Quantum signal processing in Hilbert space fragmented systems

This paper proposes a protocol that extends Quantum Signal Processing to nonintegrable systems by leveraging Hilbert space fragmentation in a pair-hopping model, enabling flexible nonequilibrium control in integrable sectors and parallel manipulation of multiple quantum dynamics within a single system.

The Gist

The Big Picture: Taming the Quantum Chaos

Imagine you are trying to conduct a massive orchestra. In a normal quantum system (like a chaotic jazz band), every instrument is listening to every other instrument. If you try to change the tempo of just the violins, the drums and trumpets immediately react, and the whole song turns into a messy, predictable noise called thermalization. In physics, this means the system "forgets" its starting state and settles into a boring, uniform heat.

For a long time, scientists could only conduct these quantum orchestras if the musicians were perfectly disciplined (a system called integrable). If the musicians were chaotic (non-integrable), you couldn't control them.

This paper introduces a brilliant new trick: Hilbert Space Fragmentation (HSF). Think of this as building invisible, soundproof walls inside the orchestra hall. Suddenly, the chaotic jazz band is split into separate rooms. In some rooms, the musicians are still chaotic. But in other rooms, they are perfectly disciplined.

The authors show that by using a technique called Quantum Signal Processing (QSP), they can conduct the music in the "disciplined rooms" with incredible precision, while the "chaotic rooms" just do their own thing. Even better, they can conduct multiple different songs in different rooms at the exact same time, all within the same building.

---

The Key Concepts (The Metaphors)

1. The Orchestra and the Walls (Hilbert Space Fragmentation)

Imagine a large dance floor (the Hilbert Space) where people are dancing.

• Normal Chaos: Everyone can bump into everyone else. Eventually, everyone just jiggles in place (thermal equilibrium).
• The Fragmentation: Now, imagine you drop giant, invisible glass walls on the floor. Some dancers get trapped in small, isolated bubbles.
  • The Frozen Bubbles: Some dancers are stuck in a corner and can't move at all.
  • The Chaotic Bubbles: Some groups are still bumping into each other wildly.
  • The Organized Bubbles: Some groups are dancing in a perfect, synchronized line.

The magic of this paper is that the "walls" are created by the rules of the dance itself (the physics of the system), not by building physical barriers.

2. The Conductor's Baton (Quantum Signal Processing - QSP)

QSP is like a super-advanced conductor's baton.

• In the past, this baton only worked if the musicians were already perfectly organized (integrable).
• The baton works by alternating between two moves: a "signal" move (letting the music play) and a "processing" move (adjusting the rhythm). By doing this in a specific pattern, the conductor can force the music to play exactly the song they want, turning a simple beat into a complex melody.

The authors figured out how to use this baton inside the "Organized Bubbles" of our fragmented dance floor.

3. The "Domain Wall" Trick (Parallel Control)

This is the coolest part. The authors realized they could create a "domain wall"—a specific arrangement of dancers that acts like a permanent glass wall.

• Left of the wall: A group of dancers is in an "Organized Bubble."
• Right of the wall: Another group is in a different "Organized Bubble."

Because of the glass wall, the conductor can tell the Left group to dance a Waltz and the Right group to dance a Tango simultaneously. They don't interfere with each other. This is "parallel control" in a single quantum system.

---

What Did They Actually Do?

1. The Setup: They used a specific model of particles (spinless fermions) that naturally creates these invisible walls. It's like a game of "pair-hopping" where particles can only move if they jump in pairs, which naturally traps them in certain patterns.
2. The Experiment: They applied their "QSP baton" (a specific sequence of pulses) to this system.
3. The Result:
   • In the Organized Rooms: The system did exactly what the conductor wanted. They could design complex, non-repeating dance moves (nonequilibrium dynamics) that stayed stable for a long time.
   • In the Chaotic Rooms: The system ignored the conductor's complex instructions and just melted into a uniform, hot mess (thermalization).
   • The Comparison: They showed that by simply choosing where to start the dance (the initial state), they could decide whether the system would be a masterpiece of control or a chaotic mess.

Why Does This Matter?

• Breaking the Rules: Previously, if you wanted to control a quantum system, it had to be perfectly ordered. This paper says, "No, you can have a messy system, as long as you know how to find the clean pockets inside it."
• Doing More with Less: Instead of needing two different quantum computers to run two different simulations, you can run both on one computer by using these invisible walls to separate the tasks.
• Future Tech: This is a huge step toward building programmable quantum simulators. Imagine a quantum computer that can simulate a superconductor in one corner and a magnetic material in another, all at the same time, without them interfering.

The Takeaway

The authors found a way to turn a chaotic quantum system into a set of isolated, controllable islands. By using Quantum Signal Processing, they can conduct complex "songs" on these islands simultaneously, proving that even in a chaotic world, you can find pockets of perfect order and control them with precision.

Technical Summary

1. Problem Statement

Quantum Signal Processing (QSP) is a powerful framework for designing quantum algorithms and controlling nonequilibrium dynamics, originally developed for nuclear magnetic resonance and later applied to integrable systems (e.g., the 1D Transverse-Field Ising Model). However, extending QSP to generic non-integrable many-body systems faces two fundamental obstacles:

1. Lack of Conserved Quantities: Non-integrable systems typically lack the extensive set of local conserved quantities required to map dynamics onto a solvable structure (like the Bogoliubov–de Gennes (BdG) form) used in standard QSP.
2. Thermalization: Generic non-integrable systems tend to thermalize under unitary dynamics, causing a loss of local information and making it difficult to sustain or control specific nonequilibrium states.

The central challenge is to determine if QSP can be realized in non-integrable systems and how to control dynamics in such chaotic environments without relying on global integrability.

2. Methodology

The authors propose a protocol that leverages Hilbert Space Fragmentation (HSF) to overcome these obstacles. HSF is a phenomenon where the Hilbert space decomposes into dynamically disconnected Krylov subspaces due to kinetic constraints and global symmetries.

• Model System: They utilize a 1D pair-hopping model of spinless fermions with open boundary conditions. The Hamiltonian involves a pair-hopping term ($J$) and a staggered on-site potential ($h$) with four-fold periodicity.
  • $H(t) = H_{PH}(t) + H_{stag}(t)$
• HSF Structure: The model exhibits HSF, splitting the Hilbert space into:
  • Integrable Sectors: Subspaces composed of specific pseudospin configurations (e.g., Néel states $|\uparrow\downarrow\uparrow\downarrow\dots\rangle$) where the dynamics map exactly to an integrable spin-1/2 XX model.
  • Non-integrable Sectors: Subspaces containing "fracton" defects (e.g., $|\dots \uparrow \downarrow \uparrow - + + - \downarrow \dots\rangle$) where the dynamics are chaotic and thermalize.
  • Frozen Regions: Configurations with consecutive identical fractons (e.g., $++$ or $--$) act as domain walls, dynamically isolating different regions of the chain.
• QSP Protocol: The authors implement a piecewise-constant driving protocol alternating between the pair-hopping Hamiltonian ($H_{PH}$) and the staggered potential ($H_{stag}$).
  • The sequence is designed such that in the integrable sector, the evolution maps to a QSP sequence acting on BdG subspaces.
  • The signal parameter $a_\lambda$ is encoded via the momentum-dependent energy spectrum of the integrable sector.
  • The control parameters are the phases (durations) of the alternating Hamiltonians.

3. Key Contributions

1. Extension of QSP to Non-Integrable Systems: The paper demonstrates that QSP can be successfully implemented within a single non-integrable system by restricting the dynamics to specific integrable Krylov subspaces defined by the initial state.
2. Parallel Control via Domain Walls: By introducing domain walls (frozen regions) via specific initial configurations, the system can be partitioned into multiple independent integrable sectors. This allows for the parallel emulation of distinct quantum dynamics within a single physical system.
3. Coexistence of Control and Thermalization: The study provides a unified framework where QSP-controlled non-thermal dynamics and thermalization occur simultaneously in different sectors of the same system, depending solely on the initial state configuration.
4. Analytical and Numerical Verification: The authors provide an analytical proof of the QSP mapping in the integrable sector and numerical simulations (using QuSpin) to verify the contrast between integrable and non-integrable behaviors.

4. Key Results

• Integrable Sector (QSP Success):
  • In sectors initialized with Néel-like states, the time evolution exactly reproduces the QSP unitary transformation $U_{PQ}(a_\lambda)$.
  • By tuning the phase sequence (e.g., using a BB1 pulse sequence), the authors achieve robust polynomial transformations of the signal.
  • Local observables (pseudospin expectation values) remain spatially inhomogeneous and do not decay to thermal values, confirming the preservation of nonequilibrium dynamics.
• Non-Integrable Sector (Thermalization):
  • When the same driving protocol is applied to initial states in non-integrable sectors (containing fracton defects), the system fails to map to the desired BdG structure.
  • Numerical results show that local observables relax to spatially homogeneous values consistent with the infinite-temperature ensemble average restricted to the specific Krylov subspace. This confirms Krylov-restricted thermalization.
• Parallel Emulation:
  • Simulations of a system with a domain wall show that the left and right integrable regions evolve independently according to their respective QSP protocols, while the domain wall remains frozen. This validates the ability to run multiple distinct quantum algorithms in parallel on one hardware platform.

5. Significance

• Beyond Integrability: This work breaks the reliance on global integrability for QSP, suggesting that "islands" of integrability within chaotic systems can be exploited for quantum control.
• Scalability and Efficiency: The ability to perform parallel emulation in a single system using domain walls offers a route toward scalable quantum emulation without requiring additional ancilla qubits or non-local operations.
• New Control Paradigm: It introduces a new perspective on nonequilibrium control where the initial state acts as a switch between controllable (non-thermal) and uncontrollable (thermal) regimes.
• Experimental Relevance: The proposed model (pair-hopping with staggered potentials) is feasible in cold-atom quantum simulators, particularly in the Stark many-body localization regime, making these results experimentally accessible.

In conclusion, the paper establishes that Hilbert Space Fragmentation is not merely a constraint on dynamics but a resource for organizing and implementing Quantum Signal Processing in complex, non-integrable many-body systems.