Back to Home

Reliability Analysis of Dynamic Surface Code Deformation

Design, Analysis, and Validation of Dynamic Surface Code Deformation for Quantum Error Correction under Correlated Faults

Implementation & Materials Read Report

Project Summary & System Overview

This project investigates the reliability of quantum information encoded using the surface code, focusing on the impact of spatially correlated faults—such as those induced by ionizing radiation—on logical qubit preservation. The study compares two architectural strategies: static redundancy (maintaining a high code distance) and dynamic code deformation (temporarily reducing code distance in response to detected bursts).

Using a rotated surface code lattice, the analysis quantifies how burst faults affect logical error rates and overall system performability, providing design guidance for future quantum hardware.

Surface Code Architecture

The lattice model uses:

  • Data Qubits — carry logical information.
  • Syndrome Qubits — stabilizer measurements.
  • Code Distance (d) — controls error suppression capability.
Rotated surface code lattice diagram
Rotated surface code lattice (data + syndrome qubits)

Validation & Methodology

The methodology employs fault injection by simulating spatially localized bursts in the code lattice, stressing the decoder and evaluating both static and dynamic policies. Performance is measured using a performability metric that combines logical error rate and computational utility, reflecting real-world trade-offs.

Phase 1 — Baseline Validation

Establish logical error rates under independent depolarizing noise across code distances.

Phase 2 — Correlated Fault Injection

Inject a spatially localized high-intensity burst with parameter p_burst in the chosen quadrant.

Phase 3 — Policy Evaluation

Compare Static Redundancy and Dynamic Deformation policies using the reward metric R = (1 - P_L) × V_config.

Implementation & Materials

All simulation scripts, Jupyter notebooks, and supporting materials are available in the shared project folder. The primary repository, including code, data, and presentation slides, can be accessed via the button below:

Implementation & Materials (Zoho WorkDrive)

Implementation snapshot (pseudocode):

Implementation snapshot (pseudocode): 

    // Pseudocode: Correlated Fault Injection Loop
    for each QEC_round in rounds:
      apply_background_noise(p_phys)
      if burst_condition_met:
        inject_correlated_noise(Q_burst, p_burst)
      extract_syndromes()
      result = decode_with_MWPM()
      record_logical_outcome(result)
    end

The Stim circuits, PyMatching decoder pipeline, and plotting utilities are included in the shared notebook and repository.

Project Report — View & Download

The full project report Design_Analysis_Final_Paper.pdf is available for inline viewing and for download. The document contains the complete problem statement, methodology, simulation setup (Stim + PyMatching), fault-injection model, performability analysis, SAN-based extension, results and reproducibility instructions.

Download PDF

Your browser does not support inline PDFs. You can open the PDF in a new tab or download it.

About this Report

This report presents a design analysis and validation study that evaluates dynamic surface code deformation ("hole punching") as an active reliability strategy against spatially correlated faults such as ionizing radiation–induced phonon events. Using Stim for stabilizer circuit simulation and PyMatching for decoding, the study injects localized high-intensity bursts and compares two control policies: maintaining a nominal distance-5 code (static redundancy) and dynamically reconfiguring to a distance-3 code (degraded operation).

Key Modeling Assumptions

  • Spatially correlated burst region Qburst is modeled as the top-left quadrant (≈50% of the lattice).
  • Physical noise: background depolarizing channel 𝒟(pphys); burst adds pburst inside Qburst.
  • Performability evaluated as R(p) = (1 - PL(p)) × Vconfig with Vd=5 = 1.0 and Vd=3 = 0.6.
  • Current simulations assume instantaneous, error-free switching; latency-aware analysis is proposed via SANs.

Findings (summary)

While dynamic code deformation physically preserves the logical qubit during localized bursts, the performability analysis in the simulated parameter regime indicates that the utility penalty of downgrading (40% in this study) outweighs the survival benefit. The report frames this as a quantitative design boundary and proposes SAN-based latency-aware extensions to determine when dynamic reconfiguration becomes beneficial.