reduce the problem, but a wrong assumption can make a confident-looking reconstruction misleading.
Check measurement quality, not just completeness
A measurement set may be informationally complete and still be poorly conditioned. In that case, distinct state changes can produce data that are difficult to distinguish, so estimates may be sensitive to finite-shot fluctuations. Completeness answers whether reconstruction is possible in principle; conditioning and statistical efficiency help determine how useful it is with the data and hardware you actually have. The APS review discusses both conditioning and experimental error sources such as shot noise and laboratory drift.
- Measurement settings: Confirm which effects the apparatus can implement and whether together they span the operators needed for the chosen reconstruction.
- Conditioning: Determine whether the measurement design distinguishes the state variations relevant to your intended conclusions.
- Finite-shot uncertainty: Plan for uncertainty from the finite number of detected outcomes; a setting count is not a shot count or runtime estimate.
- Systematic effects: Consider drift and other laboratory errors that can bias data even when statistical fluctuations are small.
- Uncertainty reporting: Decide what uncertainty information the downstream conclusion requires, and whether the estimator and data support it.
Include detector calibration in the method choice
Ordinary state tomography assumes that uncertainty in the measurement operators is negligible. As the article Joint Quantum-State and Measurement Tomography with Incomplete Measurements states, “The QST formalism assumes that uncertainty about the measurement operators of the POVMs is negligible.” If that assumption does not fit your apparatus, treating the detector model as exact can transfer calibration error into the state estimate.
Joint state-and-measurement tomography is one possible response: infer the state and measurement effects together using a suitable model. It requires appropriate trusted preparations or control operations, so it changes the inference problem rather than eliminating calibration needs. Do not confuse it with detector tomography, whose target is the measurement apparatus rather than the prepared state.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Make the decision in a practical sequence
- Write down the required output. Specify whether you need a complete density matrix or a finite set of named properties. If the latter fully answers the experiment’s question, assess a property-estimation method before committing to full reconstruction.
- Establish whether low rank is credible. Use compressed sensing only if low-rank or approximate-purity structure is justified and you can evaluate the consequences of rank mismatch.
- Map the apparatus to candidate measurements. List implementable settings and their measurement effects. Check informational completeness for full tomography and method-specific recovery requirements for compressed sensing.
- Assess conditioning and calibration. Evaluate whether the measurement design can distinguish the features you need, and whether detector uncertainty is negligible enough for ordinary state tomography.
- Plan for noise and uncertainty. Account for finite shots and plausible systematic drift, then determine what uncertainty analysis is needed for the conclusions you intend to draw.
- Compare the candidate methods on the same deliverable. A reduced setting count is useful only if the method’s assumptions, measurements, and uncertainty are adequate for the claim you will make.
What the published comparisons do—and do not—establish
Struchalin and coauthors experimentally demonstrated classical-shadow property estimation using high-dimensional photon spatial states. In that experiment, they report an advantage over conventional reconstruction for fidelity estimation under limited measurements. This is evidence for that platform and task, not a universal guarantee for other hardware or property families.
Rank #4
Likewise, the compressed-sensing scaling is expressed in measurement settings under a specific low-rank recovery framework. It should not be read as a universal shot budget, elapsed-time estimate, or promise of experimental savings. The appropriate method remains dependent on the state structure, target properties, measurement design, and calibration quality of the experiment.
Free tools Windows power users keep installed
One-click scans. No signup required.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
