Development of a Boston-area 50-km fiber quantum network testbed

TL;DR

This work characterizes a Boston-area telecom fiber testbed (BARQNET) to quantify how phase/frequency noise, polarization drift, and optical path-length drift affect quantum signals. Using differential and round-trip configurations, the authors model phase noise as a Brownian process and quantify per-span noise, environmental dependencies, and common-mode effects, enabling a resilient compensation system. They demonstrate a memory-compatible time-bin qubit distribution protocol across a 50 km deployed link, achieving nanosecond-level timing and a mean X-basis error rate of $2.3\%$, with frequency conversion to the visible for SiV memories and periodic polarization correction. The results show BARQNET's noise levels are compatible with narrow-band quantum memories and highlight practical pathways to near-term quantum networking demonstrations, while identifying splicing-related fiber-loss as a key improvement area for scaling to multi-node testbeds and memory-assisted protocols.

Abstract

Distributing quantum information between remote systems will necessitate the integration of emerging quantum components with existing communication infrastructure. This requires understanding the channel-induced degradations of the transmitted quantum signals, beyond the typical characterization methods for classical communication systems. Here we report on a comprehensive characterization of a Boston-Area Quantum Network (BARQNET) telecom fiber testbed, measuring the time-of-flight, polarization, and phase noise imparted on transmitted signals. We further design and demonstrate a compensation system that is both resilient to these noise sources and compatible with integration of emerging quantum memory components on the deployed link. These results have utility for future work on the BARQNET as well as other quantum network testbeds in development, enabling near-term quantum networking demonstrations and informing what areas of technology development will be most impactful in advancing future system capabilities.

Figures (5)

• Figure 1: (a) Schematic of the BARQNET fibers connecting MIT Lincoln Laboratory in Lexington, MIT in eastern Cambridge, and Harvard University in central Cambridge. The solid red line shows a $\sim$40 km segment where the exact route is known, and the dashed line shows the final portion where the exact route could not be obtained. (b--d) Three different connectivity topologies explored in this work. The two copropagating fibers connecting MIT-LL and MIT are labeled A and B, and the two copropagating fibers connecting MIT and Harvard are labeled C and D.
• Figure 2: Measurement of phase drift over the deployed link in both (a) Differential and (b) Round-Trip Configurations, downsampled to 50 kHz to remove noise from our measurement equipment. Differentiating these data provides a measurement of frequency shift over time. Due to the random nature of these fluctuations, this shift is effectively a frequency broadening, with a profile given by the histograms shown in (c) and (d), which are fit to Gaussian profiles (red) with variances $V_D=1.72$ kHz$^2$ and $V_R=21.2$ kHz$^2$ for the Differential and Round-Trip data respectively.
• Figure 3: Measurement of polarization drifts. (a) Example trace of polarization drift measured in the differential configuration, showing the Stokes parameters over a typical 12 hour period. The 10 minute rolling average of this drift is calculated for (b) the Differential Configuration and (c) the Round-Trip Configuration, each compared with the concurrent average wind speed recorded at MIT-LL. The correlations between drift rate and the square of the wind speed are plotted against one another in (d--e), and fit with an exponential dependence (red). Point colors in (d) and (e) correspond to the like-colored time-points in (b) and (c), respectively.
• Figure 4: Optical path length drifts. Time-of-flight measurements in the (a) Differential Configuration measured relative to the nominal delay $t_D=108.4$ ns, and in the (b) Round-Trip Configuration measured relative to the nominal delay $t_R=415.045$ µ s. For each we also plot the temperature measured concurrently outside of MIT-LL. The resultant correlation yields a linear fit (red dashed line) with a dependence of (c) 33.9(2) ps/$^\circ$C for the Differential Configuration and (d) 2.59(1) ns/$^\circ$C for the Round-Trip Configuration. Point colors in (c) and (d) correspond to the like-colored time-points in (a) and (b), respectively.
• Figure 5: Optical setup for memory-compatible time-bin qubit (a) transmitter and (b) receiver. A-EOM = amplitude electro-optic modulator, P-EOM = phase electro-optic modulator, AOM = acousto-optic modulator, EDFA = erbium-doped fiber amplifier, VOA = variable optical attenuator, PID = proportional-integral-derivative feedback controller.