--- title: PDH Qubit Readout Technique url: https://www.emergentmind.com/topics/pdh-qubit-readout type: topic --- # PDH Qubit Readout Technique The Pound–Drever–Hall (PDH) qubit readout technique is an ultrastable superconducting-qubit measurement protocol adapted from optical cavity stabilization for dispersive readout in circuit quantum electrodynamics (cQED). It employs a multi-tone, self-phase-referenced microwave scheme that renders readout signals insensitive to slow phase drifts, enhances measurement signal-to-noise ratio (SNR), and suppresses measurement-induced state transitions. The approach enables high-fidelity, scalable readout, crucial for the operation of large-scale quantum processors [2512.03138]. ## 1. Multi-Tone Self-Phase-Referenced PDH Readout Architecture Unlike conventional single-tone heterodyne readout, which is sensitive to absolute phase drift between the carrier and the local oscillator (LO), the PDH method utilizes three phase-coherent tones: a carrier at frequency $\omega_0$ and two sidebands at $\omega_0 \pm \omega_m$. These are generated via IQ modulation and sent into the cQED readout resonator. After reflection (or transmission), the combined signal is detected using a square-law detector, with beat notes at the modulation frequency $\omega_m$ providing a sensitive probe of cavity frequency shifts induced by the qubit state. ``` ┌──────────────┐ ┌──────────┐ ┌──────────────┐ ┌────────┐ │ Vector gen. │─IQ│ Cryo │──►│ Power-sq. │──■──▶│ RF LO │ │ (ω₀, ±ωₘ) │ │ atten │ │ detector (|E|²→RF) │ │mixers │ └──────────────┘ └──────────┘ └──────────────┘ └───┬────┘ ▼ Digitizer (I/Q at DC) ``` In the absence of a cryogenic microwave square-law detector, the error signal can be reconstructed synthetically by recording all three tones with a room-temperature heterodyne setup, followed by digital recombination of their I/Q traces [2512.03138, SI Sec. S5]. ## 2. Derivation of the PDH Error Signal Consider the detector input field composed of: - Carrier: $E_0 e^{i(\omega_0 t + \phi_0)}$ - Upper sideband: $E_+ e^{i(\omega_+ t + \phi_+)}$, $\omega_+ = \omega_0 + \omega_m$ - Lower sideband: $E_- e^{i(\omega_- t + \phi_-)}$, $\omega_- = \omega_0 - \omega_m$ After the resonator, each acquires the scattering coefficient $S(\omega)$. The total field at the detector is: $$\mathcal{E}_{\mathrm{det}}(t) = \sum_{j = -,0,+} E_j e^{i(\omega_j t + \phi_j)}.$$ The detector measures the total power, $$P(t) = |\mathcal{E}_{\mathrm{det}}(t)|^2,$$ which contains beat notes at $\pm \omega_m$ and $2\omega_m$. Isolating terms at $\omega_m$ frequency: $$ P(t) \supset 2\,\mathrm{Re}\left[ E_+E_0 e^{i(\phi_0 - \phi_+)} + E_-E_0 e^{i(\phi_- - \phi_0)} \right]\cos(\omega_m t) + 2\,\mathrm{Im}\left[ E_+E_0 e^{i(\phi_0 - \phi_+)} + E_-E_0 e^{i(\phi_- - \phi_0)} \right] \sin(\omega_m t), $$ yielding the complex error signal (Eqs. S7–S10, [2512.03138]): \[ \epsilon(\omega) = B_+ + B_-, \quad B_+ \equiv E_+E_0 e^{i(\phi_0 - \phi_+)}, \quad B_- \equiv E_-E_0 e^{i(\phi_- - \phi_0)} \] with in-phase (I) and quadrature (Q) components: \[ \epsilon_I = E_+E_0 \cos(\phi_0-\phi_+) + E_-E_0 \cos(\phi_0-\phi_-), \\ \epsilon_Q = E_+E_0 \sin(\phi_0-\phi_+) - E_-E_0 \sin(\phi_0-\phi_-) \] When $\omega_0$ is near the bare resonator frequency $\omega_r$, $\epsilon_Q$ exhibits a steep, nearly linear zero-crossing, permitting the conversion of dispersive frequency shifts $\Delta\omega_0$ (from the qubit state) into voltage changes. ## 3. Lock Slope, Signal-to-Noise Ratio, and Signal Enhancement The lock slope $S$ of the PDH error signal is: \[ S \equiv \left. \frac{\partial\epsilon_Q}{\partial\omega} \right|_{\omega_0 = \omega_r} \approx \frac{2 E_0 E_{sb}}{\kappa} \] for equal sideband amplitudes $E_{sb}$, and resonator linewidth $\kappa$. For integration time $T$ and white noise spectral density $S_n$ (V$^2$/Hz), the root-mean-squared noise is $\sigma = \sqrt{S_n/(2T)}$, giving single-shot SNR: \[ \mathrm{SNR} = \frac{\Delta\epsilon_Q}{\sigma} = \frac{S \cdot \Delta\omega}{\sqrt{S_n/(2T)}} \] Amplifying the sideband power by $G_{\mathrm{dBc}}$, the gain in the error signal is \[ E_{sb}/E_0 = 10^{G_{\mathrm{dBc}} / 20} \implies \epsilon_Q \propto E_0^2 10^{G_{\mathrm{dBc}}/20}. \] Relative to single-tone heterodyne readout, this provides a $+28$ dB power ($+14$ dB amplitude) enhancement in SNR for sidebands at $+28$ dBc [2512.03138]. ## 4. Phase Drift Immunity and Suppression of Measurement-Induced Transitions The PDH error signal depends only on relative carrier–sideband phases $\phi_0 - \phi_\pm$. Any common-mode phase shift $\alpha$ in all tones cancels: \[ (\phi_0 + \alpha) - (\phi_\pm + \alpha) = \phi_0 - \phi_\pm. \] Small differential phase errors $\beta = (\omega_m \Delta L / v_p) \ll 1$ due to path differences or digitizer delays merely rotate the I/Q readout frame, which can be compensated by choosing the optimal projection axis or using the “scissors phase” $\Sigma \equiv 2\phi_0 - (\phi_+ + \phi_-)$, invariant to both common and differential drifts (SI Sec. S6). Experimental benchmarks demonstrate that, without any phase locking, the ordinary heterodyne phase $\phi_0$ drifts by approximately $200^\circ$ over 2 hours, whereas the PDH differential phase $\phi_0 - \phi_-$ exhibits an rms fluctuation of only $0.44^\circ$ over the same period. Cluster separation in single-shot histograms remains stable with $<1^\circ$ rms over minutes [2512.03138, SI Fig. S4]. Measurement-induced state transitions (MIST) are suppressed when sideband detuning $\omega_m \ge 20$ MHz ($\simeq 30\kappa$) and sideband power is limited to $+28$ dBc relative to the carrier. No measurable increase in MIST was observed up to this power, establishing that the full PDH heterodyne gain can be exploited without degrading qubit lifetime or QND readout fidelity [2512.03138, SI Fig. S9]. ## 5. Readout Fidelity and Assignment Error Single-shot readout is implemented by projecting each $\epsilon_Q$ onto the optimal discrimination axis and applying a threshold. Letting $\Delta V$ be the mean separation between ground and excited state in the projected space and $\sigma$ the rms noise, the single-shot SNR is $SNR_0 = \Delta V/\sigma$. The assignment error for Gaussian readout statistics is: \[ P_{\mathrm{err}} \simeq \frac{1}{2} \mathrm{erfc}(SNR_0/2) \] For PDH readout with JPA preamplification, measured $SNR_0$ is $6–8$ for $T=1\ \mu$s integration, yielding single-shot assignment fidelities $F = 1 - P_{\mathrm{err}} \approx 98.5\%–99\%$. Here, $\Delta V = \epsilon_Q(e) - \epsilon_Q(g)$, with amplifier and vacuum noise included in $S_n$ [2512.03138]. ## 6. Scaling to Multi-Qubit Architectures: Bandwidth, Crosstalk, and Instrumentation For simultaneous multi-qubit readout, the following considerations apply: - **Tone allocation**: Each readout resonator and associated sidebands must avoid overlap with other resonators. With $\omega_m \sim 20$ MHz and inter-resonator spacings $\ge 100$ MHz, approximately five qubits per octave of readout bandwidth can be accommodated. - **Crosstalk**: Unintended excitation arises only if PDH sidebands of one qubit overlap with another’s resonance. Slightly detuning modulation frequencies between qubits and careful frequency planning mitigate this risk. - **Room-temperature electronics and cryogenic detection**: Implementation demands a multi-channel AWG or vector source for per-qubit carrier/sideband generation, a broadband cryogenic square-law detector or low-reflection mixer at 4 K, traveling-wave parametric amplifiers covering wide bandwidths, and multi-tone LO chains with digital downconversion. Replacing single-mode JPAs with TWPAs extends coverage to multiple qubit bands. A summary of key implementation benchmarks: | Parameter | Value/Comment | Source | |------------------------|------------------------------------------------------------|----------------| | Sideband spacing | $\omega_m \ge 20$ MHz, typically $\sim 30\,\kappa$ | SI, S9 | | Inter-resonator gap | ≥100 MHz (limits per-band qubits to $\sim$5/octave) | SI, multi-qubit| | Sideband/carrier ratio | up to +28 dBc (no MIST increase at this level) | SI, S9 | | Phase stability | $<0.5^\circ$ rms over hours (PDH differential phase) | SI, S4 | | Single-shot fidelity | 98.5–99% at 1 µs integration with JPA | Main text | | SNR enhancement | +14 dB amplitude vs. heterodyne for +28 dBc sidebands | Main text | ## 7. Context and Significance Embedding qubit-state information in the beatnotes of phase-coherent carrier and sidebands, PDH qubit readout offers robust immunity to slow phase fluctuations, large intrinsic heterodyne gain, and scalability for parallel multi-qubit operation without increased decoherence or loss of quantum nondemolition (QND) character. These attributes address several bottlenecks in superconducting qubit readout, enabling high-fidelity, stable, and scalable state discrimination essential for error-corrected quantum computation [2512.03138]. A plausible implication is that, with further integration of wideband cryogenic detectors and traveling-wave parametric amplifiers, the PDH protocol can be extended to the simultaneous readout of many qubits with minimal hardware overhead and maximum SNR, positioning this technique as a standard approach for next-generation cQED processor architectures. Source: https://www.emergentmind.com/topics/pdh-qubit-readout