⚛️ quantum physics

The perfect entangler spectrum as a tool to analyze crosstalk

This paper introduces the "perfect entangler spectrum," a geometric tool that identifies and characterizes dynamic crosstalk in quantum computers by analyzing peaks in the spectrum during frequency scans of spectator qubits, thereby aiding the scaling of quantum systems.

Original authors: Matthias G. Krauss, Christiane P. Koch

Published 2026-07-09

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Matthias G. Krauss, Christiane P. Koch

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to teach two friends, let's call them Alice and Bob, how to dance a perfect, complicated tango together. In the world of quantum computers, this dance is called a "two-qubit gate," and it's the fundamental move needed to build a super-smart machine. But there's a problem: a third friend, Charlie, is standing right next to them. Even though Charlie isn't supposed to dance, he keeps accidentally stepping on Alice's toes or pulling Bob's arm. This unwanted interference is called crosstalk, and it's the biggest headache stopping quantum computers from getting bigger and better.

For a long time, scientists have been good at spotting the "static" crosstalk—the kind where Charlie is just standing there, constantly bumping into Alice and Bob because they are too close. But the real troublemaker is dynamic crosstalk. This happens only when Alice and Bob start dancing (running a gate). It's like Charlie only starts tripping them when the music gets loud and fast. Because this interference is so sneaky and changes depending on the rhythm, it's incredibly hard to catch.

Enter Matthias G. Krauss and Christiane P. Koch, who have invented a new tool called the Perfect Entangler Spectrum. Think of this as a high-tech "crosstalk radar" or a special kind of musical spectroscopy. Instead of just looking at the dancers, they scan the frequency of Charlie's voice (his energy level) to see if it causes the dance to go wrong.

The Magic Radar: How It Works

The authors suggest a clever way to measure this. They imagine a system with three superconducting qubits (the quantum dancers): two main ones (Alice and Bob) and one "spectator" (Charlie). They run a specific dance routine (a gate) on Alice and Bob while slowly changing Charlie's "voice pitch" (frequency).

As they scan Charlie's frequency, they look for peaks in their radar reading.

Low readings mean the dance is perfect. Alice and Bob are entangled beautifully, and Charlie is just watching quietly.
High peaks mean disaster. The dance has gone wrong because Charlie interfered.

The brilliance of this tool is that it doesn't just say "something is wrong." It tells you why. The authors found that these peaks appear for two very different reasons:

The "Static" Clashes: Sometimes, Charlie's natural pitch just happens to match Alice or Bob's pitch perfectly. This is like two tuning forks vibrating at the same note; they start ringing together even without anyone hitting them. The paper shows these are easy to predict just by looking at the hardware's blueprint.
The "Drive-Induced" Chaos: This is the tricky part. Sometimes, the music (the control pulses) itself creates a new, hidden rhythm. Even if Charlie's pitch doesn't match Alice or Bob's, the way the music is played can create a "beat" that makes Charlie jump in. The authors found that these dynamic peaks happen at specific multiples of the driving frequency (like $1\times$ , $2\times$ , $3\times$ , up to $6\times$ the base rhythm).

What the Radar Found (and What It Didn't)

In their simulations, the authors tested this radar on two popular dance moves: the Controlled-Z (CZ) gate and the $\sqrt{iSWAP}$ gate.

The CZ Gate: When they scanned the spectator's frequency, they saw a forest of peaks. Some were the obvious static clashes, but many were the sneaky, drive-induced ones. For example, they found that even if the main dance move relies on a specific rhythm, the "music" can accidentally trigger a swap of energy between the spectator and the coupler (the device connecting the dancers) through complex, multi-step processes.
The $\sqrt{iSWAP}$ Gate: This dance move turned out to be a bit more robust. The radar showed fewer peaks, suggesting this gate is less sensitive to the spectator's frequency. However, they still found a weird split in one of the peaks that the simple radar couldn't explain immediately. They suspect this is because the music itself changes the "weight" of the dancers (an AC-Stark shift) as the dance goes on, a time-dependent effect that static calculations miss.

Crucially, the paper rules out the idea that all crosstalk is just a simple frequency match. They explicitly show that you cannot just look at the hardware's natural frequencies to find all the problems. You have to look at how the control pulses interact with the system. If you only check for static resonances, you will miss the drive-induced peaks that cause the most trouble.

Can We Actually Use This?

The authors don't just stop at a computer simulation; they sketch out a plan to do this in a real lab. They suggest that you don't need to measure the entire three-qubit system at once (which is hard). Instead, you can run the dance twice: once with Charlie in a "sleeping" state and once with Charlie in an "awake" state. By comparing the results of these two two-qubit dance routines, you can reconstruct the "Perfect Entangler Spectrum."

They admit that doing this at a fixed time (stopping the dance at exactly 690 nanoseconds) might add a little bit of "noise" to the radar picture, making the peaks look a bit jagged. However, their simulations show that all the main features are still there. So, while it might not be a perfect, crystal-clear image, it's definitely good enough to spot the troublemakers.

The Bottom Line

This paper doesn't claim to have "solved" crosstalk forever. Instead, it offers a powerful new diagnostic tool. It suggests that by scanning the spectator qubit's frequency and looking for these specific peaks, engineers can identify exactly which frequencies cause the most trouble and why.

The authors propose that this spectrum can become a standard test for future quantum chips. If you are building a quantum computer with a tunable coupler (a device that connects qubits), you can use this radar to pick the best frequencies for your qubits, avoiding the "danger zones" where the dance goes wrong. It turns the mystery of "why is my gate failing?" into a map of "here is exactly where the interference is coming from," helping scientists design better, less noisy quantum computers.

Technical Summary: The Perfect Entangler Spectrum as a Tool to Analyze Crosstalk

Problem Statement
Crosstalk remains a primary obstacle to scaling quantum computers, particularly in frequency-addressed architectures like superconducting transmons and trapped ions. While persistent crosstalk (static ZZ-interactions) has been extensively analyzed and mitigated, dynamic crosstalk occurring during gate operations is more difficult to detect and characterize. This dynamic crosstalk arises from unwanted interactions with spectator qubits or signal leakage, potentially degrading the entangling power of two-qubit gates. Existing analysis methods are often tailored to specific gate types (e.g., Cross-Resonance gates) or fail to distinguish between deviations that destroy entanglement and those that merely introduce correctable single-qubit rotations. There is a need for a general metric to identify when spectator qubits cause a two-qubit gate to fail in being a "perfect entangler."

Methodology
The authors introduce the Perfect Entangler (PE) Spectrum, a diagnostic tool based on the geometric classification of two-qubit gates using local invariants.

Theoretical Framework:
- The method extends the measure of distance to the set of perfect entanglers (gates capable of generating maximally entangled states from separable ones) to systems involving a third "spectator" qubit.
- A functional $J$ $J$ is defined to quantify the quality of the two-qubit entangling operation in the presence of a spectator. This functional combines:
  - The distance of the effective two-qubit gate to the set of perfect entanglers within the subspace where the spectator is in state $|0\rangle$ and $|1\rangle$ respectively ( $J_{PE}$ ).
  - A similarity measure $S$ ensuring the gates in both subspaces are identical (preventing unwanted entanglement with the spectator).
  - A term $\Delta_U$ accounting for non-unitarity if the logical subspace is embedded in a larger Hilbert space.
- The PE Spectrum is generated by varying the frequency of the spectator qubit ( $\omega_3$ ) and plotting the minimum value of the functional $J$ over the gate duration. Peaks in this spectrum indicate frequencies where crosstalk prevents the gate from being a perfect entangler.
System Model:
- The authors apply this framework to a system of three fixed-frequency transmons coupled via a central tunable coupler.
- Two parametrically driven gates are analyzed: the Controlled-Z (CZ) gate and the $\sqrt{iSWAP}$ gate.
- The analysis distinguishes between static resonances (degeneracies in the bare Hamiltonian) and drive-induced resonances (transitions mediated by the flux drive, including multi-photon processes).
Resonance Analysis:
- To identify the physical mechanisms behind spectral peaks, the authors define a resonance measure $M^{(n)}_{\omega_r}$ . This metric calculates the proximity of transition frequencies (up to $n$ -th order perturbation theory) to integer multiples of the drive frequency components ( $\omega_\phi$ and $2\omega_\phi$ ).

Key Results

Spectral Signatures: The PE spectrum exhibits distinct peaks at specific spectator frequencies, signaling dynamic crosstalk.
- Static Resonances: Peaks align with degeneracies between spectator levels and the logical qubits (e.g., $\omega_3 = \omega_1$ ) or higher-order transitions (e.g., involving $|2\rangle$ states).
- Drive-Induced Resonances: The spectrum reveals complex crosstalk pathways not obvious from the static Hamiltonian. These include multi-photon processes (e.g., $3\omega_\phi$ , $4\omega_\phi$ ) and transitions involving the coupler and spectator.
Mechanism Identification:
- For the CZ gate, the protocol is highly sensitive to the precise phase accumulation required for the population swap between $|11\rangle$ and $|20\rangle$ . Small deviations in transition frequencies caused by the spectator (dressing effects) lead to significant crosstalk peaks.
- For the $\sqrt{iSWAP}$ gate, the protocol is more robust to frequency shifts as it relies primarily on population swapping rather than precise phase accumulation. Consequently, the PE spectrum shows fewer and less pronounced peaks compared to the CZ gate.
- The analysis successfully attributes specific spectral features to specific mechanisms, such as excitation swaps between the spectator and the tunable coupler, or time-dependent AC-Stark shifts.
Experimental Feasibility:
- The authors propose an experimental protocol requiring only two two-qubit gate tomographies (one for the spectator in $|0\rangle$ and one for $|1\rangle$ ) plus a measurement of the relative phase on the spectator qubit.
- Simulations indicate that measuring the functional at a fixed gate time (rather than minimizing over time) is sufficient to detect the relevant crosstalk mechanisms, despite introducing some noise.

Significance and Claims
The paper claims that the Perfect Entangler Spectrum provides a versatile and practical tool for:

Quantifying Crosstalk: It precisely identifies frequencies where spectator qubits degrade the entangling power of a gate, distinguishing this from mere single-qubit phase errors.
Mechanism Diagnosis: By correlating spectral peaks with resonance measures, it allows researchers to identify whether crosstalk arises from static degeneracies or complex, drive-induced multi-photon processes.
Hardware Design: The spectrum can guide the selection of qubit frequencies and coupler parameters to avoid crosstalk-prone regimes, facilitating the design of scalable tunable coupler architectures.
Experimental Implementation: The method is presented as experimentally feasible, requiring significantly fewer resources than full three-qubit process tomography.

The authors emphasize that while their study focuses on three qubits and specific gate types, the approach is generalizable to larger systems and can be adapted to analyze crosstalk for universal gate sets by optimizing for average fidelity instead of the perfect entangler functional.

The Magic Radar: How It Works

What the Radar Found (and What It Didn't)

Can We Actually Use This?

The Bottom Line

Technical Summary: The Perfect Entangler Spectrum as a Tool to Analyze Crosstalk

More like this