Open Access
Add to BibSonomy
Abstract
A wafer-level and high-efficiency radio frequency (RF) testing of a photonic device is highly desired in the fabrication and characterization of large-scale photonic integration circuits. In this work, we propose on-wafer probing kit designs, and demonstrate a damage-free, self-calibrated RF characterization of an integrated silicon photonic transceiver with a heterodyne mixing approach. Reduced or even free of fiber coupling off chip operation can be achieved with the on-wafer probing-kit to extract the frequency responses of broadband modulators and photodetectors in the photonic integration transceiver, with no requirement of electro-optical or opto-electrical calibration. A proof-of-concept probing kit is designed and fabricated with an on-chip electroabsorption modulator (EAM) and photodetectors by heterogeneously integrated III-V material on silicon substrate. On-wafer RF measurements with the self-calibration method are experimentally demonstrated with an accuracy analysis compared with the conventional swept-frequency method. The on-wafer and full-electrical test nature of the probing kit significantly advances performance monitoring of photonic integration circuits during chip fabrication, and promisingly offers predictable outcome and yield analysis before packaging.
© 2017 Optical Society of America
1. Introduction
Large bandwidth density and cost-effective photonic integration circuits (PICs) promise future optical communication system use in high performance computer and datacenters [1]. The vision of high-level photonic integration spurs the interests in silicon photonics, which is scalable up to large substrate (up to 450 mm today) and compatible with complementary metal oxide semiconductor (CMOS) fabrication techniques. In particular, some silicon based integration technologies, e.g. the heterogeneous integration on silicon, allow solutions for on-chip lasers by integrating other materials system to silicon substrate, enabling a full library of functional components for high level integration in varied applications [2–4]. Like the “Moore’s Law” on electrical integrated circuits, the level of integration on silicon PICs increases dramatically, with number of integrated elements in one chip doubles each year [5]. Silicon phase array with over 4000 silicon phase tuners in a die [4], and integrated network-on-circuit (NoC) chip with over 400 components with on-chip lasers with heterogeneous silicon integration [6] have been reported, showing the trend toward very large-scale photonic integration. This is to say at the wafer-level, a total of 106 effective photonic unit can be reached, e.g. on a 300 mm or 450 mm silicon wafer.
This dense photonic integration on large wafer, however, raises challenges for process monitoring and automatic testing. Particularly, a fast and automatic evaluation method for high-speed optoelectronic devices is still lacking. In an integrated optical communication system, the overall performance is normally restrained by the high-speed components such as the modulator at transmitter and photodetector at receiver. For highly integrated transceiver system at wafer level, the dynamic characteristic of those components has the following requirements and preferences: (a) reduce or avoid optical coupling off chip to enable efficient automatic probing, (b) avoid intense and complex module calibrations, and (c) damage-free for the purpose of wafer level analysis. However, the most commonly used test techniques cannot fulfill all these requirements.
A variety of optical or electrical methods have been developed for measuring the frequency response of optoelectronic devices. Optical spectrum analysis method [7–10], swept frequency method [11,12], and optical down-conversion method [13] are developed to characterize the electro-optical (E/O) frequency response, such as for modulators. Intensity noise method [14, 15], optical heterodyne method [16–19], swept frequency method [11, 12], and the harmonics analysis method [20–22] are for opto-electrical (O/E) frequency response measurement, such as for photodetectors. Among those measurement techniques the optical methods are simple and efficient for measuring pigtailed discrete devices. However, in the case of wafer characterization, the required fiber-to-waveguide optical coupling brings in extra optical loss, and causes unstable and time-consuming measurements. The optical spectrum analysis method for E/O frequency response measurement requires the use of an optical spectrum analyzer (OSA). However, the grating-based OSA has a limited optical resolution of about 1 GHz at around 1550 nm [7,9]. Even though a Brillouin-based or heterodyne-based OSA could improve the resolution up to tens of MHz, this approach cannot exclude the line-width influence from the optical source [21]. For O/E characterization with optical methods, the intensity noise method is based on the ultra-wideband amplified spontaneous emission beat noise, but it suffers from low frequency stability and poor signal-to-noise ratio [19]; the optical heterodyne method is based on wavelength beating of single or two tunable lasers, in which the beating line-width and fluctuated power limits the measurement resolution [22]. Normally, the measurement resolution depends on the line-width of the tunable lasers, otherwise complicated feedback will be required to guarantee phase locking between the two heterodyne lasers.
Alternatively, efforts have been consequently directed to performing the RF characteristic in electrical domain, benefiting from the ultra-narrow line-width of RF sources as well as hyper-fine resolution of electrical spectrum analyzing systems. As is known, the swept-frequency method is widely adopted for measuring the frequency response of optoelectronic devices with a vector network analyzer (VNA) combined with an O/E or E/O transducer which converts optical into electrical measurements [23]. This method requires either intense calibrations to remove the impact of transducers in the measured frequency response, or expensive equipment, such as a light-wave component analyzer (LCA) with built-in calibrated E/O and O/E transducers [11, 23]. Furthermore, the photonic integrated circuit has to be broken down to discrete devices to be probed individually with transducers. Optical fiber couplings are still required in this procedure, as shown in Fig. 1(a) and 1(b). Recently, a self-calibrated method in electrical domain was developed to characterize high-speed optoelectronic devices by a frequency-shifted heterodyne method, which is free of extra O/E or E/O calibrations [24–27]. As shown in Fig. 1(c), this method enables a pure electrical measurement to extract frequency responses of both the modulator and photodetector assistedby an acousto-optic frequency shifter with same setup. However, an efficient frequency shifter is still in infancy to date with silicon technology [28].
Fig. 1 Schematic diagram of the conventional discrete measurement with swept-frequency method for measuring frequency response of (a) modulators and (b) photodetectors, and (c) the self-calibration measurement with frequency-shifting heterodyne method. VNA: Vector network analyzer; LCA: Lightwave component analyzer; MS: Microwave source; ESA: Electrical spectrum analyzer; LD: Laser diode; PC: Polarization controller; MOD: Modulator; PD: Photodetector; FS: Frequency shifter; DUT: Device under test.
In order to achieve an efficient and automatic on-wafer testing, we propose in this work a heterodyne-mixing based probing kit for self-calibrated RF characterization of broadband modulators and photodetectors in integrated transceivers. As is shown in Fig. 2(a), the on-wafer probing kit consists of an optical stimulus, an optical modulator and a photodiode cascaded in serial. Electrical driving signals are applied on the optical stimulus and modulator, respectively. The twice-modulated optical sidebands heterodyne with each other and generate the desired mixing products at the output of photodetector, allowing extraction of frequency responses of modulator and photodetector without any extra O/E or E/O calibration. The on-wafer probing kit can be operated with minimized or even free of fiber coupling off chip, benefiting from the pure electrical measurements. In this paper, the operation principle is elaborated, and proof-of-concept probing kit devices are designed and fabricated with heterogeneous integration technology. The on-wafer RF characterization is experimentally demonstrated for a photonic integration transceiver circuit, and comparatively measured with the conventional swept-frequency method to check the accuracy.
Fig. 2 Schematic diagram for self-calibrated RF measurement of (a) modulators and (b) photodetectors with heterodyne mixing method, and (c) typical designs of the on-wafer probing kit. VNA: Vector network analyzer; MS: Microwave source; OS: Optical stimulus; DML: Directly modulated laser; LD: Laser diode; MOD: Modulator; PD: Photodetector; PC: Polarization controller; DUT: Device under test. On-wafer probing kit can be configured with on-chip optical (c-I) modulated and (c-II) CW source, or off-chip optical (c-III) modulated and (c-IV) CW source.
2. Probing kit design and principle
In Fig. 2(a) and 2(b), the proposed probing kit consists of an optical stimulus, an optical modulator and a photodetector, accompanying the photonic transceiver circuit on the same wafer. In practice, the probing kit can be designed with either on-chip or off-chip optical stimulus, depending on whether the optical source can be integrated with the photonic transceiver or not. Figure 2(c) shows four typical probing kit designs, in which the first two are designed for on-wafer measurement of modulator and photodetector with on-chip modulated (Type I) or CW (Type II) optical source, while the last two are designed for on-wafer measurement with off-chip modulated (Type III) or CW (Type IV) optical source. In the cases of off-chip optical source, the probing kit is terminated with a grating- or edge- coupler for fiber coupling. As the optical stimulus provides an optical modulation signal, the assistant modulator (MOD1 in Type II and IV kit) can be either an EAM or Mach-Zehnder modulator (MZM), which is recommended to be biased at linear operation range.
As shown in Fig. 2(a) and 2(b), the on-wafer probing kit are driven by two microwave signals v0(t) = V0cos(2πf0t + φ0) and v1(t) = V1cos(2πf1t + φ1) at the optical stimulus and the modulator, respectively, and the twice modulated optical signal is then sent to the photodetector to generate an instantaneous photocurrent, which can be written as [21, 25]
with the power P0 and modulation depth m0 of optical stimulus, the modulation depth m1 of modulator, and the responsivity R of photodetector. From Eq. (1), there will be several electrical components generated by the heterodyne mixing, and it is easy to quantify these components in the frequency domain as followingIn the case of modulator under test shown in Fig. 2(a), two microwave signals are set with about half frequency relationship, i.e. f1≈2f0, so that R(f0)≈R(f1-f0) is satisfied. Thus, the modulation depths can be extracted from the output mixing products, given by
andIn the case of photodetector under test shown in Fig. 2(b), two microwave signals are chosen with close frequency relationship, i.e. f1≈f0, so that the lowest frequency f1-f0 is fixed and close to DC, and the sum-and-difference frequency sidebands at f1 ± f0 are investigated. As we know, each pair of optical sideband will keep equalized in the optical domain, and their amplitude difference in the electrical domain only depends on the frequency response of photodetector [26], which can be seen from the common factor P0m0(f0)m1(f1) in Eq. (2b). Therefore, the relative responsivity Rf of photodetector can be determined from the relative amplitude at f1 + f0 with respect to that at f1-f0 by
From Eqs. (3a), (3b) and (4), frequency responses of the modulator and photodetector can be selectively extracted by electrically analyzing the mixing products, and the influence from other than test device including the optical stimulus is cancelled out by carefully choosing the two microwave driving frequencies. The proposed method makes the best use of pure electrical measurements and eliminates any extra O/E or E/O calibration, as compared to the swept-frequency method. It is worth noticing that the frequency response of modulator at f1 is extracted from the electrical spectra components at around f1/2 (f0≈f1-f0≈f1/2), verifying the halved bandwidth requirement for the photodetector, while the frequency response of photodetector at about f1 + f0 (f1≈f0) is obtained with two driving microwave signals at f0 and f1, proving the doubled measuring frequency range for the modulator. Besides, the phases (φ0, φ1) of the driving microwave signals (at f0, f1) will not affect the spectrum amplitudes of the desired mixing products (at f0, f1 ± f0), so it is not necessary to keep the two microwave signals synchronized, which brings ease to the measurement.
3. Device design and fabrications
In our demonstration, we specially choose the third type probing kit (Type III in Fig. 2(c)) consisting of an on-chip optical modulator and a photodetector, in order to make a comparative characterization between on-wafer measurement with the self-calibration method and discrete measurement with conventional swept-frequency method. Since without on-chip optical source, a laser at 1550.92 nm wavelength with modulation signal is used as optical stimulus and coupled with the silicon photonic chip through an optical mode convertor on silicon waveguide. The modulator under test in this work is a heterogeneous EAM with 100 um cavity length; the photodetector under test is PIN type with 100 um long cavity. The probing devices locate at multiple places on the photonic transceiver die for a better representativeness and uniformity for the actually used devices, as shown in Fig. 3.
Fig. 3 (a) Diagram of integrated transceiver and (b) its on-wafer probing kit, where (c) is the scanning electron microscope images of EAM and PD after the device mesa etch and bottom electrode metal depositions and (d) is the photo of one transceiver die after dicing the chip.
4. Characterizations with probing kit
In the on-wafer measurement, the optical stimulus comes from an off-chip modulated optical source driven by a microwave source (MS, Agilent 8257d), and the probing kit is electrically driven at the modulator and collected at the photodetector by a VNA (Agilent E8361), respectively. The VNA is operated at frequency-offset mode, which allows independently setting the built-in source (port 1) and receiver (port 2) frequencies for characterizing conversion loss and reflection coefficient of two electrical ports [23]. Note that a four-port VNA can also be qualified for the measurement, where port 3 works as the extra MS. The conversion loss defines the conversion efficiency from the input microwave signal at f1 to the output mixing signals at f0 or f1+/−f0. With the scalar-mixer calibration (SMC) [29], the VNA is electrically calibrated to two coaxial ports to generate a RF input signal (leveled at 0 dBm) for the modulator and to analyze the received mixing signal from the photodetector. Two microwave probes (Cascade ACP40-GSG-150) are used for port extension from coaxial cables to coplanar electrodes. The effect of the probes is automatically de-embeded by the built-in two-tier calibration procedure (Agilent AdapterChar macro).
In order to characterize the modulator, two microwave signals are set with the relationship of, e.g. f1 = 2f0 + 0.004 GHz, while the mixing products from the photodetector are collected and analyzed at f0 and f1-f0. Figure 4(a) and 4(b) show the measured conversion loss at the desired frequencies. For example, the conversion loss from 10.102 GHz (f1) to 5.049 GHz (f0) and 5.053 GHz (f1-f0), are measured to be −41.18 dB and −72.09 dB, respectively, from which the ratio is determined to be −30.91 dB (−72.09 + 41.18). Therefore, the modulation depth of the modulator at 10.102 GHz (f1) is solved to be 0.057 (m1) or −24.91 dB (−30.91 + 6, 20*Log10 m1) under the RF driving power of 0 dBm. Note that the modulation depth at 10.102 GHz is extracted from the two electrical components at 5.049 and 5.053 GHz, verifying the halved bandwidth requirement for the photodetector. The conversion loss can be measured at other driving frequencies, from which frequency response of the modulator can be obtained as illustrated in Fig. 4(c). Moreover, the normalized frequency response can be easily obtained by normalizing the modulation depth to that of DC, which starts from 0 dB at DC. As we know, the modulation depths are more informative than a normalized frequency response, since it reflects not only the relative change of modulation efficiency but also the modulation efficiency itself [25, 27]. Note that the normalized frequency response can be obtained from the modulation depth under the same driving level. However, if only with the normalized frequency response, you can’t get the modulation depths (m1) [25], since the normalized frequency response just represents a relative change of modulation efficiency.
Fig. 4 Modulator characterization: measured conversion loss at frequencies (a) f0 and (b) f1- f0, and (c) extracted frequency response of modulator as a function of frequency.
In order to extract the photodetector characteristics, the two microwave signals are chosen with the relationship of, e.g. f1 = f0 + 0.1 GHz, while the mixing products from the photodetector is collected and analyzed at the desired frequencies f1 ± f0. Figure 5(a) and 5(b) show the conversion loss at the sum-and-difference frequencies. For instance, in the case of f1 = 5.101 GHz and f0 = 5.001 GHz, the conversion loss is measured to be −70.55 dB at 10.102 GHz (f1 + f0) and −62.77 dB at 100 MHz (f1-f0), respectively, from which the frequency response Rf of photodetector are determined to be −7.78 dB (−70.55 + 62.77) at 10.102 GHz with respect to 100 MHz. It is worth noticing that the frequency response at 10.102 GHz is obtained with two driving signals at 5.101 GHz and 5.001 GHz, verifying the doubled measuring frequency range for the modulator. The conversion loss is also measured at other driving frequencies, from which frequency response of the photodetector can be extracted, as shown in Fig. 5(c).
Fig. 5 Photodetector characterization: measured conversion loss at frequencies (a) f1-f0 and (b) f1 + f0, and (c) extracted frequency response of photodetector as a function of frequency.
From the two electrical measurements, the frequency response of the probing kit is estimated by combining the obtained frequency responses of modulator and photodetector. For comparison, the combined frequency response of the probing kit is also measured with the swept-frequency method under the same driving condition. In the swept-frequency measurements, the built-in source and receiver of VNA are swept at the same frequency [11]. The VNA (Agilent E8361A) is calibrated with the Short-Open-Load-Thru (SOLT) procedure and extended from two coaxial to coplanar ports [23]. Figure 6(c) illustrates both the combined frequency response from the self-calibration method and the measured one with the swept-frequency method, where the consistency proves the on-wafer probing kit and the self-calibrated measurement.
Fig. 6 Frequency responses of (a) probing kit, (b) modulator and (c) photodetector with the self-calibration method (blue line) and the swept-frequency method (red lines).
For a further verification, the modulator and the photodetector are separately characterized with the swept-frequency method by using the VNA (Agilent E8361A) plus a LCA (Agilent N4373C). In the modulator testing, the modulator is electrically driven by the VNA and optically collected by the off-chip O/E transducer of LCA. The optical input and output of modulator are coupled with two polarization maintaining (PM) lensed fibers via edge-couplers. In the case of photodetector testing, the photodetector is optically fed by the off-chip E/O transducer of LCA through a PM lensed fiber and the electrical signal is collected by the VNA. As shown in Fig. 6(b) and 6(c), the frequency responses of modulator and photodetector with and without LCA calibration are also illustrated for comparison. The measured results with the self-calibration method agree well with the calibrated results with the swept-frequency method, proving the selective on-wafer extraction of frequency responses of modulator and photodetector through the probing kit. Moreover, the reflection coefficients of the modulator and photodetector are simultaneously obtained and illustrated in Fig. 7(a) and 7(b) for reference, indicating a full characterization of frequency response and microwave impedance with the self-calibration method. The measured reflection coefficients reveal the mismatched impedance of devices under test, indicating that the self-calibration method is robust to impedance mismatch of the on-wafer devices. From the agreeable results, our measurement is ready for scattering matrix calculation of optoelectronic devices.
Fig. 7 Reflection coefficients of (a) modulator and (b) photodetector measured with the self-calibration method (blue lines) and the swept-frequency method (red lines).
5. Discussion and conclusion
In contrast to the conventional discrete measurement with swept-frequency method, the probing kit enables on-wafer and self-calibrated RF measurement for broadband optoelectronic devices integrated on chip. Unlike our earlier approach using the frequency-shifted heterodyne method, the probing kit is based on heterodyne mixing and free of acousto-optic frequency shifting, which allows on-wafer characterization for a photonic integration transceiver without adding more complicated devices.
We also compare the on-wafer and discrete measurements in terms of fiber coupling and calibration. The discrete measurement will require two fiber couplings for measuring a modulator and one for measuring a photodetector, as well as extra O/E and E/O calibrations. From our experiment, the on-wafer probing kit (Type III and IV) only requires one fiber-coupling for measuring frequency response of both modulator and photodetector with an off-chip optical stimulus. The probing kit (Type I and II) would enable fiber-coupling-free operation for on-wafer and self-calibration measurement of a modulator and a photodetector with an on-chip optical stimulus. It would be reasonable to expect that the reduced fiber coupling will improve the testing efficiency in terms of labor and time, and bring the testing accuracy by the enhanced signal power [30]. For example, the on-wafer testing makes vertical cavity surface-emitting lasers (VCSELs) more cost effective, even though the production process is more labor- or material- intensive, as compared to edge-emitting lasers. It also provides VCSELs with ease of large-scale integration in one-dimensional and two-dimensional arrays [31]. Besides, the measured frequency response of modulators and photodetectors also provides a reference to measuring other devices, like optical filter, MUX and DEMUX, in the integrated transceiver.
In conclusion, we have proposed an on-wafer probing kit and demonstrated a self-calibrated RF characterization of high-speed modulators and photodetectors in a photonic integration transceiver based on heterodyne mixing. A proof-of-concept probing kit was designed and fabricated with the heterogeneous integration of III-V materials on silicon, and the frequency responses of modulators and photodetectors in the photonic transceiver were on-wafer extracted without any extra O/E or E/O calibration. The on-wafer characterization with probing kits enables minimized or even free of fiber coupling for performance monitoring during chip fabrication, which is promising for more predictable yield before device packaging. The probing kit was demonstrated and verified with the silicon-based photonic integration transceiver, and it can also be extensively used for InP-based photonic integration circuits.
Funding
DARPA; National Natural Science Foundation of China (61377037, 61421002, and 61435010); Innovation Funds of Collaboration Innovation Center of Electronic Materials and Devices (ICEM2015-2001); Science Foundation for Youths of Sichuan Province (2016JQ0014); Fundamental Research Funds for the Central Universities (ZYGX2016J072).
References and links
1. Y. Arakawa, T. Nakamura, Y. Urino, and T. Fujita, “Silicon photonics for next generation system integration platform,” IEEE Commun. Mag. 51(3), 72–77 (2013). [CrossRef]
2. C. R. Doerr, “Silicon photonic integration in telecommunications,” Front. Phys.3(37), (2015). [CrossRef]
3. C. Sun, M. T. Wade, Y. Lee, J. S. Orcutt, L. Alloatti, M. S. Georgas, A. S. Waterman, J. M. Shainline, R. R. Avizienis, S. Lin, B. R. Moss, R. Kumar, F. Pavanello, A. H. Atabaki, H. M. Cook, A. J. Ou, J. C. Leu, Y.-H. Chen, K. Asanović, R. J. Ram, M. A. Popović, and V. M. Stojanović, “Single-chip microprocessor that communicates directly using light,” Nature 528(7583), 534–538 (2015). [CrossRef] [PubMed]
4. J. Sun, E. Timurdogan, A. Yaacobi, E. S. Hosseini, and M. R. Watts, “Large-scale nanophotonic phased array,” Nature 493(7431), 195–199 (2013). [CrossRef] [PubMed]
5. T. Komljenovic, M. Davenport, J. Hulme, A. Liu, C. Santis, A. Spott, S. Srinivasan, and C. Eric Stanton, “Zhang and J. E. Bowers, “Heterogeneous silicon photonic integrated circuits,” J. Lightwave Technol. 34(1), 20–35 (2016). [CrossRef]
6. C. Zhang, S. Zhang, J. D. Peters, and J. E. Bowers, “8 × 8 × 40 Gbps fully integrated silicon photonic network on chip,” Optica 3(7), 785–786 (2016). [CrossRef]
7. Y. Shi, L. Yan, and A. Willner, “High-speed electrooptic modulator characterization using optical spectrum analysis,” J. Lightwave Technol. 21(10), 2358–2367 (2003). [CrossRef]
8. N. Courjal, J. Dudley, and H. Porte, “Extinction-ratio-independent method for chirp measurements of Mach-Zehnder modulators,” Opt. Express 12(3), 442–448 (2004). [CrossRef] [PubMed]
9. S. Oikawa, T. Kawanishi, and M. Izutsu, “Measurement of chirp parameters and halfwave voltages of Mach-Zehnder-type optical modulators by using a small signal operation,” IEEE Photonics Technol. Lett. 15(5), 682–684 (2003). [CrossRef]
10. H. Kim and A. H. Gnauck, “Chirp characteristics of dual-drive Mach-Zehnder modulator with a finite DC extinction ratio,” IEEE Photonics Technol. Lett. 14(3), 298–300 (2002). [CrossRef]
11. P. D. Hale and D. F. Williams, “Calibrated measurement of optoelectronic frequency response,” IEEE Trans. Microw. Theory Tech. 51(4), 1422–1429 (2003). [CrossRef]
12. X. M. Wu, J. W. Man, L. Xie, Y. Liu, X. Q. Qi, L. X. Wang, J. G. Liu, and N. H. Zhu, “Novel method for frequency response measurement of optoelectronic devices,” IEEE Photonics Technol. Lett. 24(7), 575–577 (2012). [CrossRef]
13. A. A. Chtcherbakov, R. J. Kisch, J. D. Bull, and N. A. F. Jaeger, “Optical heterodyne method for amplitude and phase response measurements for ultrawideband electrooptic modulators,” IEEE Photonics Technol. Lett. 19(1), 18–20 (2007). [CrossRef]
14. E. Eichen, J. Schlafer, W. Rideout, and J. McCabe, “Wide-bandwidth receiver/photodetector frequency response measurements using amplified spontaneous emission from a semiconductor optical amplifier,” J. Lightwave Technol. 8(6), 912–916 (1990). [CrossRef]
15. D. M. Baney, W. V. Sorin, and S. A. Newton, “High-frequency photodiode characterization using a filtered intensity noise technique,” IEEE Photonics Technol. Lett. 6(10), 1258–1260 (1994). [CrossRef]
16. T. S. Tan, R. L. Jungerman, and S. S. Elliott, “Optical receiver and modulator frequency response measurement with a Nd:YAG ring laser heterodyne technique,” IEEE Trans. Microw. Theory Tech. 37(8), 1217–1222 (1989). [CrossRef]
17. S. Hou, R. S. Tucker, and T. L. Koch, “High-speed photodetector characterization by delayed self-heterodyne method,” Electron. Lett. 25(24), 1632–1634 (1989). [CrossRef]
18. T. Dennis and P. D. Hale, “High-accuracy photoreceiver frequency response measurements at 1.55 µm by use of a heterodyne phase-locked loop,” Opt. Express 19(21), 20103–20114 (2011). [CrossRef] [PubMed]
19. N. H. Zhu, J. M. Wen, H. S. San, H. P. Huang, L. J. Zhao, and W. Wang, “Improved optical heterodyne methods for measuring frequency response of photodetectors,” IEEE J. Quantum Electron. 42(3), 241–248 (2006). [CrossRef]
20. B. H. Zhang, N. H. Zhu, W. Han, J. H. Ke, H. G. Zhang, M. Ren, W. Li, and L. Xie, “Development of swept frequency method for measuring frequency response of photodetectors based on harmonic analysis,” IEEE Photonics Technol. Lett. 21(7), 459–461 (2009). [CrossRef]
21. M. Yoshioka, S. Sato, and T. Kikuchi, “A method for measuring the frequency response of photodetector modules using twice-modulated light,” J. Lightwave Technol. 23(6), 2112–2117 (2005). [CrossRef]
22. K. Inagaki, T. Kawanishi, and M. Lzutsu, “Optoelectronic frequency response measurement of photodiodes by using high-extinction ratio optical modulator,” IEICE Electron. Express 9(4), 220–226 (2012). [CrossRef]
23. S. Iezekiel, “Measurement of microwave behavior of optical links,” IEEE Microw. Mag. 9(3), 100–120 (2008). [CrossRef]
24. S. Zhang, H. Wang, X. Zou, Y. Zhang, R. Lu, and Y. Liu, “Self-calibrating measurement of high-speed electro-optic phase modulators based on two-tone modulation,” Opt. Lett. 39(12), 3504–3507 (2014). [CrossRef] [PubMed]
25. S. Zhang, X. Zou, H. Wang, Y. Zhang, R. Lu, and Y. Liu, “Calibration-free absolute frequency response measurement of directly modulated lasers based on additional modulation,” Opt. Lett. 40(20), 4727–4730 (2015). [CrossRef] [PubMed]
26. S. J. Zhang, H. Wang, X. H. Zou, Y. L. Zhang, R. G. Lu, H. P. Li, and Y. Liu, “Optical frequency-detuned heterodyne for self-referenced measurement of photodetectors,” IEEE Photonics Technol. Lett. 27(9), 1014–1017 (2015). [CrossRef]
27. S. J. Zhang, C. Zhang, H. Wang, X. Zou, Y. Zhang, Y. Liu, and J. Bowers, “Self-calibrated microwave characterization of high-speed optoelectronic devices by heterodyne spectrum mapping,” J. Lightwave Technol. 35(10), 1952–1961 (2017). [CrossRef]
28. T. Spuesens, Y. Li, P. Verheyen, G. Lepage, S. Balakrishnan, P. Absil, and R. G. F. Baets, “Integrated Optical Frequency Shifter on a Silicon Platform,” in Conference on Lasers and Electro-Optics, OSA Technical Digest (2016) (Optical Society of America, 2016), paper SF2G.1. [CrossRef]
29. Keysight Technologies, “Improving network analyzer measurements of frequency-translating device, Application note,” literature.cdn.keysight.com/litweb/pdf/5966–3318E.pdf.
30. St. Grillanda, M. Carminati, F. Morichetti, P. Ciccarella, A. Annoni, G. Ferrari, M. Strain, M. Sorel, M. Sampietro, and A. Melloni, “Non-invasive monitoring and control in silicon photonics using CMOS integrated electronics,” Optica 1(3), 129–136 (2014). [CrossRef]
31. E. Lannone, Telecommunication Networks (CRC, Taylor & Francis Group, 2012).
