Open Access
Add to BibSonomy
Abstract
A novel photonic method of linearly frequency-modulated (LFM) signal generation with high purity based on the monolithically integrated semiconductor laser (MISL) subject to the dynamical optoelectrical feedback is proposed and demonstrated in this paper. In this approach, the MISL is firstly operated in period-one state. By introducing the dynamical optoelectrical feedback to modulate the MISL, the generated LFM signals would be constantly optimized as long as the delay of the feedback loop is matched with the repetition period of the LFM signal. In this system, no additional high-speed external modulator, high-frequency electrical LFM oscillator are required, highly simplifying the framework and reducing the power consumption. In the current proof-of-concept experiment, one LFM signal with the bandwidth as large as 5.6 GHz is generated and the corresponding frequency comb contrast can be drastically improved by 51 dB. Furthermore, the effect of the delay mismatch is also discussed in this paper.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Linearly frequency-modulated (LFM) signals have been extensively employed in radar and sonar systems due to their unique characteristics in the application of detecting long-distance targets with high degree of resolution [1–3]. Especially, the LFM signal with a large time-bandwidth product (TBWP) has been most appealing for that it can further advance the detection range and the range resolution, and thus meet the urging requirements of the future radar and sonar. Here, the TBWP is calculated as the product of the pulse duration and spectral width, and it can reflect the pulse compression capacity of the LFM signal. However, attributing to the limitations of the electrical bottleneck, basically all of the LFM signals generated by traditional electrical methods for example voltage-controlled oscillator [4], current-controlled oscillator [5], surface acoustic wave [6] or direct digital synthesizer technique [7], are subject to one of the following conditions such as the limited performance parameters (bandwidth, central frequency as well as the small TBWP), poor tunability, and high phase noise. To overcome these shortcomings, a large number of photonic approaches have been proposed [8–14]. What’s more, with the help of the unique characteristics of the photons such as low-loss transmission, anti-electromagnetic interference and so on, the LFM signals generated by optical means can be capable of long-haul transmission to the remote antenna through optical cables, largely improving the survivability of the command center on the battlefield. Among all photonic methods, a common approach is making full use of the spectral shaping and frequency-to-time mapping (SS-FTT) technique [8–10]. In such a system, the ultra-short optical pulse is firstly shaped in frequency domain, and then is transmitted through a dispersive element to perform the nonlinear FTT mapping. After the photodetector (PD), a chirp electrical pulse with a high central frequency and large chirp rate can be generated. The drawback in this mechanism is that the tunability is seriously poor because the spectrum response of the complex spectral shaper is usually fixed and cannot be arbitrarily changed, leading to the LFM microwave signals with restricted parameters. Another photonic scheme is based on the optical heterodyning technique of two diverse optical signals [11]. Compared to the former method, this scheme is flexible and can own a great tunability in the chirped frequency. However, the high phase noise is inescapable as a result of the utilization of two irrelevant laser sources.
Recently, a novel promising method by employing the unique characteristics of the nonlinear dynamics of the distributed feedback (DFB) semiconductor lasers has drawn much attention [12–14]. One method is based on the dual-mode state of the optically feedback semiconductor laser system, while the other focuses on the period-one (P1) state of the optically injected semiconductor laser system. The former makes full use of the linear region of the feedback strength with the frequency of the generated microwave signal. In this method, only one LFM signal with a small bandwidth of 3.3 GHz has been reported as a result of the limited linear region until now [12]. In addition, though this integrated laser is compact, the manufacturing technology in active-passive integration is still complicated. As to the latter one, LFM continuous waveform signals with 12-GHz bandwidth based on the discrete optically injection system [13], and 6.7-GHz bandwidth based on the monolithically integrated semiconductor laser (MISL) [14] were separately achieved, indicating a great superiority. However, there is a common shortcoming with the above schemes in that the signal quality is seriously poor with a large linewidth of several MHz level. During the last two decades, optical feedback and the optoelectrical feedback systems have been widely discussed in the stabilizing the microwave signals [15–17], which can be also applied in improving the signal quality of the LFM microwave signals [18–21]. However, these LFM generators are based on the discrete optical injection systems, in which the external modulator, multi semiconductor lasers, optical circulator, and polarization controller are included, leading to the configurations large bulk, high power consumption, degraded stability and expensive cost. Therefore, for future miniaturization of radar system, a compact LFM generator on the premise of ensuring signal quality is highly demanded.
In this paper, one high-purity LFM signal generation based on the MISL subject to the dynamical optoelectrical feedback (DOEF) is proposed and experimentally demonstrated. The core device, i.e., MISL is highly integrated with three sections sharing one substrate and it can be flexibly fabricated on a large scale by making full use of the Reconstruction Equivalent Chirp (REC) technique [22]. After giving the appropriate control parameters, the MISL would work in P1 state and one LFM signal would be generated preliminary. Subsequently, by introducing the DOEF to modulate the MISL, the generated LFM signals would be constantly optimized in noise suppression in each mode and the coherence between the adjacent modes as long as the delay of the feedback loop is matched with the repetition period of the LFM signal. In this system, no additional high-speed external modulator, high-frequency electrical LFM oscillator are required, highly simplifying the framework and reducing the power consumption. In the current proof-of-concept experiment, one LFM signal with the bandwidth as large as 5.6 GHz is generated. The corresponding frequency comb contrast is improved by 51 dB, indicating a significant enhancement in the performance after using the DOEF. To the best of our knowledge, it is the first time to research the improvement of the generated LFM signal based on the MISL and it has been the optimal result reported until now. In the end, the case of the delay mismatch is also discussed in this paper.
2. Operation principle
Figure 1 illustrates the experimental schematic of the proposed scheme. The system mainly includes two components, where one is the production configuration of the LFM signal and the other is the optimization system of the LFM signal. As the production device, the internal structure of the MISL is the same as that in [14] and it can be mass-produced flexibly on one wafer at a time by taking advantage of the REC technique. Due to no existence of the isolator or circulator within it, the oscillating lights from two DFB laser sections would be mutually injected and drastically interact with each other after passing through the phase tuning section. Thus, abundant nonlinear dynamical states from mutual injection locking, P1 oscillation, period-two state, four-wave-mixing state to chaos can be achieved by giving different combinations of driving currents. In view of that there is no essential difference between the two laser sections, for the sake of distinction, we define the laser which can be modulated by radio frequency (RF) signal as the front DFB laser (FL) and the other is defined as the rear DFB laser (RL). As to the optimization configuration, one optoelectrical feedback loop composed of long single-mode fiber (SMF), PD, and electrical amplifier (EA) is utilized. Here, one fact supposed to be emphasized is that different from the single-tone signal in the conventional optoelectrical oscillator (OEO) link, the signals in our system are all time-varying in frequency. Hence, to realize the optimization of each mode in the frequency-modulated signals, the DOEF must be ensured. In the DOEF mechanism, the feedback loop delay is significantly important, which is mainly determined by the length of the SMF and the electrical cable, as well as the response speed of the EA, PD, RF modulation and so on. Only when the optical signal, which is regenerated through directly modulating the MISL with feedback microwave signal, is perfectly overlapped with the optical modes generated by the control signal under the free-running P1 state at every moment in a period, the signal quality of the frequency-modulated signal output from the MISL can be optimized. Thus, in order to meet the above requirement, the feedback loop delay must be equal to the integral multiple of the repetition period.
Fig. 1. The experimental setup of the high-purity LFM signal generation module. MISL: monolithically integrated semiconductor laser, RL: rear DFB laser, FL: front DFB laser, SMF: single-mode fiber, PD: photodetector, RF: radio frequency, OSC: oscilloscope.
To acknowledge whether it is feasible to realize the LFM signal generation based on the P1 state in the MISL, here, the theoretical simulation is researched. As mentioned in [23], the expression of the MISL subject to the short-coupled mutual injection regime can be written as the following set of delay differential equations.
Attributing to the fact that the DFB semiconductor laser belongs to the B-class laser, both the wavelength and the strength of the output light would be changed when the driver current is tuned. Thereby, in the MISL, the coupling strength $\kappa $, coupling time delay $\tau $ as well as the detuning frequency $\mathrm{\Delta }f$ would be altered subsequently. Here, to verify the high-speed frequency switching feature of the MISL, these parameters are supposed to be changed at the mean time and the concrete values are given as followed.
Fig. 2. Simulation results, (a) the temporal domain curve, (b) the critical section from the low-frequency oscillation to the high-frequency oscillation, (c) the critical section from the high-frequency oscillation to the low-frequency oscillation, (d) the instantaneous frequency variation curve.
Table 1. Laser specific values of the MISL in the simulation
3. Experiment results
3.1 Experimental setup
A proof-of-concept experiment is performed following the structure in the Fig. 1. For ease of use, the MISL chip is packaged in butterfly housing as a module and the phase section is set unconnected to simplify the system. During the entire process, the current of RL is directly supplied by the laser diode controller (Thorlabs LDC205C) while the FL is voltage-controlled by arbitrary waveform generator (AWG, Agilent 33250A) with a typical impedance value at 50 Ω through the electrode. One temperature controller (Thorlabs 200C) is utilized to provide a stable working environment with the temperature at 25°C. The output optical signal from the MISL is firstly divided by an optical splitter. 50% is extracted to the real-time oscilloscope (Lecroy SDA 830Zi-A) to observe the time-domain waveforms after the opto-electrical conversion. The rest of the optical signal is transmitted to 1.088 km SMF to form the feedback loop. After the PD, the generated electrical signal is amplified by the two-stage EAs (Mini-Circuit ZX60–24-S+, SHF L810A) and subsequently modulated onto the MISL through RF port.
3.2 Experimental results
Firstly, the feedback loop is not closed for verifying the high-speed frequency switching characteristic of the P1 state within the MISL. By setting the control signal with a similar pattern to Costas sequence of (2, 4, 3, 1), one microwave Costas sequence would be generated. The corresponding temporal curve and the frequency-time curve are depicted as the Fig. 3 shows. Apparently, with the control voltage varied, the output microwave waveforms would be changed and in one period, four kinds of the oscillating microwave signals with different frequencies appear. It can be also found that the switching time between the different frequencies is less than 40 ps, indicating that once the tuning speed is far below 25 GHz, the output signal from the MISL would be rapidly changed and operate in a new P1 state stably until the control signal alters again. This result is fully consistent with the simulation result and in this case, this MISL has the ability of generating LFM signal.
Fig. 3. (a) The generated temporal waveform of the microwave Costas sequence, (b) the corresponding frequency-time curve.
Subsequently, by changing the driver control signal from the AWG to the sawtooth signal, the LFM signal would be generated. As Fig. 4 shows, one LFM signal with a 4.6-us duration time and a 4.5-GHz bandwidth from 16 GHz to 20.5 GHz is achieved when the RL current is set at 81.44 mA and the voltage of the control signal applied on the FL changes linearly from 1.185 V to 2.295 V. The duty cycle is not 100% and this phenomenon can be eliminated as long as the control signal pattern is adjusted appropriately to guarantee the MISL work in the P1 state from beginning to end. Unfortunately, due to the existence of the intrinsic laser noise and the incoherence between different frequency points, the LFM signal quality is not very well and thus the curve of the instantaneous frequency is vague in Fig. 4(b). The detailed frequency spectrum with the span at 5 MHz is also given as the Fig. 4(c) depicts. It can be seen that the spur suppression ratio is lower than 2 dB, which means that the signal is nearly drown in the noise. Obviously, this kind of the signal can be not directly applied.
Fig. 4. (a) The generated temporal waveform when the FL is injected by the sawtooth signal, (b) the calculated instantaneous frequency curve, (c) the frequency spectrum with the center frequency at the 17.3 GHz and the span at 5 MHz.
Next, the DOEF mechanism is considered. In this case, the feedback delay is supposed to be the integral multiple of the repetition period of the LFM signal. To measure the whole delay of the feedback system, the free spectral range (FSR) of the link is firstly researched because the two values are fully opposite. By fixing the control currents of the twin DFB laser sections at a constant combination and prompting the MISL-based system to form the single-frequency OEO, the FSR can be acquired. As the Fig. 5 shows, when the MISL is stably oscillated under the P1 state with the frequency of the 17.43 GHz, the FSR is measured at 181.16 kHz.
Fig. 5. (a) The generated microwave signal at 17.43 GHz in the single-frequency OEO, (b) the detailed drawing of the spectrum within 1-MHz span.
By changing the repetition frequency of the control signal to 181.16 kHz and performing the fine tuning simultaneously, the DOEF would work. Figure 6 shows the optimized LFM signal, which is generated when specifically adjusting the repetition frequency to be 181.2591 kHz. Obviously, the amplitude of the signal in temporal domain becomes stable and the curve of the instantaneous frequency becomes clearer and finer when comparing the experimental results in the Fig. 4 and the Fig. 6. Further, the concrete drawings of the spectrum with the 10-MHz and 2-MHz span are also given. It can be observed that attributing to the DOEF, a 51-dB improvement in the frequency comb contrast at every frequency point in the spectrum can be allocated. Consequently, after introducing the DOEF, the generated LFM signal shows a greater coherence in spectral purity and lower phase noise.
Fig. 6. (a) The generated temporal waveform after introducing the DOEF, (b) the calculated instantaneous frequency curve after introducing the DOEF, (c) the frequency spectrum with the span at 10 MHz, (d) the frequency spectrum with the 2-MHz span.
By changing the control parameters to different combinations, various optimizations can be observed provided that the feedback system is not changed. As the Fig. 7 shows, another LFM signal with a 3.25-us duration time and a 5.6-GHz bandwidth from 14.6 GHz to 20.2 GHz is also achieved when the RL current is set at 85.02 mA and the voltage of the control signal applied on the FL changes from 1.15 V to 2.33 V, suggesting an 18200 of TBWP.
Further, the conditions when the loop delay is not matched with the repetition period are also discussed in our experiment as the Fig. 8 shows. When the repetition frequency is adjusted to be 180.0591 kHz, the optimization degree is largely reduced, and the comb contrast is only 26 dB. When the repetition frequency is adjusted to be 160 kHz, the comb contrast between the adjacent frequency modes is smaller than 10 dB, indicating a poor signal quality.
Fig. 8. The frequency spectra under different repetition frequencies, (a) 180.0591 kHz, (b) 160 kHz.
4. Discussion and conclusion
Actually, this optimization system is not limited to LFM signal and the common radar signals including frequency-hopping signal, triangular-modulated signal and so on can be also generated with high purity. In addition, based on the continuous tunability in a large range of the P1 frequency, and the all-optical gain OEO [24], the characteristics of the generated LFM signal can be further advanced, which is expected to realize the LFM signal with higher center frequency and bandwidth. However, what we need to pay attention to is that, in this system, to advance the quality of the generated LFM signal, the repetition period of the LFM signal is adjusted to match the feedback loop delay while in the futural application, the repetition period is required to be changeable. Thus, how to realize the flexibly and precisely tuning of the feedback loop delay is worthy of being researched. In addition, the coherence between the different pulses is of great importance in the long-distance detection but not considered here. Besides, though the generation configuration of the LFM signal is high-integrated, the whole optimization system is not fully integrated. The above questions are the next key research directions.
To conclude, in this paper, we have proposed and experimentally demonstrated a novel photonic method of the LFM signal generation with high purity based on the MISL subject to the DOEF. By adjusting the repetition period of the LFM signal to make it matched with the delay of the feedback loop, the generated LFM signal would be optimized. In the current proof-of-concept experiment, one high-linearity LFM signal with the bandwidth as large as 5.6 GHz is generated and the corresponding frequency comb contrast can be improved as high as 51 dB, which is the optimal result reported until now. Compared with previous approaches, this paper indicates that the generated LFM signal based on the MISL can be also improved, which is expected to be a more minimized and cost-effective configuration applied to the radar and sonar system.
Funding
National Natural Science Foundation of China (61974165, 62071487, 62201615).
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. M. I. Skolnik, Radar Handbook (McGraw-Hill, 2008).
2. S. Maresca, G. Serafino, C. Noviello, et al., “Field Trial of a Coherent, Widely Distributed, Dual-Band Photonics-Based MIMO Radar With ISAR Imaging Capabilities,” J. Lightwave Technol. 40(20), 6626–6635 (2022). [CrossRef]
3. X. Li, Y. Zhou, S. Zhao, et al., “Research progress of integrated radar-communication waveform based on linear frequency modulation,” Laser & Optoelectronics Progress 60(5), 0500003 (2023). [CrossRef]
4. H. Kwon and B. Kang, “Linear frequency modulation of voltage-controlled oscillator using delay-line feedback,” IEEE Microw. Wirel. Compon. Lett. 15(6), 431–433 (2005). [CrossRef]
5. J. Tratnik and Matjaž Vidmar, “2.8 GHz - 5.7 GHz very fast UWB CCO using discrete- packaged SiGe RF transistors,” Informacije MIDEM 41(1), 70–72 (2011).
6. D. G. Johnson and G. M. Brooker, “Wide Band Linearization of a Millimetre-Wave, Linear Frequency Modulated Radar Employing a Surface Acoustic Wave, Delay Line Discriminator,” in Smart Sensors and Sensing Technology, G. S. Gupta, (Springer Verlag, 2008).
7. E. Adler, E. Viveiros, T. Ton, et al., “Direct digital synthesis applications for radar development,” in Proceedings of International Radar Conference (IEEE, 1995), pp. 224–226.
8. C. Wang and J. Yao, “Photonic Generation of Chirped Millimeter-Wave Pulses Based on Nonlinear Frequency-to-Time Mapping in a Nonlinearly Chirped Fiber Bragg Grating,” IEEE Trans. Microw. Theory Tech. 56(2), 542–553 (2008). [CrossRef]
9. C. Wang and J. Yao, “Chirped Microwave Pulse Generation Based on Optical Spectral Shaping and Wavelength-to-Time Mapping Using a Sagnac Loop Mirror Incorporating a Chirped Fiber Bragg Grating,” J. Lightwave Technol. 27(16), 3336–3341 (2009). [CrossRef]
10. C. Wang and J. Yao, “Photonic generation of chirped microwave pulses using superimposed chirped fiber Bragg gratings,” IEEE Photonics Technol. Lett. 20(11), 882–884 (2008). [CrossRef]
11. J. M. Wun, C. C. Wei, J. Chen, et al., “Photonic chirped radio-frequency generator with ultra-fast sweeping rate and ultra-wide sweeping range,” Opt. Express 21(9), 11475–11481 (2013). [CrossRef]
12. L. Guo, D. Lu, B. Pan, et al., “Linearly Chirped Microwave Generation Using a Monolithic Integrated Amplified Feedback Laser,” IEEE Photonics Technol. Lett. 29(21), 1915–1918 (2017). [CrossRef]
13. P. Zhou, F. Z. Zhang, Q. S. Guo, et al., “Linearly Chirped Microwave Waveform Generation with Large Time-Bandwidth Product by Optically Injected Semiconductor Laser,” Opt. Express 24(16), 18460–18467 (2016). [CrossRef]
14. J. Li, T. Pu, J. Zheng, et al., “Photonic generation of linearly chirped microwave waveforms using a monolithic integrated three-section laser,” Opt. Express 26(8), 9676–9685 (2018). [CrossRef]
15. L. C. Lin, S. H. Liu, and F. Y. Lin, “Stability of period-one (p1) oscillations generated by semiconductor lasers subject to optical injection or optical feedback,” Opt. Express 25(21), 25523–25532 (2017). [CrossRef]
16. R. J. Steed, F. Pozzi, M. J. Fice, et al., “Monolithically integrated heterodyne optical phase-lock loop with RF XOR phase detector,” Opt. Express 19(21), 20048–20053 (2011). [CrossRef]
17. L. Naglic, L. Pavlovic, B. Batagelj, et al., “Improved phase detector for electro-optical phase-locked loops,” Electron. Lett. 44(12), 758 (2008). [CrossRef]
18. P. Zhou, Z. Zhang, and P. Shilong, “Generation of linear frequency-modulated waveforms by a frequency-sweeping optoelectronic oscillator,” J. Lightwave Technol. 36(18), 3927–3934 (2018). [CrossRef]
19. J. P. Zhuang, X. Z. Li, S. S. Li, et al., “Frequency-modulated microwave generation with feedback stabilization using an optically injected semiconductor laser,” Opt. Lett. 41(24), 5764 (2016). [CrossRef]
20. X. D. Lin, G. Q. Xia, Z. Shang, et al., “Frequency-modulated continuous-wave generation based on an optically injected semiconductor laser with optical feedback stabilization,” Opt. Express 27(2), 1217–1225 (2019). [CrossRef]
21. T. F. Hao, Y. Z. Liu, J. Tang, et al., “Recent advances in optoelectronic oscillators,” Adv. Photonics 2(04), 1 (2020). [CrossRef]
22. Y. Shi, S. Li, X. Chen, et al., “High channel count and high precision channel spacing multi-wavelength laser array for future PICs,” Sci. Rep. 4(1), 7377 (2014). [CrossRef]
23. J. Li, J. Zheng, T. Pu, et al., “Monolithically integrated multi-section semiconductor lasers: towards the future of integrated microwave photonics,” Optik 226, 165724 (2021). [CrossRef]
24. J. Li, J. Zheng, T. Pu, et al., “All-optical gain optoelectronic oscillator based on the dual-frequency integrated semiconductor laser: potential to break the bandwidth limitation in the traditional OEO configuration,” Opt. Express 29(2), 1064–7046 (2021). [CrossRef]
