A reproducible, hardware-aware benchmarking framework to study the impact of realistic quantum noise on Variational Quantum Algorithms (VQAs) across NISQ-era quantum hardware.
Variational Quantum Algorithms (VQAs) are leading candidates for achieving near-term quantum advantage on Noisy Intermediate-Scale Quantum (NISQ) devices. However, their performance is highly sensitive to hardware-specific noise, gate fidelities, and circuit depth.
This repository presents a systematic and reproducible benchmarking framework that evaluates the degradation of entanglement, energy estimation, and circuit depth under realistic noise models inspired by IBM superconducting qubit systems and IonQ trapped-ion architectures.
All experiments are implemented using Qiskit and Aer simulators, enabling hardware-aware yet fully reproducible analysis suitable for academic research and IEEE-style publications.
- Benchmark entanglement degradation under realistic noise
- Compare IBM-like and IonQ-like hardware behavior
- Analyze VQA energy estimation robustness under noise
- Study depth-normalized entanglement decay
- Demonstrate hardware-aware error mitigation techniques
- Provide a fully reproducible research pipeline
- Hardware-aware noise modeling for superconducting (CZ-based) and trapped-ion systems
- Bell-state entanglement benchmarking under increasing noise
- Entanglement metrics: Concurrence, Fidelity, Negativity
- Energy vs noise analysis for variational ansätze
- Depth-normalized entanglement decay comparison
- Zero-Noise Extrapolation (ZNE) for error mitigation
- Transpilation-aware circuit depth and gate analysis
- Publication-ready, reproducible experimental workflow
- Bell-state preparation
- Ideal statevector simulation
- Analytical entanglement verification
- Depolarizing and gate-specific noise
- CZ gate noise for IBM-like superconducting hardware
- Two-qubit interaction noise for IonQ-like trapped-ion systems
- Density matrix simulation
- Concurrence, fidelity, and negativity computation
- Noise-dependent entanglement decay curves
- Parametric variational ansatz
- Expectation value estimation under noise
- Energy degradation analysis vs hardware error
- Zero-Noise Extrapolation (ZNE)
- Comparison of noisy vs mitigated results
All figures are generated programmatically and saved to the figures/ directory.
| Figure | Description |
|---|---|
entanglement_decay.png |
Concurrence vs noise strength |
hardware_comparison.png |
IBM vs IonQ entanglement decay |
energy_vs_noise.png |
VQA energy vs hardware noise |
depth_analysis.png |
Depth-normalized entanglement loss |
zne_comparison.png |
Error-mitigated vs noisy results |
- Python ≥ 3.9
- pip
- Git
git clone https://github.com/MrigankJaiswal-hub/vqa-hardware-benchmark.git
cd vqa-hardware-benchmark
**🚀 Future Work**
Execution on real IBM Quantum and IonQ hardware
Extension to QAOA and larger VQE benchmarks
Advanced error mitigation (PEC, CDR)
Pulse-level noise modeling
ML-assisted hardware-aware ansatz optimization