Source-linked AI summary

Microcomb-based true-time-delay network for microwave beamforming with arbitrary beam pattern control

Xiaoxiao Xue, Yi Xuan, Chengying Bao, Shangyuan Li, Xiaoping Zheng, Bingkun Zhou, Minghao Qi, Andrew M. Weiner

arXiv:1710.00045v2physics.app-pheess.SP

TL;DR

Phase-shifter PAAs are bandwidth-limited by beam squint, while large photonic TTD systems remain challenging because of hardware complexity. This paper demonstrates a microcomb-based TTD network using optical phase modulation and programmable spectral shaping. It supports a 21-element PAA, with an 8–20 GHz operating range and beam scanning over ±60.2°; the beamforming evaluation is simulated rather than performed with a real antenna array.

  • Problem

    Traditional phase-shifter PAAs have limited bandwidth because of beam squint, and scaling photonic TTD networks is challenging because of hardware complexity.

    Source: §I · p. 1
  • Method

    The network uses a miniature microcomb source, dispersive time delays, optical phase modulation, and programmable spectral shaping to control microwave beam patterns.

    Source: §I · p. 2, §III · p. 5
  • Results

    The demonstrated network supports 21 PAA elements, operates over 8–20 GHz, and scans beams across ±60.2°; its simulated TTD beams show no obvious divergence across 12, 15, and 18 GHz.

    Source: §IV · p. 6, §III · p. 6
  • Takeaways & Limitations

    Microcombs are promising compact multi-wavelength sources for microwave TTD beamforming, and increasing comb-line count could potentially support larger PAAs.

    Source: §IV · pp. 6–7

    The experiments used EDFAs, which introduce ASE noise; optical SNR ranges from 11 to 31 dB across channels under a standard 10 GHz bandwidth assumption.

    Source: §IV · p. 7

Abstract

from arXiv · show

Microwave phased array antennas (PAAs) are very attractive to defense applications and high-speed wireless communications for their abilities of fast beam scanning and complex beam pattern control. However, traditional PAAs based on phase shifters suffer from the beam-squint problem and have limited bandwidths. True-time-delay (TTD) beamforming based on low-loss photonic delay lines can solve this problem. But it is still quite challenging to build large-scale photonic TTD beamformers due to their high hardware complexity. In this paper, we demonstrate a photonic TTD beamforming network based on a miniature microresonator frequency comb (microcomb) source and dispersive time delay. A method incorporating optical phase modulation and programmable spectral shaping is proposed for positive and negative apodization weighting to achieve arbitrary microwave beam pattern control. The experimentally demonstrated TTD beamforming network can support a PAA with 21 elements. The microwave frequency range is $\mathbf{8\sim20\ {GHz}}$, and the beam scanning range is $\mathbf{\pm 60.2^\circ}$. Detailed measurements of the microwave amplitudes and phases are performed. The beamforming performances of Gaussian, rectangular beams and beam notch steering are evaluated through simulations by assuming a uniform radiating antenna array. The scheme can potentially support larger PAAs with hundreds of elements by increasing the number of comb lines with broadband microcomb generation.

I. INTRODUCTION

Traditional phase-shifter PAAs have limited bandwidth because of beam squint, while photonic true-time-delay systems face hardware-complexity challenges at large scale. This work presents a microcomb-based network using optical phase modulation and programmable spectral shaping for beam-pattern control.

I. INTRODUCTION

  • Photonic TTD networks can address beam squint, but large systems face hardware-volume and complexity challenges, including the use of multiple lasers and tunable delay units.WDM and photonic integration reduce some hardware burdens, but the cited integrated systems have not matched discrete systems in bandwidth, delay range, and loss.Source: §I · p. 1
  • Microcombs offer closely spaced comb lines and broad bandwidth that can potentially support large antenna arrays with reduced source complexity compared with laser arrays.The cited passage notes microcomb line spacing can reach tens to hundreds of GHz, yielding a large Nyquist zone.Source: §I · pp. 1–2
  • The proposed photonic TTD network uses a miniature microcomb source and programmable spectral shaping with optical phase modulation for arbitrary microwave beam-pattern control.The paper reports measured microwave amplitudes and phases and evaluates beamforming performance by simulation because a real antenna array was not built.Source: §I · p. 2

II. SCHEME OF THE MICROCOMB-BASED MICROWAVE TTD

The network generates a broadband comb, modulates it with a microwave signal, and uses a programmable dispersive delay matrix to set relative delays across comb lines. Spectral shaping controls the beam pattern, while switching the matrix controls beam direction.

II. SCHEME OF THE MICROCOMB-BASED MICROWAVE TTD

  • A broadband comb is microwave-modulated and spectrally shaped before passing through a programmable binary dispersion matrix built from optical switches and dispersive fibers.Different comb lines acquire different delays in the dispersive matrix.Source: §II · p. 2
  • After delay, the comb lines are demultiplexed and photodetected, and the resulting microwave signals are amplified for the antenna array.The figure identifies the continuous-wave laser, phase modulator, wavelength demultiplexer, photodetector, and microwave amplifier.Source: §II · p. 2, Figure 2 · p. 2
  • Switching the dispersion matrix controls beam direction, whereas shaping the comb spectrum controls the beam pattern.The illustrated system uses a SiN microring to generate the comb; a downstream spectral shaper and EDFA smooth and boost it.Source: §II · p. 2, Figure 2 · p. 2

A. On-chip microcomb generation

A thermally tuned SiN microring generates the microcomb, with tuning also allowing the comb spectrum to be shifted for alignment with wavelength demultiplexers. The measured spectrum is uneven, so an additional shaper smooths it before modulation.

A. On-chip microcomb generation

  • The SiN microring is fabricated in a CMOS-compatible process and has a loaded quality factor of approximately 8 × 10^5, a 100 µm radius, and a 231 GHz free spectral range.A microheater enables thermal tuning; tuning the pump laser and microring together can shift the generated comb for alignment with wavelength demultiplexers.Source: A. On-chip microcomb generation · p. 3
  • The comb spectrum has approximately 28 dB of fluctuation across the C band, beyond the practical equalization capability of the post-modulation shaper.Its 35 dB attenuation-control range would make arbitrary shaping impractical and equalization would add substantial insertion loss, motivating an additional shaper before modulation.Source: A. On-chip microcomb generation · p. 3
  • Twenty-one C-band comb lines were selected and shaped to a truncated Gaussian window after smoothing and amplification.The comb index n runs from 1 to 21.Source: A. On-chip microcomb generation · p. 3

B. Characterization of the dispersion matrix and higher-order dispersion compensation

The dispersion matrix was characterized and fiber-induced nonlinear delay errors quantified. Spectral-shaper compensation reduces comb-band residual errors to below 1 ps, while larger arrays require additional TOD compensation.

B. Characterization of the dispersion matrix and higher-order dispersion compensation

  • Time delay is retrieved by linearly fitting microwave phase against angular frequency; 50 measurements are spaced 1 second apart.The four measured fiber lengths are 1007.992 m, 503.995 m, 251.999 m, and 125.998 m; estimated cutting errors are below 1 cm.Source: B. Characterization of the dispersion… · p. 3, Figure 3 · p. 3
  • Comb-band residual delay errors are below 1 ps after TOD compensation, with 0.18 ps standard deviation; correcting an approximately 30 ps error adds about 3 dB of spectral-shaper loss.The added loss is higher near the comb-spectrum edges than its center.Source: Figure 4 · p. 4, B. Characterization of the dispersion… · p. 4
  • In the “1111” state, fiber TOD makes delay nonlinear, with comb-band errors reaching almost 30 ps, equivalent to 162° at 15 GHz.The retrieved dispersion coefficients are β2 = −21.08 ps2/km and β3 = 0.1254 ps3/km.Source: B. Characterization of the dispersion… · pp. 3–4, Figure 4 · p. 4
  • For large-scale antenna arrays, additional TOD compensation is required; proposed options are opposite-sign-TOD fiber or fiber Bragg gratings designed to control higher-order dispersion.Source: B. Characterization of the dispersion… · p. 4
  • With microcomb source and spectral-shaper compensation for fiber TOD, measured delays across all dispersion-matrix states from “0000” to “1111” were linear versus comb frequency, with RMS error ∼1.1 ps.The estimated error was <1 ps; authors say the larger measured error is likely due to fiber non-uniformity and cross talk from other comb lines during measurement.Source: B. Characterization of the dispersion… · p. 4
  • A delay offset is needed for symmetric beam scanning around the antenna normal; proposed adjustment uses 126 m of dispersion-compensating fiber or differential channel-fiber lengths of about 6 mm.The fiber lengths target a 15 GHz center frequency and a ±60° scan range.Source: B. Characterization of the dispersion… · pp. 3–4

III. BEAM PATTERN CONTROL VIA SPECTRAL SHAPING

Measured tap amplitudes and phases are used to calculate beam patterns, while spectral shaping controls tap weights and dispersion controls steering. The section evaluates Gaussian and rectangular beams, notch steering, and frequency-dependent beam behavior.

III. BEAM PATTERN CONTROL VIA SPECTRAL SHAPING

  • Beam direction can be steered independently of microwave frequency by controlling relative inter-antenna time delay, while tap coefficients determine beam shape.Ideally, both positive and negative taps are needed for flexible pattern control such as a rectangular beam.Source: §III · p. 5
  • Selective suppression of the lower or upper modulation sideband generates positive or negative microwave taps, respectively.The approach uses an optical phase modulator and programmable spectral shaping.Source: §III · p. 5
  • 8 GHz is the measured 3-dB lower cutoff; positive and negative taps show a clear π-phase difference.At low microwave frequencies, limited spectral-shaper resolution prevents efficient sideband suppression and produces a high-pass response.Source: §III · p. 5, Figure 6 · p. 5
  • Local dispersion compensation with the spectral shaper compensates high-frequency microwave phase deviations in the “1111” dispersion state.The uncompensated response deviates from 0 and π; the programmed shaper phase combines local compensation with global third-order dispersion compensation.Source: §III · p. 6
  • −60.2° to 60.2° scanning at 15 GHz was calculated for the Gaussian beam, with 7° to 14° 3-dB widths and out-of-beam suppression above 30 dB.The scan used an average step of 8°; the widths were 7° at ±3.3° and 14° at ±60.2°.Source: §III · p. 6
  • 25° to 50° 3-dB widths were calculated for rectangular beams at scan angles ±3.3° to ±60.2°.The beam patterns were calculated from measured microwave amplitudes and phases after shaping tap coefficients to a sinc function.Source: §III · p. 6
  • ∼25 dB notch suppression was calculated when the 11th tap had the opposite sign from the others; the notch ratio was not optimized.The null occurs where the negative-tap omnidirectional beam cancels the positive beam from the other taps.Source: §III · p. 6
  • 12, 15, and 18 GHz calculations showed no obvious beam divergence for TTD, versus phase-shifter beam directions of 40.4°, 31.4°, and 25.6°.The comparison used a Gaussian beam directed at 31.4° and phase-shifter phases set to those of 15 GHz.Source: §III · p. 6

IV. SUMMARY AND DISCUSSION

The discussion presents a 21-element microcomb TTD network with arbitrary beam-pattern control and describes its operating range, potential scale-up, and remaining implementation constraints.

IV. SUMMARY AND DISCUSSION

  • The demonstrated network supports 21 elements, microwave frequencies from 8 to 20 GHz, and beam scanning from −60.2° to 60.2°.The upper frequency is limited by the measurement equipment.Source: §IV · p. 6
  • At 15 GHz, calculated patterns based on measured amplitudes and phases show Gaussian, rectangular, and dark-beam control.Figure 8 pairs the shaped optical spectra with the corresponding calculated beam patterns.Source: Figure 8 · p. 7
  • With 50 GHz comb spacing, comb lines in the S, C, and L bands could support more than 400 sub-antennas.This is a potential scale-up, not the demonstrated array size.Source: §IV · p. 7
  • Practical applications still require coherent broadband 50 GHz microcombs; type-II comb noise can compromise RF performance, while spectral variation reduces apodization efficiency.The discussion also notes that high intensity noise commonly accompanies type-II combs.Source: §IV · p. 7
  • The current setup uses EDFAs, whose amplified spontaneous emission is identified as the main microwave noise source.The reported electrical SNR is 10–30 dB for a 2.5 GHz electrical bandwidth.Source: §IV · p. 7
  • Chip-scale spectral shapers and more highly dispersive components are discussed as routes to reducing system volume.Photonic crystal fibers, chirped fiber Bragg gratings, or integrated waveguides could provide tens to hundreds of times the dispersion of singlemode fiber, requiring shorter lengths for the same dispersion.Source: §IV · pp. 7–8
  • Beamforming simulations assume identical channel components; real detector, amplifier, and antenna differences could distort beam shape and degrade mainlobe-to-sidelobe suppression.Evaluation involving a real antenna array is identified as future work.Source: §IV · p. 8
Loading 1710.00045v2…