Source-linked AI summary
Coherent optical communications using coherence-cloned Kerr soliton microcombs
Yong Geng, Heng Zhou, Wenwen Cui, Xinjie Han, Qiang Zhang, Boyuan Liu, Guangwei Deng, Qiang Zhou, Kun Qiu
TL;DR
Coherent optical systems need receiver local oscillators that preserve the frequency and phase coherence of transmitted carriers, a receiver-side Kerr soliton microcomb regeneration target that has not been comprehensively investigated. The paper regenerates a microcomb over 50 km using pump conveyance and two-point locking, then demonstrates terabit coherent interconnect with simplified frequency-offset and carrier-phase estimation.
Problem
Receiver-side regeneration of a Kerr soliton microcomb that preserves transmitted carriers’ frequency and phase coherence has not been comprehensively investigated, despite its importance for coherent communications.
Source: Introduction · pp. 2–3Method
The authors convey the pump laser over 50 km and use a second pilot tone for two-point locking to regenerate a coherence-cloned receiver microcomb.
Source: Introduction · p. 3Results
1.68 Tbit s−1 across 20 channels was demonstrated using coherence-cloned microcombs, dispensing with traditional FOE and substantially simplifying CPE.
Source: Results · p. 6, Discussion · p. 9Takeaways & Limitations
The authors conclude that regulating microcomb frequency and phase coherence can improve coherent communication and propose local optical frequency standards for multi-node networks.
Source: Discussion · p. 9Detailed receiver-circuit design for fully exploiting coherence-cloned microcombs is outside the scope of this work.
Source: Results · p. 8
Abstract
from arXiv · showhide
Dissipative Kerr soliton microcomb has been recognized as a promising on-chip multi-wavelength laser source for fiber optical communications, as its comb lines possess frequency and phase stability far beyond independent lasers. In the scenarios of coherent optical transmission and interconnect, a highly beneficial but rarely explored target is to re-generate a Kerr soliton microcomb at the receiver side as local oscillators that conserve the frequency and phase property of the incoming data carriers, so that to enable coherent detection with minimized optical and electrical compensations. Here, by using the techniques of pump laser conveying and two-point locking, we implement re-generation of a Kerr soliton microcomb that faithfully clones the frequency and phase coherence of another microcomb sent from 50 km away. Moreover, leveraging the coherence-cloned soliton microcombs as carriers and local oscillators, we demonstrate terabit coherent data interconnect, wherein traditional digital processes for frequency offset estimation is totally dispensed with, and carrier phase estimation is substantially simplified via slowed-down phase estimation rate per channel and joint phase estimation among multiple channels. Our work reveals that, in addition to providing a multitude of laser tones, regulating the frequency and phase of Kerr soliton microcombs among transmitters and receivers can significantly improve coherent communication in terms of performance, power consumption, and simplicity.
Introduction
Coherent WDM depends on frequency and phase coherence between data carriers and receiver local oscillators, yet commercial systems commonly use weakly coherent independent lasers. The paper targets receiver-side regeneration of a Kerr soliton microcomb that preserves the transmitted comb’s coherence.
Motivation
- Independent carrier and local-oscillator lasers have weak mutual coherence and require guard bands and power-hungry DSP to estimate frequency and phase uncertainties.Optical frequency combs offer many evenly spaced, phase-locked tones with greater spectral stability than individual lasers.Source: Introduction · p. 2
Research gap
- Receiver-side regeneration of a DKS microcomb that inherits the transmitted comb’s frequency and phase coherence remains incomprehensively investigated.The paper identifies this as important for coherent communication networks.Source: Introduction · pp. 2–3
DKS coherence control
- A shared pump sets the comb’s central frequency, cavity tuning adjusts mode spacing, and soliton-line phases align to the pump phase.Together, these properties allow the comb’s spectral parameters to be manipulated through pump-laser tuning.Source: Introduction · p. 3
Contribution
- Pump-laser conveyance over 50 km and a second pilot tone for two-point locking regenerate a microcomb that clones the transmitter’s frequency and phase coherence.The approach uses two-point locking and optical frequency division to stabilize the regenerated comb.Source: Introduction · p. 3
Results
Two-point locking and optical frequency division yield a coherent receiver microcomb and support 1.68 Tbit s−1 interconnect. The resulting frequency and phase stability simplifies frequency-offset and carrier-phase estimation.
Coherence-cloned re-generation of DKS microcomb
- The transmitter pump regenerates the receiver DKS microcomb after propagation through 50 km of standard single-mode fiber, with auxiliary laser heating used at both cavities.The combs can operate for weeks when maintained by stabilization techniques.Source: Results · p. 4
- Two-point locking narrowed the 17th-line beat-note linewidth to ∼5 Hz and improved frequency stability by about 4 orders of magnitude at 1 s gate time.Optical frequency division suppresses noise in the intervening comb lines; the 17th line was chosen because of the phase comparator’s bandwidth limitation.Source: Results · pp. 4–5
Coherent data interconnect using coherence-cloned microcombs
- After data-channel co-propagation, the pilot-to-receiver beat note showed almost no change, indicating negligible XPM-induced linewidth distortion in this experiment.Fiber-dispersion walk-off smooths XPM from different data channels into a quasi-constant phase envelope without a high-frequency component.Source: Results · p. 6
- 1.68 Tbit s−1 across 20 data channels was received using the cloned combs, with frequency offsets precalculated rather than estimated by conventional DSP.The precalculated offsets differed from measured offsets by within ±500 Hz across all 20 channels; without two-point locking, errors were about 3 orders of magnitude larger.Source: Results · pp. 6–7
- Two-point-locked combs enabled 1 order of magnitude lower CPE rate than unlocked microcombs and 3 orders of magnitude lower than independent lasers at the 3.8e-3 BER target.For lower-indexed channels, one 32-symbol CPE block sufficed to meet the target BER over a 400,000-symbol frame.Source: Results · p. 7
- Strong cross-channel phase correlations enabled master-slave CPE, with only minor BER penalty when channel 10’s CPE result detected channel 1.Detailed receiver-circuit design for fully exploiting coherence-cloned microcombs is outside the current work’s scope.Source: Results · p. 8
Discussion
The discussion presents a trade-off between BER and CPE simplicity, with large CPE savings in a 10-channel example. It also notes that conveying pump and pilot tones is difficult in multi-node networks, while local frequency standards are a possible alternative.
Discussion
- The authors frame CPE rate and BER as a system-specific trade-off and report that source savings can scale up as the number of data channels increases.The example achieves its target BER by running CPE every 1001 data blocks on channel 10 and sharing its result with channels 1–9.Source: Discussion · pp. 8–9
- At a 3.8e-3 BER target, 10 channels required 13 CPE operations with coherence-cloned combs versus 12,500 with independent lasers over 4,000,000 decoded symbols.The comparison uses channel 10’s CPE result to detect channels 1–9 and reports more than 3 orders of magnitude fewer pilot symbols and CPE operations.Source: Discussion · p. 9
- The demonstrated point-to-point scheme conveys a pump laser and pilot tone, a challenge for multi-node networks; local optical frequency standards are proposed as an alternative.The authors state that sufficiently stable standards across network nodes could retain the discussed frequency-offset and phase-estimation benefits.Source: Discussion · p. 9
Methods
The transmitter soliton comb is regenerated at the receiver using the conveyed pump laser, then locked to the transmitter comb at a pilot line. The setup transmits modulated comb lines over 50 km and uses the regenerated comb for coherent reception.
Generation and locking of CTx and CRx.
- The receiver regenerates its soliton comb using the transmitter pump conveyed over 50 km, while an auxiliary laser controls receiver-cavity detuning and thermal stability.The transmitter comb is generated using auxiliary-laser heating to balance heat flow and access a single-soliton state.Source: Methods · p. 10
- Two-point locking detects the pilot-line beat note and feeds back auxiliary-laser power to adjust receiver-comb repetition rate and lock the receiver pilot line to the transmitter.The phase-lock-loop bandwidth is about 100 kHz.Source: Methods · pp. 10, 12
Coherence analysis between CTx and CRx.
- The phase model tracks pump and comb-line phase terms, including intrinsic pump noise, fiber-acquired phase noise, nonlinear phase modulation, and repetition-rate fluctuations.The model assumes aligned intrinsic comb-line phases under DKS mode locking and neglects nonlinear cross-phase modulation because it is small in the 50 km fiber experiment.Source: Methods · p. 11
- Fiber dispersion causes comb lines to walk off in time, producing low-frequency phase noise below 3.0 kHz that the authors say hardly affects carrier-phase estimation or data reception.The mode-index-linear phase term is used for master-slave carrier-phase estimation.Source: Methods · p. 13
Coherent data modulation and receiving using CTx and CRx.
- Twenty comb lines carry single-polarization 16-QAM data through 50 km of fiber, and the regenerated receiver comb serves as the local oscillator for coherent detection.The setup also transmits the pump and a pilot tone; offline processing includes dispersion compensation, Volterra equalization, and pilot-aided carrier phase estimation.Source: Methods · pp. 13–14