Integrated optoelectronic oscillator

Jian Tang; Tengfei Hao; Wei Li; David Domenech; Rocio Baños; Pascual Muñoz; Ninghua Zhu; José Capmany; Ming Li

doi:10.1364/OE.26.012257

1. Introduction

Applications relying on the interfacing between radiofrequency (RF) signals and optical fiber network segments, call for flexible, agile and compact devices to carry the transition between the wireless and wired signals [1–4]. Such as global wireless fifth generation (5G) communications [5], radar based civil surveillance [6], fast analog-to-digital converters [7] and clock recovery systems [8]. Moreover, RF systems with increased complexity, large frequency tuning range and broad bandwidth [4,9] are urgent for signal generation and processing areas [2,5,7]. Microwave Photonics (MWP) aims at the design and implementation of hybrid optoelectronic systems that can meet the aforementioned targets [2,10–12]. Specifically, an area that urgently needs these solutions within a short/middle term scope is communication networks, which are facing the 5G revolution that aims to service billions users [13,14]. Since the spectrum has become scarce at microwave frequencies, 5G networks need to work at, and span several regions of the millimeter wave (mm-Wave) within RF bandwidth [15]. In addition, the expected characters of 5G networks will call for mm-Wave sources with high-spectral purity, frequency tunability, drastically reduction in the size of the coverage cell area and reduced base station dimensions [5]. These requirements will need to feature reduced power consumption, compact design and very small footprint. In order to meet the above requirements, integrated microwave photonics (IMWP) is proposed by combining the integrated microwave technology and the integrated photonics technology [3].

Among various reported techniques for mm-Wave generation, the optoelectronic oscillator (OEO) is considered as an optimum candidate due to its ability to generate high-frequency signals with ultra-low phase noise [1,16]. Several schemes have been reported to realize and refine the original OEO scheme after the first OEO proposed in 1996 [17,18]. Such as those based on multiple loops [19], stimulated Brillouin scattering effects [16], fiber Bragg gratings [20], injection-locked [21] and coupled cavity designs [22]. The recent development of different photonics integration material platforms, including Silicon on Insulator (SOI), Indium Phosphide (InP) and Silicon Nitride (Si₃N₄), opens the prospective way for the potential of integrating an OEO on chip and has attracted great interests. As the spectral purity of the output microwave signal is directly related to the Q-factor of the OEO loop, traditional OEOs normally use long fiber delay line to obtain the high Q-factor. This feature is hard to realize in principle for an integrated OEO configuration. However, alternative design approaches can be used to achieve long time delay compatible with the reduced chip footprint. One example is the high-Q whispering gallery mode (WGM) resonator which is used in a previously reported miniaturized OEO [1,23]. In this configuration, all of the three optical parts of the proposed OEO (laser, modulator + WGM filter and detector) were discretely packaged together with the RF devices into a metal box with size of $4 \times 5.5 \times 1.2$ cm³. This miniaturized OEO featured tunable oscillation frequency operation and excellent phase noise values below −122 dBc/Hz @100KHz when the oscillation frequency is located at 9.8 GHz [9,11]. Optical chip based OEO with partially integrated optical parts have been reported in recently years [24,25]. For example, in [25], on-chip stimulated Brillouin scattering effect has been used to provide high-Q factor for OEO. It realized wideband microwave frequency tunability with low phase noise. Table 1 summarize the size and degree of integration and phase noise performance of selected ways to reduce the footprint of OEO in recently years. A significant step for next generation OEO should be the fully integrated OEO with all of the photonic and electrical components are monolithically integrated on the same substrate.

Table 1. Selected Ways to Reduce the Footprint of OEO.

View Table

In this letter, we report and experimentally demonstrate an integrated OEO (IOEO) with ultra-compact size. All the photonics components are monolithically integrated in the same chip substrate, and both the optical part and electrical part are packaged on a print circuit board (PCB) within size of $5 \times 6$ cm². In contrast to the miniaturized OEO mentioned above, a spiral-shape optical waveguide is employed as delay line in our prototype to decrease the phase noise. The oscillation frequency of the IOEO is depending on the injection current. RF signal frequency of 7.30 GHz with phase noise value of −91 dBc/Hz@1MHz is obtained by tuning the injection current at 44 mA. When the injection current tuned to 65 mA, the output RF signal frequency is located at 8.87 GHz with phase noise value of −92 dBc/Hz@1MHz. By strictly tuning the injection current, both of the oscillation frequency can be slightly tuned in 20 MHz bandwidth. To the best of our knowledge, it is the first report of an IOEO configuration by fully integrating the photonic section and electrical section on the same board. We believe this result sets a first and important landmark towards the full integrated OEO devices.

2. Design and fabrication

Figure 1(a) shows the schematic of the proposed IOEO. A photograph of the fabricated device is shown in Fig. 1(b). Lightwave from the DML is detected by the PD after propagation through an ODL. As the phase noise of the output RF signal decreases quadratically with the loop delay time [26], the ODL is designed using spiral configuration to obtain the maximum delay time. The total length of the spiral shaped ODL is 8.97 mm, which is limited by the large loss (2 dB/cm) of the InP-based waveguide and the size of the chip. The detected microwave signal then passes through a microwave filter and a tunable amplifier. In order to realize variable gain and minimize the gain noise, a tunable attenuator is inserted into the OEO loop. The amplified signal is split into two sections by an RF splitter with coupled quarter-wavelength topology. One section with 90% power is feedback to modulate the DML, while another section is directly routed to the output port. Moreover, two bias tees are set on the PCB to properly drive the DML and PD.

Fig. 1 Schematic and photograph of the IOEO. (a). Schematic of the proposed IOEO, (b) Photograph of the fabricated IOEO, (c) The photonic part of the IOEO. DML: directly modulated laser; ODL: optical delay line; PD: photo detector; EA: electrical amplifier; ATT: attenuator.

Download Full Size | PDF

Overall, the IOEO is composed of an optical and electrical part. The optical part (shown in the right part of Fig. 1(b) and zoomed in Fig. 1(c)) is manufactured on InP integrated photonics platform which monolithically integrated the directly modulated laser (DML), the optical delay line (ODL) and the photo detector (PD). All of the electrical devices needed in the proposed IOEO loop are assembled on a single PCB as shown in Fig. 1(b). The optical chip is manufactured at the InP fab at Fraunhofer Heinrich Hertz Institute (HHI) in Berlin [27]. The material we chosen is Fe-doped semi-insulating InP wafer with the cell size is $6 \times 4$ mm². The passive components are based on rib waveguides which consists of a bulk InGaAsP quaternary core layer (the bandgap is 1.06 µm) with no other cladding layer deposited on the top (i.e. air confined). The DFB laser is based on a Multi-Quantum-Well (MQW) ridge-waveguide structure. The complex-coupled grating of the DFB laser is defined on the top and etched into the MQW structure. The PD is based on a high-speed diode which is placed on the top of the waveguide. The PCB was manufactured by employing standard fabrication processes in a high-speed Rogers 4003 substrate [28]. The RF components were soldered to the board with a reflow oven and tin-lead based solder paste. The gain of each RF amplifier is 15 dB with a bandwidth larger than 15 GHz. The variable attenuator can reach up to 30 dB attenuation over the bandwidth from 5GHz to 30 GHz. The filter has been implemented by the Ni-Au conductive layer on the PCB in a 9 GHz stub topology with a 6th order Chebyshev response. The InP die is attached to the PCB by the thermo-electric conductive epoxy. Finally, the electrical connections from the die to the RF/DC lines on the PCB are implemented by Au wire bonds. The overall size of the PCB is $5 \times 6$ cm². The total length of the OEO loop is only several centimeters which is much shorter than those fiber based OEOs. Consequently, the corresponding free spectrum range (FSR) is on the order of several GHz which is confirmed by the experimental results shown in Figs. 2(a) and 2(d).

Fig. 2 Frequency response of the IOEO. (a) and (d) represents multimode oscillation spectrum for an injection current of 44 mA and 60 mA, respectively. The solid (empty) triangle represents the primary resonance (second harmonic). (b) Frequency spectrum of the stable output RF signal with the injection current of 44.1 mA. (c) Frequency fluctuation of the primary oscillation frequency shown in (b) during 1 minute time window. (e) Frequency spectrum of the generated stable RF signal with the injection current of 65 mA. (f) Frequency fluctuation of the primary oscillation frequency shown in (e) during 1 minute time window. The spectrum in (a), (b), (d) and (e) was measured under resolution bandwidth of 10 MHz while (c) and (f) was measured under resolution bandwidth of 1 KHz. The cyan dashed line in (a), (b), (d) and (e) represents the response of the RF filter.

Download Full Size | PDF

3. Experiments and results

The first row in Fig. 2 shows the results for an injection current of 44 mA, while the second row presents the results for an injection current of 65 mA. One can observe from Figs. 2(a) and 2(d) that the FSR of the integrated OEO are 3.614 GHz and 2.236 GHz when the injection current is 44 mA and 60 mA, respectively. Once the injection current exceeded a certain value, only the primary resonance mode is retained through mode competition. Such large FSR value reduces the stringent band-pass requirements for the RF filter. Here, the RF filter is centered at 9 GHz with a 3-dB bandwidth of 7 GHz, which can select three oscillation modes at most.

By tuning the injection current, the proposed integrated OEO can oscillate at different frequency with different injection current which are shown in Figs. 2(b) and 2(e). The first stable resonance frequency is 7.30 GHz, featuring amplitude of 0.4 dBm with an injection current of 44.1 mA. The second one is 8.87 GHz with amplitude of −0.5 dBm when the current is tuned to 65 mA. The frequency tuning mechanism can be explained by the fact that the central laser wavelength increases significantly with the increased injection current. As the effective index of the optical waveguide on the chip is inversely related to the laser wavelength, the oscillation loop length will be decreased with the increasing laser wavelength. The oscillation frequency change is given by $Δ f = - f_{0} Δ L / L$ where $f_{0}$ is original oscillation frequency, $L$ is the oscillation loop length and $Δ L$ is the loop length change [26]. Thus, the resonance frequency is increased from 7.30 GHz to 8.87 GHz by increasing the injection current.

The max hold function of the frequency spectrum analyzer (Tektronix, RSA5126B) is employed to characterize the stability of the resonance frequency. Figures 2(c) and 2(f) is drawn by using the recorded oscillation frequency minus the center oscillation frequency during one minute. One can see that the oscillation frequency shifts around 1.5 MHz (1 MHz) in one minute for the oscillation frequency of 7.30 GHz (8.87 GHz). This fast frequency shifting can be attributed to the following two reasons. Firstly, the lasing frequency of the DML is affected by the injection current. Thus, the modulation signal injected into the DML will cause laser frequency noise. After transmitting through the spiral waveguide with chromatic dispersion, this laser frequency noise will cause oscillation delay time variation in OEO loop which is finally transformed to oscillation frequency fluctuation (the dispersion coefficient is −1700 $p s / (n m • k m)$ ) [29]. Secondly, the thermal crosstalk influences the length of the oscillation loop and the response of the DML including the carrier wavelength and power which is finally induced the oscillation frequency shifting. The fast fluctuation of the oscillation frequency also explains the large power difference shown in Fig. 2(b) (Fig. 2(e)) and 2(c) (Fig. 2(f)). When the resolution bandwidth of the ESA is reduced to 1 KHz, the sweep time is slowed down to slower than the oscillation frequency shifting speed. The maximum value of the oscillation frequency may not fall inside the passband of the filter of the ESA. Then the measured power value shown in Fig. 2(c) (Fig. 2(f)) is 4.5 dB lower than the measured result under the resolution bandwidth of 10 MHz shown in Fig. 2(b) (Fig. 2(e)). The oscillation frequency fluctuation can be suppressed by replacing the DML to a single mode laser with an external modulator. Controlling the temperature of the optical chip and PCB separately is expected as an effective method to further stabilize the oscillation frequency. There is a double frequency located at 14.60 GHz with amplitude of −17.93 dBm (shown in Fig. 2(b)) which arises due to the nonlinearity effect in the oscillation loop (caused by DML, PD, EA, etc.) and the insufficient rejection of the RF filter. This situation is improved when the resonance frequency shifts to 8.87 GHz. In this case, the oscillation amplitude is −0.5 dBm as shown in Fig. 2(e) and the double frequency component is located at 17.74 GHz with amplitude of −42.2 dBm. This is mainly caused by the enhanced rejection around 17.74 GHz for RF filter which can be clearly seen from the RF filter response in Fig. 2(e).

Moreover, by changing the injection current, the oscillation frequency can be continuously tuned from 7.30 GHz to 7.32 GHz with a current tuning rate of 15 MHz/mA (Figs. 3(a) and 3(b)). The measured phase noise of the output RF signal is around −91 dBc/Hz@1MHz and only small phase noise changes are observed when the frequency is tuned within the continuously oscillation frequency tuning range (shown in Fig. 3(c)). Similarly, as shown in Figs. 3(d)-3(f), the oscillation frequency at 8.87 GHz can be tuned from 8.86 GHz to 8.88 GHz with a 25 MHz/mA tuning rate. The measured phase noise is around −92 dBc/Hz@1MHz. This frequency tuning mechanism should be the same as the frequency changing mechanism mentioned above. As a consequence of the large FSR of the OEO cavity, mode hopping is prevented while tuning the oscillation frequency. This robustness of the oscillation mode indicates that feedback control is suitable for the chip to stabilize the oscillation frequency.

Fig. 3 Oscillation frequency tunability and phase noise response of the IOEO. (a), (d), The oscillation frequency tunability around 7.3 GHz and 8.8 GHz. (b) Dependence of the oscillation frequency around 7.3 GHz with the injection current. (e) Similar to (b) dependence of the oscillation frequency around 8.8 GHz with the injection current. (c), (f), Measured phase noise of the oscillation frequency at 7.30 GHz and 8.87 GHz, respectively. (a) is measured under resolution bandwidth of 1 KHz while (d) was measured under resolution bandwidth of 100 KHz.

Download Full Size | PDF

The instability of oscillation frequency and high phase noise are induced by several identified limitations of this first version of the integrated chip, which can be summarized into the following points:

a) Low quality factor of the oscillation loop.
Based on previous research, the quality factor Q can be expressed as [21]:
$Q = \frac{f_{0}}{Δ f_{F W H M}} = Q_{D} \frac{τ P_{o}}{G_{A}^{2} ρ_{N}}$
Where, $f_{o}$ is the oscillation frequency of the OEO, $Δ f_{F W H M}$ is the full width at half-maximum of the oscillation mode, $τ$ is the loop group delay, $P_{o} / G_{A}^{2}$ describes the oscillation power before the RF amplifier and the equivalent input noise density is defined as $ρ_{N}$ . Here, $Q_{D} = 2 π f_{0} τ$ is the quality factor of the loop delay. Thus, one can clearly see that the quality factor is quadratically increased with the loop group delay. On the other hand, the oscillator phase noise $L_{φ} (f)$ in terms of the phase noise of the oscillator components $L_{ψ} (f)$ can be expressed as [22]:
$L_{φ} (f) = [1 + {(f_{L} / f)}^{2}] L_{ψ} (f)$
where $f$ is the offset frequency, and $f_{L}$ is Leeson frequency which can be written as: $f_{L} = ν_{0} / (2 Q)$ . Here $ν_{0}$ is oscillation frequency. Combining with the Eq. (1), one can clearly see that the long group delay can decrease the phase noise by increasing the Q factor. However, the small footprint of the delay line limits the group delay which increases the phase noise.
b) Inherent defects of the DML.
The first impact is the high relative intensity noise (RIN) of the DML. Since the temperature control of the laser is really hard to realize on chip, the RIN of laser should be large which will greatly worsen the phase noise. Then the large frequency noise of the laser from DML induced a large delay time fluctuation via the chromatic dispersion of the spiral shaped InP waveguide. The delay time fluctuation leads to the fast oscillation frequency shift which is finally transformed into phase noise [29]. Besides, the DML is characterized by a wide linewidth which increasing the equivalent input noise density in the denominator of Eq. (1) [8,29,30] and then decreasing the Q factor which will directly enhance the phase noise.
c) Thermal effects in the IOEO system.
Since the PCB is athermanous, assembling all the active devices including photonics chip on the PCB will face serious heat dissipation issues. Inherent heat accumulation together with environmental influence affects each device status which further enhances the instability of the system.

According to the above analyses, we can concluded that the ways to improve the performance of the OEO should focus on the following four methods. First of all, a whispering gallery mode resonator can be used to increase the Q factor. Secondly, replacing the directly modulation mechanism by external modulation mechanism to overcome the defects of the DML. In third place, thermal control is important for such integrated chip to suppress the thermal crosstalk and thermal noise. Finally, as the oscillation frequency can be continuously tuned around the center frequency, feedback control should be an effective way to stabilize the oscillation frequency. The phase noise of the IOEO can be estimated as [31]:

S_{ψ} (ω) = \frac{ω_{o}^{2}}{4 Q_{R F}^{2}} \frac{{| {\tilde{χ}}_{F N} (ω) |}^{2} + \frac{{| {\tilde{ρ}}_{r i n} (ω) |}^{2} + {| {\tilde{η}}_{m} (ω) |}^{2}}{4 Q_{R F}^{2}} + \frac{2 Γ_{a}}{{| ν_{o} |}^{2}}}{{| i ω + \frac{ω_{o}}{2 Q_{R F}} [1 - e^{- i ω T}] |}^{2}}

where $Q_{R F} = \frac{ω_{o}}{Δ ω}$ is the quality factor of the RF filter (1.28 in our case). The noise term $χ_{F N} (ω)$ is the effect of the laser frequency noise. $ρ_{r i n} (ω)$ is the effect of the RIN, $η_{m} (ω)$ is the multiplicative noise come from the relative gain fluctuations which is uncorrelated to microwave oscillations, the related power spectral density is $Γ_{a}$ , $ν_{o}$ is the steady state complex amplitude of the microwave, $T$ is the time delay. By taking all the parameters related to the IOEO and some reference parameters from [31,32], we can predict that the phase noise of the IOEO can be improved 20 dB by decreasing the laser RIN of 20 dB. By replacing the DML to the narrow linewidth DFB laser with MZM and controlling the temperature, reducing the RIN of laser by 20 dB can be expected [32]. Furthermore, Ring resonator based discretely packaged OEO has been realized by employing an Z-cut stoichiometric lithium tantalate ring resonator. The phase noise of the OEO reached −122dBc/Hz @ 100KHz for oscillation frequency of 9.8 GHz without employing fiber delay line [9]. By reducing laser RIN of 10~50 dB over the bandwidth of 10 MHz, the phase noise of the proposed OEO decreased 10 dB for offset frequency from 10 Hz to 10 KHz [33]. Feedback control is also realized by phase-locked loop which achieved long term frequency stability [34]. Thus, we expect the phase noise of the IOEO can be as low as −110dBc/Hz @1MHz after the improvement.

4. Conclusion

In summary, we propose and experimentally demonstrate a novel integrated OEO which composes of two parts on chip. The optical part is manufactured by integration photonic platform based on InP. All the required photonic components (i.e., the DML, the ODL and the PD) are monolithically integrated on chip. The electrical part assembles all of the electrical devices needed in the OEO loop on a single PCB. Through tuning the injection current, we demonstrated the IOEO can be operated at two different oscillation frequencies with different injection current (7.3 and 8.87 GHz RF respectively). In both cases the oscillation frequency can be slightly shifted around the central oscillation frequency (i.e. within a range of 20 MHz) by tuning the injection current. Phase noise values of −91 dBc/Hz @1MHz for 7.30 GHz and −92 dBc/Hz@1MHz for 8.87 GHz has been measured respectively. Due to the low values of the cavity Q factor and the defects of the DML, the phase noise value of the signal from IOEO are higher than those obtained in OEOs featuring longer cavities implemented by fibers. The methods to improve the cavity Q factor and to increase the oscillation frequency stability have been discussed and proposed for incorporation in future IOEO designs. Integrated technology brings unique advantages (such as compact design, small footprint, large FSR and oscillation frequency tunability) to meet the requirements of emerging communication, radar and wireless services application scenarios.

Funding

National Natural Science Foundation of China (61090391, 61377002, 61522509); Beijing Natural Science Foundation (4152052); National High-Tech Research and Development Program of China (2015AA017102); Thousand Young Talent Program.

References and links

1. L. Maleki, “Sources: The optoelectronic oscillator,” Nat. Photonics 5(12), 728–730 (2011). [CrossRef]

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

3. D. Marpaung, C. Roeloffzen, R. Heideman, A. Leinse, S. Sales, and J. Capmany, “Integrated microwave photonics,” Laser Photonics Rev. 7(4), 506–538 (2013). [CrossRef]

4. D. Dolfi, “New trends in optoelectronics for radar, E.W. and communication systems,” in IEEE International Topical Meeting on Microwave Photonics (MWP, 2011).

5. M. H. Alsharif and R. Nordin, “Evolution towards fifth generation (5G) wireless networks: Current trends and challenges in the deployment of millimeter wave, massive MIMO, and small cells,” Telecomm. Syst. 64(4), 617–637 (2017). [CrossRef]

6. P. Ghelfi, F. Laghezza, F. Scotti, G. Serafino, A. Capria, S. Pinna, D. Onori, C. Porzi, M. Scaffardi, A. Malacarne, V. Vercesi, E. Lazzeri, F. Berizzi, and A. Bogoni, “A fully photonics-based coherent radar system,” Nature 507(7492), 341–345 (2014). [CrossRef] [PubMed]

7. L. Yu, W. Zou, G. Yang, X. Li, and J. Chen, “Electro-optical switching based demultiplexing in high-speed photonic analog-to-digital converter based on actively mode-lock laser,” in 2017 International Topical Meeting on Microwave Photonics (2017), paper THP.29. [CrossRef]

8. X. Li and J. Yu, “Generation and heterodyne detection of >100-Gb/s Q-band PDM-64QAM mm-wave signal,” IEEE Photonics Technol. Lett. 29(1), 27–30 (2017). [CrossRef]

9. A. A. Savchenkov, V. S. Ilchenko, J. Byrd, W. Liang, D. Eliyahu, A. B. Matsko, D. Seidel, and L. Maleki, “Whispering-gallery mode based opto-electronic oscillators,” in 2010 IEEE International Frequency Control Symposium (2010), pp. 554–557. [CrossRef]

10. S. Poinsot, H. Porte, J. P. Goedgebuer, W. T. Rhodes, and B. Boussert, “Continuous radio-frequency tuning of an optoelectronic oscillator with dispersive feedback,” Opt. Lett. 27(15), 1300–1302 (2002). [CrossRef] [PubMed]

11. L. Maleki, “The opto-electronic oscillator (OEO): Review and recent progress,” in European Frequency and Time Forum (EFTF, 2012), pp. 497–500. [CrossRef]

12. J. Capmany, J. Mora, I. Gasulla, J. Sancho, J. Lloret, and S. Sales, “Microwave Photonic Signal Processing,” J. Lightwave Technol. 31(4), 571–586 (2013). [CrossRef]

13. M. Agiwal, A. Roy, and N. Saxena, “Next Generation 5G Wireless Networks: A Comprehensive Survey,” Commun. Surv. Tutorials 18(3), 1617–1655 (2016). [CrossRef]

14. J. S. Fandiño, P. Muñoz, D. Doménech, and J. Capmany, “A monolithic integrated photonic microwave filter,” Nat. Photonics 11(2), 124–129 (2017). [CrossRef]

15. F. Boccardi, R. W. Heath, A. Lozano, T. L. Marzetta, and P. Popovski, “Five disruptive technology directions for 5G,” IEEE Commun. Mag. 52(2), 74–80 (2014). [CrossRef]

16. H. Peng, Y. Xu, X. Peng, X. Zhu, R. Guo, F. Chen, H. Du, Y. Chen, C. Zhang, L. Zhu, W. Hu, and Z. Chen, “Wideband tunable optoelectronic oscillator based on the deamplification of stimulated Brillouin scattering,” Opt. Express 25(9), 10287–10305 (2017). [CrossRef] [PubMed]

17. X. S. Yao and L. Maleki, “Optoelectronic oscillator for photonic systems,” IEEE J. Quantum Electron. 32(7), 1141–1149 (1996). [CrossRef]

18. X. S. Yao and L. Maleki, “Opto-electronic oscillator and its applications,” in Microwave Photonics Conference (1996), pp. 265–268.

19. S. García and I. Gasulla, “Multi-cavity optoelectronic oscillators using multicore fibers,” Opt. Express 23(3), 2403–2415 (2015). [CrossRef] [PubMed]

20. W. Li and J. Yao, “An Optically Tunable Optoelectronic Oscillator,” J. Lightwave Technol. 28(18), 2640–2645 (2010). [CrossRef]

21. K. H. Lee, J. Y. Kim, and W. Y. Choi, “Injection-Locked Hybrid Optoelectronic Oscillators for Single-Mode Oscillation,” IEEE Photonics Technol. Lett. 20, 1645–1647 (2008). [CrossRef]

22. E. Salik, N. Yu, and L. Maleki, “An Ultralow Phase Noise Coupled Optoelectronic Oscillator,” IEEE Photonics Technol. Lett. 19(6), 444–446 (2007). [CrossRef]

23. K. Saleh, R. Henriet, S. Diallo, G. Lin, R. Martinenghi, I. V. Balakireva, P. Salzenstein, A. Coillet, and Y. K. Chembo, “Phase noise performance comparison between optoelectronic oscillators based on optical delay lines and whispering gallery mode resonators,” Opt. Express 22(26), 32158–32173 (2014). [CrossRef] [PubMed]

24. J. Li, H. Lee, and K. J. Vahala, “Microwave synthesizer using an on-chip Brillouin oscillator,” Nat. Commun. 4, 2097 (2013). [CrossRef] [PubMed]

25. M. Merklein, B. Stiller, I. V. Kabakova, U. S. Mutugala, K. Vu, S. J. Madden, B. J. Eggleton, and R. Slavík, “Widely tunable, low phase noise microwave source based on a photonic chip,” Opt. Lett. 41(20), 4633–4636 (2016). [CrossRef] [PubMed]

26. X. S. Yao and L. Maleki, “Optoelectronic microwave oscillator,” J. Opt. Soc. Am. B 13(8), 1725–1735 (1996). [CrossRef]

27. https://www.hhi.fraunhofer.de/en/departments/pc/researchgroups/foundry-services-for-inp-based-photonic-integrated-circuits.html.

28. http://www.rogerscorp.com/acs/producttypes/9/RO4000-Laminates.aspx

29. L. Wang, N. Zhu, J. Liu, W. Li, H. Zhu, and W. Wang, “Experimental optimization of phase noise performance of optoelectronic oscillator based on directly modulated laser,” Chin. Sci. Bull. 57(31), 4087–4090 (2012). [CrossRef]

30. J. Xiong, R. Wang, T. Fang, T. Pu, D. Chen, L. Lu, P. Xiang, J. Zheng, and J. Zhao, “Low-cost and wideband frequency tunable optoelectronic oscillator based on a directly modulated distributed feedback semiconductor laser,” Opt. Lett. 38(20), 4128–4130 (2013). [CrossRef] [PubMed]

31. K. Volyanskiy, Y. K. Chembo, L. Larger, and E. Rubiola, “Contribution of laser frequency and power fluctuations to the microwave phase noise of optoelectronic oscillators,” J. Lightwave Technol. 28(18), 2730–2735 (2010). [CrossRef]

32. M. Ahmed, “Spectral lineshape and noise of semiconductor lasers under analog intensity modulation,” J. Phys. D Appl. Phys. 41(17), 175104 (2008). [CrossRef]

33. C. W. Nelson, A. Hati, and D. A. Howe, “Relative intensity noise suppression for RF photonic links,” IEEE Photonics Technol. Lett. 18(18), 1542–1544 (2008). [CrossRef]

34. Y. Zhang, D. Hou, and J. Zhao, “Long-term frequency stabilization of an optoelectronic oscillator using phase-locked loop,” J. Lightwave Technol. 32(13), 2408–2414 (2014). [CrossRef]

Architecture	Size	Degree of integration
With a WGM resonator [11]	Small with a size of $4 \times 5.5 \times 1.2$ cm³	Discretely packaged photonic and RF parts	−122 dBc/Hz@100KHz for osicllation frequency of 9.8 GHz
With a Chalcogenide chip [24]	Relative Large	Partially integrated photonic parts	−110 dBc/Hz @100KHz for osicllation frequency of 21.7 GHz
With a DML in this work	Small with a size of $5 \times 6$ cm²	Monolithically integrated photonic and RF parts.	−92 dBc/Hz@1MHz for osicllation frequency of 8.87 GHz

Integrated optoelectronic oscillator

Abstract

1. Introduction

2. Design and fabrication

3. Experiments and results

4. Conclusion

Funding

References and links

Cited By

Figures (3)

Tables (1)

Equations (3)

Optics Express