Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Frequency-selectable microwave generation based on on-chip switchable spectral shaping and wavelength-to-time mapping

Open Access Open Access

Abstract

We propose and experimentally demonstrate a scheme for the photonic generation of pulsed microwave signals with selectable frequency based on spectral shaping and wavelength-to-time mapping (WTTM) technique. The frequency selectivity is realized by channel switching on an integrated silicon-on-insulator (SOI) spectral shaping chip. The incident signal is spectrally shaped by the asymmetric Mach-Zehnder interferometer (MZI) in the selected channel, and an optical spectrum with uniform free spectral range (FSR) can be generated in a broad bandwidth up to dozens of nanometers, implying large microwave signal duration after WTTM if a pulse light source with matched bandwidth is available. Microwave pulses of frequency from 3.6 GHz to 28.4 GHz with a fixed interval are experimentally generated respectively. The realization of eight microwave frequencies selectable with only one shared dispersive element (DE) required indicates high expansibility in the frequency cover range of our scheme by tuning the dispersion value in WTTM.

© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement

1. Introduction

Microwave waveforms have numerous applications in fields like communications, radar systems, sensing, and imaging [14]. Compared with conventional electrical domain microwave waveform generation [5,6], photonic methods appear more attractive for higher frequency and processing speed, broader bandwidth, and stronger stability [7]. One frequently used approach to photonically generating microwave signals is external modulation. This method is usually based on manipulation of the driving signals applied to the modulators, for example, frequency up-conversion [8,9] and frequency shift [1012]. The usage of mature commercial modulators along with other all-fiber devices ensures the stability and precision of the system, while restraining the possibility of integration. Another widely selected method for the photonic generation of microwave waveforms is direct space-to-time (DST) pulse shaping. As proposed, the DST pulse shaping technique for microwave generation develops from spatial light modulation structure [13,14] to on-chip configurations [15,16]. DST method is advantageous especially in high-frequency microwave generation, for instance, up to 100 GHz. However, when it comes to another frequently used frequency band such as centimeter wave, the size of the system will dramatically increase due to the requirement for larger propagation length difference between adjacent pulses. In addition, the power utilization in DST implementation is relatively low. In comparison, microwave generation schemes based on wavelength-to-time mapping (WTTM) technique have drawn more attention in recent years for their flexibility and simplicity [17,18]. A typical WTTM waveform generation system incorporates a spectral shaper and a dispersive element (DE). The shaped spectrum of a broadband ultrashort pulse is mapped to the temporal domain in the DE. Various methods have been proposed for pulse shaping, such as polarization interference [1922], unbalanced interference [23,24], and spectral sampling [25]. Except for the merit of being modulator-free, an increasing number of on-chip spectral shapers are proposed and demonstrated [2528], indicating the integration potential for WTTM technique-based microwave waveform generation systems.

The main problem existing in on-chip spectral shaping now is the lack of tunability or reconfigurability. Once the parameters are set, the free spectral range (FSR) of the shaped spectrum is hard to vary, especially in a broad bandwidth. In this paper, we propose a switch-controlled multi-channel spectral shaping chip for the generation of frequency-selectable microwave signals based on the WTTM technique. The input pulses can be guided to a specified channel for spectral shaping by applying appropriate voltages to the heaters. Shaped optical spectrum with uniform FSR can be generated by the asymmetric Mach-Zehnder interferometer (MZI) in a significantly broader bandwidth compared with microring resonator or Bragg grating-based spectral shapers, which is instrumental in realizing longer signal duration. Moreover, our configuration can be viewed to have discrete tunability in microwave frequency, and the frequency cover range can be largely expanded if a tunable DE is available.

2. Principle and design

The principle of the microwave pulse generation method based on spectral shaping followed by WTTM is illustrated in Fig. 1. An ultrashort pulse with broadband spectrum is firstly shaped in the optical domain. Then a DE is used to map the profile of the optical signal to the time domain. Corresponding microwave signals can thus be generated after optical-to-electrical conversion.

 figure: Fig. 1.

Fig. 1. Schematic of a microwave waveform generation system based on spectral shaping and WTTM technique.

Download Full Size | PDF

Based on the WTTM waveform generation theory, a silicon-on-insulator (SOI) spectral shaping chip with switchable FSR is designed, and the schematic diagram is depicted in Fig. 2. The basic element for pulse shaping is the 50:50 multimode interferometer (MMI) based asymmetric MZI. Periodic notch dips can be realized in the optical domain by the asymmetric MZI, and the FSR of the asymmetric MZI spectrum can be expressed as:

$$FSR = \frac{{{\lambda ^2}}}{{{n_g}\Delta L}}$$
where λ, ng, and ΔL represent the wavelength, the waveguide group index, and the length difference between the two arms in the asymmetric MZI, respectively. The operation wavelength in our design is around 1555 nm, where the FSR of the asymmetric MZI spectrum can be seen as constant in the range of dozens of nanometers. The group index of the standard 500 nm × 220 nm single-mode silicon waveguide at such a wavelength is calculated to be 4.2. Applying these parameters, a series of asymmetric MZIs with ΔL increasing from 460 µm to 3680 µm by equal difference are designed, and the corresponding FSRs are supposed to be 1.25 nm to 1/8 of 1.25 nm. A 3-stage binary tree-like optical switch network is connected to the asymmetric MZIs for channel selection. Each switch is based on the conventional symmetric MZI structure, with a heater on one arm for thermo-optical tuning.

 figure: Fig. 2.

Fig. 2. Schematic of the FSR switchable spectral shaping chip based on the SOI platform.

Download Full Size | PDF

In the WTTM step, DE with linear dispersion is selected in our design, thus the generated microwave can be approximated as [18]:

$$y(t )\propto T{(\lambda )_{\lambda = t/\ddot{\varPhi}}}$$
where $\ddot{\varPhi}$ is the dispersion coefficient of the DE, and T(λ) is the shaped optical spectrum. Therefore, the frequency of the generated microwave can be expressed as:
$$f = \frac{1}{{|\ddot{\varPhi}|\cdot FSR}}$$

Considering our configuration, substituting the FSR expression of the asymmetric MZI in Eq. (3), the frequency expression can be rewritten as:

$$f = \frac{{{n_g}\Delta L}}{|\ddot{\varPhi}| {\lambda ^2}}$$

3. Experiment

The spectral shaping chip is fabricated by Applied Nanotools Inc. using an SOI wafer with a 220 nm thick top silicon and a 2 µm thick buried thermal oxide. The waveguides are patterned by electron beam lithography and then fully etched. A 2.2 µm thick oxide cladding layer is deposited over the waveguides by plasma-enhanced chemical vapor deposition (PECVD), with a 200 nm thick TiW alloy heater placed above. A TiW/Al routing bilayer is located between the heaters and the oxide windows for wire bonding. Deep trenches are etched on the edges to polish the side surfaces of the chip and reduce the loss of edge coupling. Microscope images of a single optical switch network and the FSR-selectable spectral shaping structure with a switch network connected to the asymmetric MZIs are shown in Fig. 3(a) and (b), respectively. The single optical switch network is fabricated on a same wafer for precise characterization of its performance. The ports labeled LT and RT in Fig. 3 represent the testing waveguides, while LCh and RCh are the selectable channels. Switches are labeled LS and RS. The input and output ports are distributed as an array with a 250 µm interval. The testing waveguides are designed to download 1% of the optical power for the measurement of switches at the first and second stages. The waveguides are tapered to the width of 164 nm at the edge of the chip for low-loss coupling with the 4.0 µm mode field diameter (4.0 MFD) fiber array. The measured coupling loss after packaging is about 4 dB/facet.

 figure: Fig. 3.

Fig. 3. Microscope images of (a) the single switch network for test and (b) the FSR-selectable spectral shaping structure. Yellow labels represent input or output ports on the chip, while white labels represent the optical switches.

Download Full Size | PDF

3.1 Switch network

The performance of the single optical switch network is first verified. The input optical wavelength is set to 1555 nm with a power of 10 dBm. The switching voltage is swept by stage at an interval of 0.02 V, with a multi-channel programmable voltage source (PVS) as the power supply. The varying output power is recorded by an optical power meter. As an example, the switching voltage scanning results of the optical switches related to the first channel at each stage are exhibited in Fig. 4. The extinction ratio between the “on” and “off” state for a single switch can reach 26.47 dB, and can improve to 44.70 dB after three stages. The complete switching voltage sweeping results are listed in Table 1. The switch numbers correspond to the labels in Fig. 3, and the switching voltage represents the larger of the “on” voltage and the “off” voltage. Every heater in our design has the same footprint of 170 µm × 3 µm. The heaters closer to the chip edge show larger resistance due to the non-uniformity during deposition, which makes these heaters thinner than those near the chip center. Fortunately, such fabrication error does not make the heaters more likely to burnout, and the power consumption of each switch does not appear to make much difference. Since only 3 optical switches are required to operate simultaneously in any operation mode, the power consumption of the whole optical switch network can be estimated to be 118.36 mW, which is the sum of the largest switching power consumption at each stage.

 figure: Fig. 4.

Fig. 4. Optical power measured while sweeping the switching voltage of (a) Switch LS1 at LT1 port; (b) Switch LS3 at LT2 port when the voltage applied on LS1 is fixed at 2.58 V; (c) Switch LS7 at LCh1 port when the voltage applied on LS1 and LS3 are fixed at 2.58 V and 3.98 V, respectively.

Download Full Size | PDF

Tables Icon

Table 1. Characteristics of each switch in the single optical switch network

The crosstalk characteristic of the optical switch network is also verified. Figure 5 exhibits the optical spectrum of each channel in the wavelength range from 1550 nm to 1560 nm when a certain channel is tuned to the “on” state. The results demonstrate that every channel in the optical switch network can realize a crosstalk smaller than −20 dB over a bandwidth of 10 nm while operating, and the power non-uniformity is less than 2 dB in the wavelength range.

 figure: Fig. 5.

Fig. 5. (a)-(h): Transmission spectrum of all channels when LCh1-LCh8 is tuned to the “on” state.

Download Full Size | PDF

The spectra of the asymmetric MZIs connected to the optical switch network are measured then. The results in Fig. 6 display the optical characteristics of each asymmetric MZI when the corresponding switching channel is tuned to the “on” state. A series of optical spectra with FSR varying from 1.26 nm to 0.157 nm are realized, matching accurately with the design values. Considering the coupling loss and the loss introduced by the asymmetric MZIs, the switch network caused loss is very small, and the extinction ratio is maintained at the level of over 20 dB, ensuring adequate signal intensity of the generated microwaves after WTTM.

 figure: Fig. 6.

Fig. 6. (a)-(h): Measured transmission spectra of the asymmetric MZIs in RCh1-RCh8 when the corresponding switching channel is tuned to the on-state. The FSR of the spectrum is labeled in each subfigure.

Download Full Size | PDF

3.2 Frequency selectable microwave generation

The microwave generation experiment has been conducted finally. The experimental setup of the whole system is illustrated in Fig. 7. A short temporal pulse with a broad optical bandwidth generated by a mode-locked femtosecond pulse laser (MLFPL) transmits through a polarization controller (PC) to the spectral shaping chip. The DC voltage applied to the heaters on the chip is supplied by a PVS with eight pairs of current lead. A temperature control (TEC) module is employed to make the chip operate in a thermo-stable state, and our experiment temperature is set to 25.5°C. The optical spectrum of the input pulse is shaped in a switch-selected channel and then propagates to an 8 × 1 optical switch placed at the output of the spectral shaping chip. This switch is applied to ensure that the shaped signal from any selected channel can be guided to a shared output port, thus only one DE is required for WTTM. The pulse is amplified by an erbium-doped fiber amplifier (EDFA) before mapping in the DE, and eventually sent to a photodetector (PD) for optoelectrical conversion. A spectrum analyzer is used to obtain the frequency spectrum of the generated microwave.

 figure: Fig. 7.

Fig. 7. Experimental setup of the frequency selectable microwave waveform generation system. MLFPL: mode-locked femtosecond pulse laser, PC: polarization controller, PVS: programmable voltage source, TEC: temperature control, EDFA: erbium-doped fiber amplifier, PD: photodetector.

Download Full Size | PDF

The DE employed in our experiment is a 1411 m long dispersion fiber with a labeled dispersion value of −200 ps/nm. According to the analysis in the previous chapter, the frequencies of the microwave pulses that can be generated via channels RCh1 to RCh8 are calculated to be 4 GHz to 32 GHz, with a 4 GHz interval. The measured results from the spectrum analyzer are shown in Fig. 8. By switching the channel, the frequency of the generated microwave signals increases from 3.6 GHz to 28.4 GHz with an average interval of 3.5 GHz, presenting a red shift compared with the design values. This phenomenon is caused by the deviation of dispersion value in the dispersive fiber. If dispersion of standard −200 ps/nm is available, the frequencies of the generated microwave waveforms will accurately match our design values. It is worth noting that these signals of different frequencies can be generated simultaneously in any combination by tuning the voltage on the heaters. Moreover, other frequency series can also be obtained if applying different dispersion values.

 figure: Fig. 8.

Fig. 8. (a)-(h): Measured frequency spectra of microwave signals generated via RCh1 to RCh8.

Download Full Size | PDF

4. Conclusion

In conclusion, we have proposed and experimentally demonstrated a microwave signal generation scheme based on a spectral shaping chip and WTTM technique. Asymmetric MZIs with various length differences placed in separate channels are able to generate broadband optical spectra with different FSRs, so that the FSR of the shaped signal transmitted to the DE is selectable by the application of the 1 × 8 optical switch network on the chip, and microwave pulses of eight different frequencies can be provided for selection with only one shared DE needed. In our experiment, microwave signals with a frequency from 3.6 GHz to 28.4 GHz at an average interval of 3.5 GHz are successfully generated by channel switching. The footprint of the FSR-selectable spectral shaping chip is within 12 mm2, and the largest power consumption is 118.36 mW. The frequency coverage from the C band to the K band indicates that our configuration has application potential in radar systems, satellite communication, radio navigation, etc.

Funding

National Key Research and Development Program of China (2018YFB2201903); National Natural Science Foundation of China (62075038).

Disclosures

The authors declare no conflicts of interest.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. J. Capmany and D. Novak, “Microwave photonics combines two worlds,” Nat. Photonics 1(6), 319–330 (2007). [CrossRef]  

2. J. Capmany, G. Li, C. Lim, and J. Yao, “Microwave photonics: current challenges towards widespread application,” Opt. Express 21(19), 22862–22867 (2013). [CrossRef]  

3. J. Yao, “Microwave photonics,” J. Lightwave Technol. 27(3), 314–335 (2009). [CrossRef]  

4. L. R. Chen, “Silicon photonics for microwave photonics applications,” J. Lightwave Technol. 35(4), 824–835 (2017). [CrossRef]  

5. H. D. Griffiths and W. J. Bradford, “Digital generation of high time bandwidth product linear FM waveforms for radar altimeters,” IEE Proc. F Radar Signal Process. UK 139(2), 160–169 (1992). [CrossRef]  

6. H. Kwon and B. Kang, “Linear frequency modulation of voltage-controlled oscillator using delay-line feedback,” IEEE Microw. Wireless Compon. Lett. 15(6), 431–433 (2005). [CrossRef]  

7. J. Yao, “Photonic generation of microwave arbitrary waveforms,” Opt. Commun. 284(15), 3723–3736 (2011). [CrossRef]  

8. C. Wang, M. Li, and J. Yao, “Continuously tunable photonic microwave frequency multiplication by use of an unbalanced temporal pulse shaping system,” IEEE Photonics Technol. Lett. 22(17), 1285–1287 (2010). [CrossRef]  

9. Y. Du, J. Zheng, L. Wang, H. Wang, N. Zhu, and J. Liu, “Widely-tunable and background-free ultra-wideband signals generation utilizing polarization modulation- based optical switch,” IEEE Photonics Technol. Lett. 25(4), 335–337 (2013). [CrossRef]  

10. Y. Zhang, C. Liu, K. Shao, Z. Li, and S. Pan, “Multioctave and reconfigurable frequency-stepped radar waveform generation based on an optical frequency shifting loop,” Opt. Lett. 45(7), 2038–2041 (2020). [CrossRef]  

11. Y. Zhang, C. Liu, K. Shao, C. Ma, L. Li, L. Sun, S. Li, and S. Pan, “Multi-functional radar waveform generation based on optical frequency-time stitching method,” J. Lightwave Technol. 39(2), 458–464 (2021). [CrossRef]  

12. H. Yang, L. Wang, C. Zhao, and H. Zhang, “Sinusoidal frequency-modulated waveforms generated by a phase-modulated frequency-shifting loop,” J. Lightwave Technol. 39(10), 3112–3120 (2021). [CrossRef]  

13. J. D. McKinney, D. E. Leaird, and A. M. Weiner, “Millimeter-wave arbitrary waveform generation with a direct space-to-time pulse shaper,” Opt. Lett. 27(15), 1345–1347 (2002). [CrossRef]  

14. J. D. McKinney, D. Seo, D. E. Leaird, and A. M. Weiner, “Photonically assisted generation of arbitrary millimeter-wave and microwave electromagnetic waveforms via direct space-to-time optical pulse shaping,” J. Lightwave Technol. 21(12), 3020–3028 (2003). [CrossRef]  

15. J. Wang, H. Shen, L. Fan, R. Wu, B. Niu, L. T. Varghese, Y. Xuan, D. E. Leaird, X. Wang, F. Gan, A. M. Weiner, and M. Qi, “Reconfigurable radio-frequency arbitrary waveforms synthesized in a silicon photonic chip,” Nat. Commun. 6(1), 5957 (2015). [CrossRef]  

16. S. Liao, Y. Ding, J. Dong, X. Wang, and X. Zhang, “125-GHz microwave signal generation employing an integrated pulse shaper,” J. Lightwave Technol. 35(13), 2741–2745 (2017). [CrossRef]  

17. V. T. Company, D. E. Leaird, and A. M. Weiner, “Dispersion requirements in coherent frequency-to-time mapping,” Opt. Express 19(24), 24718–24729 (2011). [CrossRef]  

18. L. R. Chen, “Photonic generation of chirped microwave and millimeter wave pulses based on optical spectral shaping and wavelength-to-time mapping in silicon photonics,” Opt. Commun. 373, 70–81 (2016). [CrossRef]  

19. H. Chi, F. Zeng, and J. Yao, “Photonic generation of microwave signals based on pulse shaping,” IEEE Photonics Technol. Lett. 19(9), 668–670 (2007). [CrossRef]  

20. F. Zhang, X. Ge, S. Pan, and J. Yao, “Photonic generation of pulsed microwave signals with tunable frequency and phase based on spectral-shaping and frequency-to-time mapping,” Opt. Lett. 38(20), 4256–4259 (2013). [CrossRef]  

21. F. Zhang, X. Ge, and S. Pan, “Background-free pulsed microwave signal generation based on spectral shaping and frequency-to-time mapping,” Photonics Res. 2(4), B5–B10 (2014). [CrossRef]  

22. W. Wang, J. Liu, H. Mei, W. Sun, W. Wang, H. Yuan, and N. Zhu, “Photonic generation of pulsed microwave signal based on phase shifted Lyot optical filter,” IEEE Photonics Technol. Lett. 27(17), 1845–1848 (2015). [CrossRef]  

23. J. H. Wong, H. Q. Lam, S. Aditya, K. E. K. Lee, V. Wong, P. H. Lim, K. Wu, C. Ouyang, and P. P. Shum, “Photonic generation of tunable continuous-wave microwave signals using a temporally-stretched and chirped pulse-train,” J. Lightwave Technol. 30(9), 1269–1277 (2012). [CrossRef]  

24. X. Jia, L. Tang, S. Liu, H. Ma, T. Tao, L. Chen, J. Wu, C. Li, X. Wang, L. Zhang, and J. Weng, “Frequency-tunable pulsed microwave waveform generation based on unbalanced single-arm interferometer excited by near-infrared femtosecond laser,” Appl. Sci. 11(24), 11928 (2021). [CrossRef]  

25. M. H. Khan, H. Shen, Y. Xuan, L. Zhao, S. Xiao, D. E. Leaird, A. M. Weiner, and M. Qi, “Ultrabroad-bandwidth arbitrary radio frequency waveform generation with a silicon photonic chip-based spectral shaper,” Nat. Photonics 4(2), 117–122 (2010). [CrossRef]  

26. J. Wang, R. Ashrafi, M. Rochette, and L. R. Chen, “Chirped microwave pulse generation using an integrated SiP Bragg grating in a Sagnac loop,” IEEE Photonics Technol. Lett. 27(17), 1876–1879 (2015). [CrossRef]  

27. W. Zhang and J. Yao, “Photonic generation of linearly chirped microwave waveforms using a silicon-based on-chip spectral shaper incorporating two linearly chirped waveguide Bragg gratings,” J. Lightwave Technol. 33(24), 5047–5054 (2015). [CrossRef]  

28. S. Yan, S. Gao, F. Zhou, Y. Ding, J. Dong, X. Cai, and X. Zhang, “Photonic linear chirped microwave signal generation based on the ultra-compact spectral shaper using the slow light effect,” Opt. Lett. 42(17), 3299–3302 (2017). [CrossRef]  

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (8)

Fig. 1.
Fig. 1. Schematic of a microwave waveform generation system based on spectral shaping and WTTM technique.
Fig. 2.
Fig. 2. Schematic of the FSR switchable spectral shaping chip based on the SOI platform.
Fig. 3.
Fig. 3. Microscope images of (a) the single switch network for test and (b) the FSR-selectable spectral shaping structure. Yellow labels represent input or output ports on the chip, while white labels represent the optical switches.
Fig. 4.
Fig. 4. Optical power measured while sweeping the switching voltage of (a) Switch LS1 at LT1 port; (b) Switch LS3 at LT2 port when the voltage applied on LS1 is fixed at 2.58 V; (c) Switch LS7 at LCh1 port when the voltage applied on LS1 and LS3 are fixed at 2.58 V and 3.98 V, respectively.
Fig. 5.
Fig. 5. (a)-(h): Transmission spectrum of all channels when LCh1-LCh8 is tuned to the “on” state.
Fig. 6.
Fig. 6. (a)-(h): Measured transmission spectra of the asymmetric MZIs in RCh1-RCh8 when the corresponding switching channel is tuned to the on-state. The FSR of the spectrum is labeled in each subfigure.
Fig. 7.
Fig. 7. Experimental setup of the frequency selectable microwave waveform generation system. MLFPL: mode-locked femtosecond pulse laser, PC: polarization controller, PVS: programmable voltage source, TEC: temperature control, EDFA: erbium-doped fiber amplifier, PD: photodetector.
Fig. 8.
Fig. 8. (a)-(h): Measured frequency spectra of microwave signals generated via RCh1 to RCh8.

Tables (1)

Tables Icon

Table 1. Characteristics of each switch in the single optical switch network

Equations (4)

Equations on this page are rendered with MathJax. Learn more.

F S R = λ 2 n g Δ L
y ( t ) T ( λ ) λ = t / Φ ¨
f = 1 | Φ ¨ | F S R
f = n g Δ L | Φ ¨ | λ 2
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.