Fig. 1. Experimental setup. A continuous wave laser at 1550 nm is equally divided into four paths through a 1×4 beam splitter (BS1×4). Three polarization controllers are used to align the polarization inside the different cores. The four single-mode fibers are then individually connected to the fan-in input of the MCF. Thanks to the reconfigurability of the optical system, it is possible to set the number of subsequent MCF strands to be tested in the experiment (each strand is approximately 6.29 km in length). To connect the strands to each other, an MCF connector is used. After propagation through the MCF, a fan-out device is used to divide the cores into four different single-mode fibers that are finally combined with each other by three 2×2 beam splitters (BS1, BS2, and BS3). By using three automatic control boards, each driving a phase shifter, based on the measurement output of the three detectors, we can monitor and individually control the relative phase of each core.

Fig. 2. Flow chart of the PID controller. This algorithm is used for controlling each phase shifter in the three different interferometers. M and m in stage 1 stand for the maximum and the minimum of the feedback signal (FS), respectively. LP indicates the desired locking point, while ThA and ThB are user-defined thresholds.

Fig. 3. Visibility fringes of the interferometers. (a) Interference signals as a function of time for the two independent interferometers between cores 1 and 2 and cores 3 and 4. Measured visibility of V1&2=0.981±0.008 and V3&4=0.945±0.011. (b) Interference signal from the three interferometers. In this measurement, the first two independent interferometers are locked to a fixed position. Measured visibility of V1&2&3&4=0.989±0.004.

Fig. 4. Phase drifts of the non-stabilized and stabilized four concatenated strands multicore interferometers over 10 min acquisition. (a) Phase drift between two-core and four-core interferometers without active phase stabilization. (b) Phase drift of (a) but with active phase stabilization loops. Different colors represent the three different configurations (red and blue, two-path interferometers; yellow, four-path interferometer).

Fig. 5. Frequency analysis of the interference signals. (a) Two-core intereferometers for the four different strands. (b) Four-core interferometer for the four different strands. Different colors represent different strands. Each measurement has been acquired for 30 min in the non-locked system.