Next Article in Journal
Design and Mixing Analysis of a Passive Micromixer Based on Curly Baffles
Previous Article in Journal
High-Performance All-Optical Logic Operations Using Ψ-Shaped Silicon Waveguides at 1.55 μm
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Micromachines
Volume 14
Issue 9
10.3390/mi14091794
micromachines-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Comparison of Microcomb-Based Radio-Frequency Photonic Transversal Signal Processors Implemented with Discrete Components Versus Integrated Chips

by
Yang Sun
,
Jiayang Wu
*,
Yang Li
and
David J. Moss
*
Optical Sciences Centre, Swinburne University of Technology, Hawthorn, VIC 3122, Australia
*
Authors to whom correspondence should be addressed.
Micromachines 2023, 14(9), 1794; https://doi.org/10.3390/mi14091794
Submission received: 15 August 2023 / Revised: 9 September 2023 / Accepted: 18 September 2023 / Published: 20 September 2023
(This article belongs to the Special Issue Novel Integrated Photonic Devices Incorporating 2D Materials)
Browse Figures
Versions Notes

Abstract

:
RF photonic transversal signal processors, which combine reconfigurable electrical digital signal processing and high-bandwidth photonic processing, provide a powerful solution for achieving adaptive high-speed information processing. Recent progress in optical microcomb technology provides compelling multi-wavelength sources with a compact footprint, yielding a variety of microcomb-based RF photonic transversal signal processors with either discrete or integrated components. Although they operate based on the same principle, the processors in these two forms exhibit distinct performances. This paper presents a comparative investigation of their performances. First, we compare the performances of state-of-the-art processors, focusing on the processing accuracy. Next, we analyze various factors that contribute to the performance differences, including the tap number and imperfect response of experimental components. Finally, we discuss the potential for future improvement. These results provide a comprehensive comparison of microcomb-based RF photonic transversal signal processors implemented using discrete and integrated components and provide insights for their future development.

1. Introduction

Driven by the exponential growth of data capacity, there has been a rapid increase in the demand for high-speed information processing. Radio-frequency (RF) photonics, which utilize photonic hardware and technologies to process high-bandwidth RF signals, provide speed advantages over electrical signal processing with intrinsic bandwidth limitations [1]. Additionally, they offer other advantages such as reduced loss and a strong immunity to electromagnetic interference. Although the history of RF photonics can be traced back to the 1970s [2], the advancement of RF photonic signal processing has remained highly active in recent years, along with the rapid developments of integrated optics and high-bandwidth RF devices [1,3,4,5,6,7,8,9]. Particularly, the innovative merging of RF photonics and optical microcombs provides extensive opportunities for real-world applications, such as frequency synthesis [10,11,12,13,14,15], filters [16,17,18,19], and neuromorphic computing [20,21,22,23].
Among the various schemes to implement RF photonic processors, RF photonic transversal signal processors have garnered significant interest due to their exceptional reconfigurability, allowing the realization of diverse processing functions without the need for changing any hardware. Previously, processing functions, such as differentiation [24], integration [25], Hilbert transform [26], arbitrary waveform generation [27], filtering [16,17,18], data transmission [28,29,30,31,32,33], and convolutional processing [20,21], have been successfully demonstrated.
In RF photonic transversal signal processors, a number of wavelength channels are needed to facilitate good reconfigurability and ensure a high processing accuracy. In addition, wide channel spacing between these wavelength channels is necessary to ensure a large operational bandwidth. Recent advances in optical microcomb technology provide a competitive solution to meet these requirements by generating a substantial number of widely spaced wavelengths from a micro-scale resonator, together with the added benefits of a significantly reduced device footprint, power consumption, and complexity [34]. In contrast, conventional multi-wavelength sources, such as discrete laser arrays [35,36], fiber Bragg grating arrays [37,38,39], laser frequency combs generated by electro-optic (EO) modulation [40,41,42], and mode-locked fiber lasers [43,44], suffer from limitations in one form or another, such as a limited number of wavelength channels and insufficient channel spacings.
Early implementations of microcomb-based RF photonic transversal signal processors simply replaced conventional multi-wavelength sources with optical microcombs, while retaining all the other components as discrete devices [1,20]. Although this already yields significant benefits, there is much more to be gained by increasing the level of integration for the entire processing system, particularly with respect to the system size, power consumption, and cost. Recently, several processors comprised entirely of integrated components have also been demonstrated [18,45]. Despite being based on the same operation principle, the processors implemented with discrete and integrated components present different processing performances. Previously, we analyzed the processing accuracy of processors implemented by using discrete components [46]. Given the rapid development of integrated processors, it has become crucial to conduct a comparative analysis of the performances of these two forms. This can provide not only important guidance for their selection and assessment in practical applications, but also insights into state-of-the-art and future developments.
In this paper, we provide a comparative study of the performance of microcomb-based RF photonic transversal signal processors implemented with discrete and integrated components. First, we compare the performance of state-of-the-art processors, focusing on the processing accuracy. Next, we conduct an analysis of multiple factors that induce the performance differences, including the tap number and imperfect response of experimental components. Finally, we discuss the potential for future development. These results provide a comprehensive comparison and valuable perspectives for these processors with high reconfigurability for diverse signal processing applications.

2. Microcomb-Based RF Photonic Transversal Signal Processors

RF transversal signal processors are implemented based on the transversal filter structure of digital signal processing, which features a finite impulse response and has been used in applications in a wide range of signal processing functions [1]. Implementing these processors by using RF photonic technology can yield a significantly higher processing bandwidth compared to that of their electronic counterparts [1], and the use of optical microcombs provides a powerful multiwavelength source that is critical for the RF photonic system. Figure 1a illustrates the operation principle of a microcomb-based RF photonic transversal signal processor. The processor employs an optical microcomb as a multiwavelength source, which simultaneously generates numerous wavelength channels as discrete taps. The input RF signal is modulated on each wavelength channel via EO modulation, producing multiple RF replicas. Next, optical spectral shaping is applied to weight these modulated replicas, and a time delay is introduced between adjacent wavelength channels. Finally, the weighted and delayed RF replicas are added together through photodetection to generate the final RF output of the processor. After going through the processing flow in Figure 1a, the output RF signal s(t) can be given as [1].
s ( t ) = ∑ M - 1 n = 0 a n f ( t   –   n Δ T ) ,
where f(t) is the input RF signal, M is the tap number, an (n = 0, 1, 2, …, M – 1) is the tap weight of the nth tap, and ΔT is the time delay between the adjacent wavelength channels. Therefore, the system’s impulse response can be expressed as [1].
h ( t ) = ∑ M - 1 n = 0 a n δ ( t   –   n Δ T ) ,
After the Fourier transformation of Equation (2), the spectral transfer function of the processor can be described as
H ( ω ) = FT   [ h ( t ) ] = ∑ M - 1 n = 0 a n e - j ω n Δ T ,
According to Equations (1)–(3), different processing functions can be realized by appropriately setting the tap coefficients an (n = 0, 1, 2, …, M − 1) without changing the hardware, which allows the high reconfigurability of the processor.
Based on the operation principle in Figure 1a, microcomb-based RF photonic transversal signal processors can be practically implemented into two forms. The first form is illustrated in Figure 1b, where all the components are discrete devices, except for an integrated microring resonator (MRR) used to generate optical microcombs. The second is composed entirely of integrated components, as shown in Figure 1c. To simplify our discussion, we refer to the processors implemented in these two forms as discrete and integrated processors, respectively. Early microcomb-based RF photonic transversal signal processors were developed in the form of discrete processors [1,20], while more recently, some integrated processors have been created [18,45]. Although the operation principle remains consistent across these two forms, their different components result in distinct performances. In the following, we compare their performance in Section 3 and discuss their potential for improvement in Section 4.

3. Performance Comparison of Discrete and Integrated Processors

In this section, we compare the performances of the discrete and integrated processors shown in Figure 1b,c, respectively. Although the size, weight, and power consumption (SWaP) of integrated processors are greatly reduced compared with those of the discrete processors, state-of-the-art integrated processors suffer from limited tap numbers due to the restrictions imposed by the integrated components. Currently, integrated processors with only 8 [18] and 12 taps [45] have been demonstrated, whereas discrete processors have been implemented with up to 80 taps [1,46]. The difference in the tap numbers results in a difference in the processing accuracy [46]. In addition, the imperfect response of experimental components also induces processing errors. These mainly include noise from microcombs, the chirp of the EOM, errors in the delay element, and errors in the spectral shaping module [46].
Table 1 summarizes the parameters of the components in the three processors that we investigated, including a discrete processor (Processor 1) and two integrated processors (Processors 2 and 3). There are two integrated processors: one with the same tap number as that in Ref. [18] and another with an increased tap number to demonstrate the potential for improvement. To characterize the errors induced by imperfect experimental components, optical signal-to-noise ratios (OSNRs), chirp parameters (α), error of the delay element (tv), and random tap coefficient errors (RTCEs) are introduced. The OSNR of microcombs is the ratio of the maximum optical signal to the noise power in each of the comb lines [46]. The chirp parameter is used to characterize the chirp of the EOM, which is caused by asymmetry in the electric field overlap at each electrode and can be expressed as [47].
α = γ 1 + γ 2 γ 1   - γ 2
where γ1 and γ2 are the voltage-to-phase conversion coefficients for the two arms of the modulator. For discrete processors, errors in the delay element are mainly caused by the third-order dispersion of the single-mode fibre, which introduces additional non-uniform time delays between the adjacent wavelength channels. The additional time delay Δt of the nth channel is given as [48].
Δ t   = D 3   l   Δ λ 2 n 2
where D3 is the third-order dispersion parameter, l is the fibre length, and Δλ is the comb spacing. For integrated processors, errors in the delay element are mainly induced by fabrication imperfections and temperature fluctuations [49]. The errors arising from these sources can be measured via characterizing the RF phase response of each wavelength channel [45,48]. Random tap coefficient errors are the difference between the measured power of comb lines and the ideal tap coefficients. These parameters are all set based on the real processors in Refs. [45,46,47,50,51]. For comparison, we assume that the three processors employ the same microcomb with a comb spacing of 0.4 nm (~50 GHz). We also assume that the time delay between adjacent taps in Equation (3) is ∆T = ~33.4 ps to ensure the same operation bandwidth.
Compared with the discrete EOM, the integrated EOM has a relatively high chirp parameter in Table 1, mainly because achieving an accurate bias point and precise electrode placement is more challenging for integrated devices [51]. The lower accuracy for the delay element in the discrete processor is mainly caused by the high-order dispersion of the dispersive medium (e.g., optical fibre), which results in non-uniform time delays between the adjacent wavelength channels [46]. In contrast, in integrated processors, a time delay is introduced by integrated optical delay lines (e.g., Si spiral waveguides), which exhibit a higher accuracy owing to the precise control over the amount of delay achieved by designing a specific length and refractive index profile [52]. The accuracy differences between the integrated and discrete spectral shaping modules are more noticeable. Although the integrated spectral shaping modules have much lower tap numbers, they have lower spectral shaping accuracy compared to that of their discrete counterparts. This is due to the fact that commercial discrete waveshapers based on mature liquid crystal on silicon (LCoS) technology offer much better accuracy for amplitude and phase control, as well as inter-channel synchronization [53].
In our following analysis, three typical signal processing functions, including first-order differentiation (DIF), integration (INT), and Hilbert transform (HT), are taken as examples to compare the accuracy of discrete and integrated processors. The tap numbers required to achieve these processing functions are designed based on our previous work in Ref. [1]. To quantify the comparison of processing accuracy, the root-mean-square error (RMSE) is introduced to compare the deviation between the processors’ outputs and the ideal results, which is expressed as:
RMSE = ∑ k i = 1 ( Y i   –   y i ) 2 k  
where k is the number of sampled points, Y1, Y2, …, Yn are the values of the ideal result, and y1, y2, …, yn are the values of the output of the processors.
Figure 2a–c show the outputs of Processors 1–3 in Table 1 that performed DIF, INT, and HT, respectively. The input RF signal is a Gaussian pulse with a full-width-at-half-maximum (FWHM) of ~0.17 ns. Here, we show the processors’ outputs with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors. The ideal processing results are also shown for comparison. Deviations between the processors’ outputs and the ideal results are observed for all three functions, and the deviations become more significant when taking into account the experimental errors.
Figure 3 compares RMSEs of the processors in Figure 2. The higher processing accuracy of the discrete processor compared to that of the integrated processors is reflected by the lower RMSEs of Processor 1 for all three processing functions. In addition, the RMSEs of Processor 3 are lower compared to those of Processor 2, which indicates a higher processing accuracy achieved by increasing the tap number. According to the results in Figure 3, the primary factor that contributes to the degradation of accuracy for integrated processors is the limited tap number, whereas for discrete processors with a sufficiently large tap number, the processing inaccuracy is mainly induced by the imperfect response of the experimental components. We also note that the differences in RMSEs among Processors 1–3 are more prominent for the INT than they are for the other two processing functions, indicating a higher requirement for a greater number of taps to improve the processing accuracy of INT. In addition, experimental errors have a substantial impact on the RMSEs of DIF, whereas their impact on HT is very small.

4. Potential for Improvement

In this section, we discuss the potential for improvement for both the discrete and integrated processors. In Figure 4a, we quantitively analyze the influence of tap number M on the processing accuracy, where the parameters of the input signal and the processor components are kept the same as those in Figure 2. Similar to that in Figure 2 and Figure 3, we show the results with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors. It is evident that when assuming no experimental errors, both discrete and integrated processors exhibit the same RMSE values at the same M, as they possess identical comb spacing and time delays. When experimental errors are considered, the RMSEs no longer exhibit a monotonic decrease with the tap number M as observed when no experimental errors are assumed. This is because some experimental errors, such as shaping errors in both discrete and integrated processors, as well as the increased errors in the time delay induced by higher-order dispersion in discrete processors, increase with M. The accumulation of these errors outweighs the reduction in RMSEs caused by increasing M, resulting in an overall increase in the RMSEs when M exceeds 10 in Figure 4a.
When considering experimental errors, DIF, INT, and HT require tap numbers of 20, 20, and 80, respectively, to achieve an RMSE of ~0.05, respectively. Although this can be easily achieved with the discrete processor, it is challenging for the state-of-the-art integrated processors due to the significantly increased complexity and degraded processing accuracy for M ≥ 20. The increased complexity results from the increased numbers of MRRs, micro-heaters, and spiral waveguides in Figure 1c. Although integrated processors have the advantage of monotonically integrating a large number of these building blocks, achieving their precise tuning and control can be challenging, especially when dealing with a large number of taps. On the other hand, fabrication errors, additional losses, and thermal drifts in these building blocks degrade the cooperative operation of different wavelength channels, and the processing errors resulting from these factors increase super linearly with the tap number.
Figure 4b compares the output waveforms of both discrete and integrated processors with the same tap number M = 80. Compared to the discrete processor, the experimental errors have more significant influence on the processing accuracy of the integrated processor. Note that we have not taken into account any additional processing errors resulting from factors that may degrade the cooperative operation of different wavelength channels, as discussed earlier. Considering these factors could lead to even larger errors.
To reduce the errors induced by the imperfect response of the experimental components, employing advanced mode-locking approaches [1] to reduce the noise of microcombs could be beneficial for both discrete and integrated processors. For integrated processors, the chirp of silicon EOM can be mitigated by using push–pull configurations, as well as a p-n depletion mode structure [51], and proper methods to calibrate the bias point [45]. The shaping errors of integrated spectral shapers can be alleviated via calibration procedures and gradient descent control [45]. Integrated delay elements introduce additional loss especially when using a waveguide with a high propagation loss, and adiabatic Euler bends can be employed to achieve a reduced loss and low-crosstalk waveguide bends [52]. The use of a wavelength-addressable serial integration scheme can also enable large-scale integration [54]. On the other hand, for discrete processors, there is still room for improving the processing accuracy. Errors in the delay elements induced by higher-order dispersion can be reduced by using programmable phase characteristics of optical spectral shapers (OSSs), and the shaping errors can be minimized through employing feedback control [46].
Environment factors also influence the performance of microcomb-based RF photonic processors. For the MRRs used for generating microcombs, their resonance wavelengths are affected by environment factors such as temperature, thus resulting in the instability of microcombs. The employment of advanced mode-locking technologies has proven to be an effective method to improve the robustness of microcombs, making them less susceptible to environment perturbations [1,3]. Compared to their equivalents implemented using discrete devices, integrated EOMs and spectral shapers are more sensitive to temperature variations. This makes the integrated processors more vulnerable to environmental factors when contrasted with the vulnerability of discrete processors. To minimize the influence of environmental factors in integrated processors, temperature controllers [45] and feedback control [53] have been introduced.

5. Conclusions

In summary, we provide a comparative study of the performances of microcomb-based RF photonic transversal signal processors implemented with discrete and integrated components. We first compared the performances of state-of-the-art processors, especially the processing accuracy. Next, analysis of multiple factors that contribute to the performance differences was conducted, including the tap number and imperfect response of the experimental components. Finally, we discussed the potential for future improvement. Our results show that although current integrated processors are attractive in providing a significantly reduced system size, power consumption, and cost, their processing accuracy is not as high as that of the discrete processors. In addition, there is still room for improvement for both the discrete and integrated processors. The results offer valuable insights for microcomb-based RF photonic transversal signal processors with high reconfigurability for diverse applications.

Author Contributions

Conceptualization, methodology, Y.S. and J.W.; writing—original draft preparation, Y.S., J.W. and Y.L.; writing—review and editing, J.W. and D.J.M.; supervision, J.W. and D.J.M. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported in part by the funded ARC Centre of Excellence in Optical Microcombs for Breakthrough Science (grant number CE230100006), in part by the Australian Research Council Discovery Projects Programs (grant numbers DP150104327, DP190102773, and DP190101576), and in part by the Swinburne ECR-SUPRA program.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Sun, Y.; Wu, J.; Tan, M.; Xu, X.; Li, Y.; Morandotti, R.; Mitchell, A.; Moss, D.J. Applications of optical microcombs. Adv. Opt. Photon. 2023, 15, 86–175. [Google Scholar] [CrossRef]
  2. Seeds, A.J. Microwave photonics. IEEE Trans. Microw. Theory Tech. 2002, 50, 877–887. [Google Scholar] [CrossRef]
  3. Wu, J.; Xu, X.; Nguyen, T.G.; Chu, S.T.; Little, B.E.; Morandotti, R.; Mitchell, A.; Moss, D.J. RF photonics: An optical microcombs’ perspective. IEEE J. Sel. Top. Quantum Electron. 2018, 24, 1–20. [Google Scholar] [CrossRef]
  4. Marpaung, D.; Yao, J.; Capmany, J. Integrated microwave photonics. Nat. Photonics 2019, 13, 80–90. [Google Scholar] [CrossRef]
  5. Capmany, J.; Novak, D. Microwave photonics combines two worlds. Nat. Photonics 2007, 1, 319–330. [Google Scholar] [CrossRef]
  6. Torres-Company, V.; Weiner, A.M. Optical frequency comb technology for ultra-broadband radio-frequency photonics. Laser Photonics Rev. 2013, 8, 368–393. [Google Scholar] [CrossRef]
  7. Wang, W.; Wang, L.; Zhang, W. Advances in soliton microcomb generation. Adv. Photonics 2020, 2, 034001. [Google Scholar] [CrossRef]
  8. Kippenberg, T.J.; Gaeta, A.L.; Lipson, M.; Gorodetsky, M.L. Dissipative Kerr solitons in optical microresonators. Science 2018, 361, eaan8083. [Google Scholar] [CrossRef]
  9. Yao, J. Microwave Photonics. J. Light. Technol. 2009, 27, 314–335. [Google Scholar] [CrossRef]
  10. Spencer, D.T.; Drake, T.; Briles, T.C.; Stone, J.; Sinclair, L.C.; Fredrick, C.; Li, Q.; Westly, D.; Ilic, B.R.; Bluestone, A.; et al. An optical-frequency synthesizer using integrated photonics. Nature 2018, 557, 81–85. [Google Scholar] [CrossRef]
  11. Liu, J.; Lucas, E.; Raja, A.S.; He, J.; Riemensberger, J.; Wang, R.N.; Karpov, M.; Guo, H.; Bouchand, R.; Kippenberg, T.J. Photonic microwave generation in the X- and K-band using integrated soliton microcombs. Nat. Photonics 2020, 14, 486–491. [Google Scholar] [CrossRef]
  12. Lucas, E.; Brochard, P.; Bouchand, R.; Schilt, S.; Südmeyer, T.; Kippenberg, T.J. Ultralow-noise photonic microwave synthesis using a soliton microcomb-based transfer oscillator. Nat. Commun. 2020, 11, 374. [Google Scholar] [CrossRef] [PubMed]
  13. Weng, W.; Lucas, E.; Lihachev, G.; Lobanov, V.E.; Guo, H.; Gorodetsky, M.L.; Kippenberg, T.J. Spectral Purification of Microwave Signals with Disciplined Dissipative Kerr Solitons. Phys. Rev. Lett. 2019, 122, 6. [Google Scholar] [CrossRef] [PubMed]
  14. Weng, W.; Kaszubowska-Anandarajah, A.; Liu, J.; Anandarajah, P.M.; Kippenberg, T.J. Frequency division using a soliton-injected semiconductor gain-switched frequency comb. Sci. Adv. 2020, 6, 6. [Google Scholar] [CrossRef]
  15. Wang, B.; Morgan, J.S.; Sun, K.; Jahanbozorgi, M.; Yang, Z.; Woodson, M.; Estrella, S.; Beling, A.; Yi, X. Towards high-power, high-coherence, integrated photonic mmWave platform with microcavity solitons. Light: Sci. Appl. 2021, 10, 4. [Google Scholar] [CrossRef] [PubMed]
  16. Xu, X.; Tan, M.; Wu, J.; Nguyen, T.G.; Chu, S.T.; Little, B.E.; Morandotti, R.; Mitchell, A.; Moss, D.J. Advanced Adaptive Photonic RF Filters with 80 Taps Based on an Integrated Optical Micro-Comb Source. J. Light. Technol. 2019, 37, 1288–1295. [Google Scholar] [CrossRef]
  17. Hu, J.; He, J.; Liu, J.; Raja, A.S.; Karpov, M.; Lukashchuk, A.; Kippenberg, T.J.; Brès, C.-S. Reconfigurable radiofrequency filters based on versatile soliton microcombs. Nat. Commun. 2020, 11, 4377. [Google Scholar] [CrossRef]
  18. Shu, H.; Chang, L.; Tao, Y.; Shen, B.; Xie, W.; Jin, M.; Netherton, A.; Tao, Z.; Zhang, X.; Chen, R.; et al. Microcomb-driven silicon photonic systems. Nature 2022, 605, 457–463. [Google Scholar] [CrossRef]
  19. Liu, Y.; Choudhary, A.; Marpaung, D.; Eggleton, B.J. Integrated microwave photonic filters. Adv. Opt. Photonics 2020, 12, 485–555. [Google Scholar] [CrossRef]
  20. Xu, X.; Tan, M.; Corcoran, B.; Wu, J.; Boes, A.; Nguyen, T.G.; Chu, S.T.; Little, B.E.; Hicks, D.G.; Morandotti, R.; et al. 11 TOPS photonic convolutional accelerator for optical neural networks. Nature 2021, 589, 44–51. [Google Scholar] [CrossRef]
  21. Feldmann, J.; Youngblood, N.; Karpov, M.; Gehring, H.; Li, X.; Stappers, M.; Le Gallo, M.; Fu, X.; Lukashchuk, A.; Raja, A.S.; et al. Parallel convolutional processing using an integrated photonic tensor core. Nature 2021, 589, 52–58. [Google Scholar] [CrossRef] [PubMed]
  22. Shastri, B.J.; Tait, A.N.; de Lima, T.F.; Pernice, W.H.P.; Bhaskaran, H.; Wright, C.D.; Prucnal, P.R. Photonics for artificial intelligence and neuromorphic computing. Nat. Photonics 2021, 15, 102–114. [Google Scholar] [CrossRef]
  23. Ma, W.; Liu, Z.; Kudyshev, Z.A.; Boltasseva, A.; Cai, W.; Liu, Y. Deep learning for the design of photonic structures. Nat. Photonics 2020, 15, 77–90. [Google Scholar] [CrossRef]
  24. Xu, X.; Wu, J.; Shoeiby, M.; Nguyen, T.G.; Chu, S.T.; Little, B.E.; Morandotti, R.; Mitchell, A.; Moss, D.J. Reconfigurable broadband microwave photonic intensity differentiator based on an integrated optical frequency comb source. APL Photonics 2017, 2, 096104. [Google Scholar] [CrossRef]
  25. Xu, X.; Tan, M.; Wu, J.; Boes, A.; Corcoran, B.; Nguyen, T.G.; Chu, S.T.; Little, B.E.; Morandotti, R.; Mitchell, A.; et al. Photonic RF and Microwave Integrator Based on a Transversal Filter With Soliton Crystal Microcombs. IEEE Trans. Circuits Syst. II: Express Briefs 2020, 67, 3582–3586. [Google Scholar] [CrossRef]
  26. Tan, M.; Mitchell, A.; Moss, D.J.; Xu, X.; Corcoran, B.; Wu, J.; Boes, A.; Nguyen, T.G.; Chu, S.T.; Little, B.E.; et al. Microwave and RF Photonic Fractional Hilbert Transformer Based on a 50 GHz Kerr Micro-Comb. J. Light. Technol. 2019, 37, 6097–6104. [Google Scholar] [CrossRef]
  27. Wang, B.; Yang, Z.; Sun, S.; Yi, X. Radio-frequency line-by-line Fourier synthesis based on optical soliton microcombs. Photonics Res. 2022, 10, 932–938. [Google Scholar] [CrossRef]
  28. Corcoran, B.; Tan, M.; Xu, X.; Boes, A.; Wu, J.; Nguyen, T.G.; Chu, S.T.; Little, B.E.; Morandotti, R.; Mitchell, A.; et al. Ultra-dense optical data transmission over standard fibre with a single chip source. Nat. Commun. 2020, 11, 7. [Google Scholar] [CrossRef]
  29. Pfeifle, J.; Brasch, V.; Lauermann, M.; Yu, Y.; Wegner, D.; Herr, T.; Hartinger, K.; Schindler, P.; Li, J.; Hillerkuss, D.; et al. Coherent terabit communications with microresonator Kerr frequency combs. Nat. Photonics 2014, 8, 375–380. [Google Scholar] [CrossRef]
  30. Marin-Palomo, P.; Kemal, J.N.; Karpov, M.; Kordts, A.; Pfeifle, J.; Pfeiffer, M.H.P.; Trocha, P.; Wolf, S.; Brasch, V.; Anderson, M.H.; et al. Microresonator-based solitons for massively parallel coherent optical communications. Nature 2017, 546, 274–279. [Google Scholar] [CrossRef]
  31. Fülöp, A.; Mazur, M.; Lorences-Riesgo, A.; Helgason, B.; Wang, P.-H.; Xuan, Y.; Leaird, D.E.; Qi, M.; Andrekson, P.A.; Weiner, A.M.; et al. High-order coherent communications using mode-locked dark-pulse Kerr combs from microresonators. Nat. Commun. 2018, 9, 1–8. [Google Scholar] [CrossRef] [PubMed]
  32. Fujii, S.; Tanaka, S.; Ohtsuka, T.; Kogure, S.; Wada, K.; Kumazaki, H.; Tasaka, S.; Hashimoto, Y.; Kobayashi, Y.; Araki, T.; et al. Dissipative Kerr soliton microcombs for FEC-free optical communications over 100 channels. Opt. Express 2022, 30, 1351–1364. [Google Scholar] [CrossRef] [PubMed]
  33. Geng, Y.; Zhou, H.; Han, X.; Cui, W.; Zhang, Q.; Liu, B.; Deng, G.; Zhou, Q.; Qiu, K. Coherent optical communications using coherence-cloned Kerr soliton microcombs. Nat. Commun. 2022, 13, 1070. [Google Scholar] [CrossRef]
  34. Chang, L.; Liu, S.; Bowers, J.E. Integrated optical frequency comb technologies. Nat. Photonics 2022, 16, 95–108. [Google Scholar] [CrossRef]
  35. Zhang, J.; Yao, J. Photonic True-Time Delay Beamforming Using a Switch-Controlled Wavelength-Dependent Recirculating Loop. J. Light. Technol. 2016, 34, 3923–3929. [Google Scholar] [CrossRef]
  36. Sagues, M.; Olcina, R.G.; Loayssa, A.; Sales, S.; Capmany, J. Multi-tap complex-coefficient incoherent microwave photonic filters based on optical single-sideband modulation and narrow band optical filtering. Opt. Express 2008, 16, 295–303. [Google Scholar] [CrossRef]
  37. Yu, G.; Zhang, W.; Williams, J. High-performance microwave transversal filter using fiber Bragg grating arrays. IEEE Photonics Technol. Lett. 2000, 12, 1183–1185. [Google Scholar] [CrossRef]
  38. Mansoori, S.; Mitchell, A. RF transversal filter using an AOTF. IEEE Photonics Technol. Lett. 2004, 16, 879–881. [Google Scholar] [CrossRef]
  39. Zhang, W.; Bennion, I.; Williams, J.A.R. Chromatic dispersion effect in a microwave photonic filter using superstructured fiber Bragg grating and dispersive fiber. Opt. Express 2005, 13, 6429–6437. [Google Scholar] [CrossRef]
  40. Supradeepa, V.R.; Long, C.M.; Wu, R.; Ferdous, F.; Hamidi, E.; Leaird, D.E.; Weiner, A.M. Comb-based radiofrequency photonic filters with rapid tunability and high selectivity. Nat. Photonics 2012, 6, 186–194. [Google Scholar] [CrossRef]
  41. Hamidi, E.; Leaird, D.E.; Weiner, A.M. Tunable Programmable Microwave Photonic Filters Based on an Optical Frequency Comb. IEEE Trans. Microw. Theory Tech. 2010, 58, 3269–3278. [Google Scholar] [CrossRef]
  42. Metcalf, A.J.; Kim, H.-J.; Leaird, D.E.; Jaramillo-Villegas, J.A.; McKinzie, K.A.; Lal, V.; Hosseini, A.; Hoefler, G.E.; Kish, F.; Weiner, A.M. Integrated line-by-line optical pulse shaper for high-fidelity and rapidly reconfigurable RF-filtering. Opt. Express 2016, 24, 23925–23940. [Google Scholar] [CrossRef]
  43. Ortigosa-Blanch, A.; Mora, J.; Capmany, J.; Ortega, B.; Pastor, D. Tunable radio-frequency photonic filter based on an actively mode-locked fiber laser. Opt. Lett. 2006, 31, 709–711. [Google Scholar] [CrossRef]
  44. Maram, R.; Onori, D.; Azaña, J.; Chen, L.R. Discretely programmable microwave photonic filter based on temporal Talbot effects. Opt. Express 2019, 27, 14381–14391. [Google Scholar] [CrossRef]
  45. Bai, B.; Yang, Q.; Shu, H.; Chang, L.; Yang, F.; Shen, B.; Tao, Z.; Wang, J.; Xu, S.; Xie, W.; et al. Microcomb-based integrated photonic processing unit. Nat. Commun. 2023, 14, 66. [Google Scholar] [CrossRef]
  46. Sun, Y.; Wu, J.; Li, Y.; Tan, M.; Xu, X.; Chu, S.T.; Little, B.E.; Morandotti, R.; Mitchell, A.; Moss, D.J. Quantifying the Accuracy of Microcomb-Based Photonic RF Transversal Signal Processors. IEEE J. Sel. Top. Quantum Electron. 2023, 29, 1–17. [Google Scholar] [CrossRef]
  47. Iii, C.E.R.; Carini, J.L.; Pechkis, J.A.; Gould, P.L. Characterization and compensation of the residual chirp in a Mach-Zehnder-type electro-optical intensity modulator. Opt. Express 2010, 18, 1166–1176. [Google Scholar]
  48. Xu, X.; Wu, J.; Nguyen, T.G.; Moein, T.; Chu, S.T.; Little, B.E.; Morandotti, R.; Mitchell, A.; Moss, D.J. Photonic microwave true time delays for phased array antennas using a 49 GHz FSR integrated optical micro-comb source [Invited]. Photonics Res. 2018, 6, B30–B36. [Google Scholar] [CrossRef]
  49. Ji, X.; Yao, X.; Gan, Y.; Mohanty, A.; Tadayon, M.A.; Hendon, C.P.; Lipson, M. On-chip tunable photonic delay line. APL Phontonics 2019, 4, 090803. [Google Scholar] [CrossRef]
  50. Tait, A.N.; de Lima, T.F.; Nahmias, M.A.; Shastri, B.J.; Prucnal, P.R. Continuous Calibration of Microring Weights for Analog Optical Networks. IEEE Photonics Technol. Lett. 2016, 28, 887–890. [Google Scholar] [CrossRef]
  51. Sinatkas, G.; Christopoulos, T.; Tsilipakos, O.; Kriezis, E.E. Electro-optic modulation in integrated photonics. J. Appl. Phys. 2021, 130, 010901. [Google Scholar] [CrossRef]
  52. Ji, X.; Liu, J.; He, J.; Wang, R.N.; Qiu, Z.; Riemensberger, J.; Kippenberg, T.J. Compact, spatial-mode-interaction-free, ultralow-loss, nonlinear photonic integrated circuits. Commun. Phys. 2022, 5, 84. [Google Scholar] [CrossRef]
  53. Tait, A.N.; Jayatilleka, H.; De Lima, T.F.; Ma, P.Y.; Nahmias, M.A.; Shastri, B.J.; Shekhar, S.; Chrostowski, L.; Prucnal, P.R. Feedback control for microring weight banks. Opt. Express 2018, 26, 26422–26443. [Google Scholar] [CrossRef]
  54. Kuramochi, E.; Nozaki, K.; Shinya, A.; Takeda, K.; Sato, T.; Matsuo, S.; Taniyama, H.; Sumikura, H.; Notomi, M. Large-scale integration of wavelength-addressable all-optical memories on a photonic crystal chip. Nat. Photonics 2014, 8, 474–481. [Google Scholar] [CrossRef]
Micromachines 14 01794 g001
Figure 1. (a) Schematic illustration of the operation principle of a microcomb-based RF photonic transversal signal processor. (b) Schematic of a microcomb-based RF photonic transversal signal processor implemented by discrete components. (c) Schematic of an on-chip microcomb-based RF photonic transversal signal processor implemented with integrated components. EOM: electro-optic modulator. RF: radio frequency. PD: photodetector. CW laser: continuous-wave laser. EDFA: erbium-doped fibre amplifier. PC: polarization controller. MRR: microring resonator. SMF: single-mode fibre. OSS: optical spectral shaper. BPD: balanced photodetector. BPD: balanced photodetector. ART: anti-reflection termination.
Figure 1. (a) Schematic illustration of the operation principle of a microcomb-based RF photonic transversal signal processor. (b) Schematic of a microcomb-based RF photonic transversal signal processor implemented by discrete components. (c) Schematic of an on-chip microcomb-based RF photonic transversal signal processor implemented with integrated components. EOM: electro-optic modulator. RF: radio frequency. PD: photodetector. CW laser: continuous-wave laser. EDFA: erbium-doped fibre amplifier. PC: polarization controller. MRR: microring resonator. SMF: single-mode fibre. OSS: optical spectral shaper. BPD: balanced photodetector. BPD: balanced photodetector. ART: anti-reflection termination.
Micromachines 14 01794 g001
Micromachines 14 01794 g002
Figure 2. Temporal waveform of Gaussian input pulse and output waveforms from (a) Processor 1, (b) Processor 2, and (c) Processor 3 that performed (i) differentiation (DIF), (ii) integration (INT), and (iii) Hilbert transform (HT). The processors’ outputs with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors are shown, together with the ideal processing result for comparison.
Figure 2. Temporal waveform of Gaussian input pulse and output waveforms from (a) Processor 1, (b) Processor 2, and (c) Processor 3 that performed (i) differentiation (DIF), (ii) integration (INT), and (iii) Hilbert transform (HT). The processors’ outputs with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors are shown, together with the ideal processing result for comparison.
Micromachines 14 01794 g002
Micromachines 14 01794 g003
Figure 3. Comparison of root-mean-square errors (RMSEs) of Processors 1–3 that performed (a) DIF, (b) INT, and (c) HT. The RMSEs of processors’ outputs with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors are shown.
Figure 3. Comparison of root-mean-square errors (RMSEs) of Processors 1–3 that performed (a) DIF, (b) INT, and (c) HT. The RMSEs of processors’ outputs with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors are shown.
Micromachines 14 01794 g003
Micromachines 14 01794 g004
Figure 4. Influence of the tap number and experimental errors on processing accuracy of the discrete and integrated processors. (a) RMSEs of processors that performed (i) DIF, (ii) INT, and (iii) HT as a function of tap number M. The RMSEs of processors’ outputs with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors are shown. (b) Temporal waveform of Gaussian input pulse and output waveforms of discrete and integrated processors that performed (i) DIF, (ii) INT, and (iii) HT. The processors’ outputs with errors induced by (1) limited tap number and (2) both the limited tap number and experimental errors are shown, together with the ideal processing result for comparison.
Figure 4. Influence of the tap number and experimental errors on processing accuracy of the discrete and integrated processors. (a) RMSEs of processors that performed (i) DIF, (ii) INT, and (iii) HT as a function of tap number M. The RMSEs of processors’ outputs with errors induced by (1) only limited tap numbers and (2) both limited tap numbers and experimental errors are shown. (b) Temporal waveform of Gaussian input pulse and output waveforms of discrete and integrated processors that performed (i) DIF, (ii) INT, and (iii) HT. The processors’ outputs with errors induced by (1) limited tap number and (2) both the limited tap number and experimental errors are shown, together with the ideal processing result for comparison.
Micromachines 14 01794 g004
Table 1. Comparison of components’ parameters in discrete and integrated processors.
Table 1. Comparison of components’ parameters in discrete and integrated processors.
Discrete
processor		No.Tap No.OSNR of microcombsChirp parameter of the EOMErrors of the delay elementRTCE of the spectral shaping module
1M = 80 [46]OSNR: 20 dB [46]α: 0.1 [51]tv: 4% [46]RTCE: 5% [46]
Integrated
processors		No.Tap No.OSNR of microcombsChirp parameter of the EOMError of the delay elementRTCE of the spectral shaping module
2M = 8 [18]OSNR: 20 dB [46]α: 0.8 [47]tv: 3% [45]RTCE: 9% [50]
No.Tap No.OSNR of microcombsChirp parameter of the EOMError of the delay elementRTCE of the spectral shaping module
3M = 20OSNR: 20 dB [46]α: 0.8 [47]tv: 3% [45]RTCE: 9% [50]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Sun, Y.; Wu, J.; Li, Y.; Moss, D.J. Comparison of Microcomb-Based Radio-Frequency Photonic Transversal Signal Processors Implemented with Discrete Components Versus Integrated Chips. Micromachines 2023, 14, 1794. https://doi.org/10.3390/mi14091794

AMA Style

Sun Y, Wu J, Li Y, Moss DJ. Comparison of Microcomb-Based Radio-Frequency Photonic Transversal Signal Processors Implemented with Discrete Components Versus Integrated Chips. Micromachines. 2023; 14(9):1794. https://doi.org/10.3390/mi14091794

Chicago/Turabian Style

Sun, Yang, Jiayang Wu, Yang Li, and David J. Moss. 2023. "Comparison of Microcomb-Based Radio-Frequency Photonic Transversal Signal Processors Implemented with Discrete Components Versus Integrated Chips" Micromachines 14, no. 9: 1794. https://doi.org/10.3390/mi14091794

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop