Optimizing TWPA pump conditions for Multiplexed Readout

Travelling Wave Parametric Amplifiers (TWPAs) are widely used in the readout of superconducting qubits. By providing near quantum-limited amplification over a broad bandwidth, TWPAs significantly enhance the signal-to-noise ratio (SNR), thereby enabling faster and more accurate qubit state discrimination. In particular, their large bandwidth makes them suitable for multiplexed readout, which is essential for large scale superconducting quantum processors. However, there is no standardized procedure for tuning the TWPA pump parameters to optimize performance under multiplexed operation. In turn, this limits the potential of TWPAs in shortening readout time, a critical component for faster and more efficient operation of quantum devices.

In this work, we present a systematic methodology for optimizing TWPA pumping conditions aimed at minimizing multiplexed readout duration on superconducting qubits. Our approach begins by characterizing both the qubits and the TWPA, followed by pump optimization targeted at the worst-performing readout that limits the overall multiplexed readout speed. This optimization is iteratively repeated for the newly identified bottleneck qubit until the TWPA’s amplification capability (SNR improvement) is saturated. Through this adaptive optimization process, we achieve the minimal readout duration for a given system while ensuring a certain readout fidelity across all qubits. This methodology is readily applicable to any multiplexed readout architectures, providing a scalable pathway that ultimately supports fault-tolerant quantum computing.