Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2611.07729

Unified design of short mode-evolution couplers

S. -Y. Tseng
View original source ↗Read source document ↗Export metadata ↓

Abstract

Short mode-evolution couplers have been designed with several seemingly different approaches, including fast quasi-adiabatic dynamics (FAQUAD), counter-diabatic (CD) driving realized by tailoring waveguide widths and spacing, CD driving realized by axis bending, and numerical optimization of the coupler geometry. Using exact two-mode coupled-mode theory in the Kramers--Henneberger frame of the bent coupler, we show that these approaches belong to the same design space. The three geometric parameters, width difference, spacing, and axis tilt, act on the light through two functions only: the coupling strength, set by the spacing, and an effective detuning, set by the width difference and the axis curvature. The gauge, bent, and mixed realizations of a CD coupler are therefore the same device and have the same bandwidth. The bandwidth is set by the angle $χ$ between the light and the field that steers it on the Bloch sphere, integrated along the device as $\int\sin^2χdz$, and for a CD coupler this integral is set by the reference alone. This leads to a lithography-aware design rule: make the reference as adiabatic as the minimum gap allows, by narrowing its spacing at the anticrossing, and let exact CD suppress the remaining nonadiabatic coupling. In beam propagation method (BPM) simulations on a polymer platform, a 500\um{} coupler designed this way, without numerical optimization, maintains ${>}99\%$ transfer over 274~nm at a minimum edge gap of 1.05\um, compared with 150~nm for the CD coupler of a fixed-spacing reference at nearly the same gap. Particle swarm optimization of the geometry with BPM in the loop does not exceed the design rule. Three-dimensional BPM of the BCB channel waveguides also confirms these conclusions.

Explore related subjects

physics.optics

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

S. -Y. Tseng. 2026-11-05. Unified design of short mode-evolution couplers. https://arxiv.org/abs/2611.07729

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Nonperturbative regime of low-order harmonic generation in intense low-frequency laser field

We find the atomic response to the intense femtosecond laser pulse via solving numerically the three-dimensional non-stationary Schrödinger equation (TDSE) for a model atom and calculating its dipole moment. For weak quasi-static fields, the response is well described by a perturbation approach, but for intensities higher than about $0.6 \, \, 10^{14}$ W/cm$^2$ the accuracy of this description is unsatisfactory, regardless of the order of non-linearity taken into account. We suggest fitting the numerical TDSE solution results with a Padé expansion, and show that this approximation describes the response well both in the perturbative regime and beyond it for intensities approximately up to $1.4 \, \, 10^{14}$ W/cm$^2$. To consider the non-perturbative nonlinearity beyond the quasi-static limit we use the model of nonlinear oscillator with the restoring force defined by the found Padé expression. Our model fails to predict the behaviour of the nonlinear refractive index in the nonperturbative domain, but it describes well the nonperturbative growth of the efficiency with the laser intensity for other nonlinear optical processes, namely, the third and fifth harmonic generation in the IR field and the optical rectification in a two-color field.

physics.optics↗
arXiv2026

Digitally enhanced Multi-wavelength Stabilization using a Passive Fiber Frequency Reference

High-stability frequency stabilization between lasers is essential for applications such as precision metrology, frequency dissemination, and quantum communications. We demonstrate the use of code-based multiplexing of a passive fiber-based interferometer for frequency stabilization and transfer between lasers separated by 85 GHz in frequency using a single shared passive reference. Out-of-loop characterization places an upper bound of sub-kHz/$\sqrt{\text{Hz}}$ on the transferred frequency noise at Fourier frequencies above 40 mHz, with a differential fractional stability of $7\times10^{-13}$ at 1 second. We also demonstrate continuous laser tunability and characterize the residual length noise coupling. Finally, we provide a framework for scaling the architecture to higher laser counts, offering a pathway to scalable, frequency agile stabilization of multiple light sources without the overhead of an optical frequency comb.

physics.optics↗
arXiv2026

Optical Coherent Combining in Multi-Aperture Free-Space Receivers through Atmospheric Turbulence via Optically Computed Alignment Gradients

Multi-aperture reception yields a coherent-combining gain only if the aperture fields are aligned in phase and polarization, which in a downlink must track the atmosphere. We show that the exact gradients of the combined power with respect to the phase and polarization angle of each branch can be measured optically, before photodetection: a passive tap of the combined field, split one-to-$N$, is the common self-reference and, interfered with a branch tap on a 3 dB coupler, yields the phase gradient; with a $90^\circ$ Faraday rotator and a static $π/2$ offset it yields the polarization gradient. Each branch then runs fixed-gain phase and polarization loops, without division, perturbation, dither, reference laser, or per-branch receiver, as analog integrators or a low-rate processor. We derive the gradients, their optical readout, and the loop stability condition. A phase-only simulation with identical disturbances, a common 100 MS/s clock, and retuning of every scheme compares the loop with centralized and distributed stochastic parallel gradient descent and with dither locking at the OSNR where ideal four-beam combining reaches the 28 GBd DP-QPSK FEC threshold. Up to $10^5$ rad/s of multi-tone disturbance the loop keeps the lowest residual loss, at most $5.6\times10^{-4}$ dB for four beams and 13 to 160 times below the alternatives, and reaches $90\%$ efficiency in 0.6 $μ$s for up to 64 beams. The taps leave the receiver OSNR unchanged and every retuned scheme stays within 0.06 dB of ideal combining, whereas brief dither misalignments cost 0.14 dB. Up to $10^7$ rad/s it preserves the ideal-combining bit-error ratio for four, sixteen, and sixty-four beams while every baseline crosses the threshold and slips carrier recovery. In fades it loses $4\times10^{-3}$ dB where the others lose 0.2 to 0.9 dB. Assumptions and limitations are stated; experimental validation remains open.

physics.optics↗