End-to-End Fidelity Analysis of Quantum Circuit Optimization: From Gate-Level Transformations to Pulse-Level Control

Abstract: We present a comprehensive analysis of quantum circuit fidelity across the full compilation stack, from high-level gate optimization through pulse-level control. Using a modular integration framework connecting a C++ circuit optimizer with Lindblad-based pulse simulation, we systematically evaluate the fidelity impact of four optimization passes: gate cancellation, commutation, rotation merging, and identity elimination, on IQM Garnet hardware parameters. Our simulation campaign spanning 371 circuit runs reveals that gate cancellation provides the most significant improvement (68\% of circuits improved, 14,024 gates eliminated), while pulse duration exhibits the strongest negative correlation with process fidelity ($r = -0.74$, $R^2 = 0.55$). We validate these findings through hardware execution on the IQM Resonance Garnet 20-qubit processor, demonstrating 70\% gate reduction on QFT circuits with 100\% job success rate (8 executions). Our open-source framework enables reproducible benchmarking of quantum compilation pipelines.