Open Access
Add to BibSonomy
Abstract
An InGaAs/InAlAs multiple-quantum-well (MQW) optical phase modulator integrated with a planar antenna for a millimeter-wave (MMW) radio-over-fiber (RoF) system is fabricated, and its high-speed modulation under irradiation of MMW signals is experimentally demonstrated. The modulator exhibits a carrier-to-sideband ratio (CSR) of 62.7 dB, corresponding to a phase shift Δϕ of 1.46 mrad under irradiation of MMW signals with an RF power density P of ∼77 W/m2 at a frequency of 57.5 GHz. The modulator is operated over the C band in optical communications. The modulation operation was attributed to the large change in refractive index caused by the quantum-confined Stark effect in the MQW. This is the first case where the refractive index change of a semiconductor was used as an antenna-coupled optical modulator.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Millimeter-wave (MMW) radio-over-fiber (RoF) technology has attracted increasing attention for its applications in, for example, fifth-generation mobile communications (5G) [1–3], sensor-over-fiber (SoF) systems [4], and networks in dense user environments [5]. In particular, an MMW at 60-GHz-band frequencies has been attractive for wireless communications because of abundant frequency resources in various countries and its advantages in high-speed communications as is clear from Shannon’s law [6,7]. In RoF technology, optical modulators are required to convert wireless MMW signals to light-wave (LW) signals. A typical optical modulator is driven by inputting wireless RF signals received by remote antennas to the modulator using a coaxial cable. However, wireless RF signal transmission using a coaxial cable incurs a large loss, and RF signals may be distorted or delayed during their transmission. Moreover, because coaxial cables are heavy and bulky, the RoF system may become complex.
An optical modulator integrated with antennas is promising for MMW RoF systems. As antenna-coupled modulators, electrooptic (EO) materials such as lithium niobate (LN) [8–11] and nonlinear optical polymer (EOP) [12–15]-based modulators fabricated on the basis of the Pockels effect have been reported. The modulators enable us to directly convert wireless MMW signals to LW signals by using an electric field induced in integrated antennas without using any connection cables and external power supplies. In addition, demodulation of wireless MMW signals with phase-shift keying (PSK) and quadrature amplitude modulation (QAM) has been demonstrated by using the modulators [11,14]. However, the EO materials such as LN and EOP have relatively small EO coefficients, and there is a limit for reducing the device size and power consumption. Although antenna-coupled modulators based on EOP materials and plasmonic waveguides that significantly reduce their size and power consumption have been developed [16,17], the propagation loss of the modulators is exceedingly large (> 30 dB/cm) and high optical power is required for driving. Furthermore, previous antenna-coupled modulators have an issue in integration with III-V compound semiconductor light sources such as laser diodes (LDs) and light-emitting diodes (LEDs) [18,19], which results in high coupling loss between the modulators and LDs and low productivity when being integrated in an RoF system. To resolve these issues, we have proposed and theoretically discussed a multiple-quantum-well (MQW) optical Mach–Zehnder modulator integrated with antennas [20]. The InP-based modulator can be fabricated monolithically with LDs and LEDs, which leads to high productivity and reduced losses of the modulator systems. In addition, if a novel MQW, a five-layer asymmetric quantum well (FACQW), exhibiting a large electrorefractive change Δn is adopted in the core layer of the waveguide, the modulator is expected to perform with higher modulation efficiency than those in previous research.
In this paper, an InGaAs/InAlAs MQW optical phase modulator integrated with an antenna operating at 60-GHz-band frequencies is experimentally demonstrated. In the core layer of the waveguide, an FACQW exhibiting a large Δn owing to its unique quantum-confined Stark effect (QCSE) was adopted [21–23]. The fabricated modulator exhibited a carrier-to-sideband ratio (CSR) of 62.7 dB corresponding to a phase shift Δϕ of 1.46 mrad under irradiation of wireless MMW signals with the RF power density P of ∼77 W/m2 at the frequency of 57.5 GHz. In addition, the modulator exhibited little wavelength dependence and can be operated in a broad C band of optical communications. By comparing Δϕ obtained from the measured CSR with that calculated from Δn of the FACQW, we demonstrated that the modulation operation was due to the QCSE. This is the first case where the refractive index change of a semiconductor was used as an antenna-coupled optical modulator. We also described the design, simulation, and fabrication of the InGaAs/InAlAs MQW antenna-coupled optical phase modulator.
2. Device structure and simulation of millimeter-wave characteristics
In this section, we describe the modulator structure, design, and simulation. The refractive index change characteristics of the FACQW used in the modulator are presented. In addition, we explain the derivation of the optical phase modulation under the irradiation of wireless MMW signals.
2.1 Device structure
Figure 1 shows a schematic view of the antenna-coupled modulator based on a conductive n-type InP (n-InP) substrate. The modulator consists of an optical phase modulator and a gap-embedded patch antenna placed on the modulator. As the most specific feature of the modulator, a deep trench with a thickness in the range of less than 20 µm is formed in the n-InP substrate, and SU8, which is a negative-type UV photoresist, is embedded in the trench. The embedded SU8 and the n-InP substrate function as an effective antenna substrate and the ground, respectively. This is a special design for fabricating the modulator using the n-InP substrate since an antenna does not work if formed directly on a conductive substrate. For an optimal design, we proposed an antenna-coupled modulator based on a semi-insulating Fe-doped InP (Fe-InP) substrate [20]. However, since we were unable to prepare the Fe-InP substrate with the FACQW core layer, the modulator based on the n-InP substrate with the unique structure was proposed. On the top of the modulator, a glass substrate is attached using a UV-curable resin (NOA81) to reduce the effective dielectric constant of the modulator ɛeff, resulting in an increase in the electric field induced in the antenna [9]. The waveguides are located on a waveguide base placed at the center of the deep trench, whose structure is designed such that the induced electric field can be effectively applied to the core layer. The waveguides are buried with benzocyclobutene (BCB) to enhance the electric field applied to the core layer. At the edges of the waveguide base, there are “base edges,” which are a by-product in the device fabrication and have the same cross-sectional structure as the actual waveguide. A SiO2 film is formed as a buffer layer between the AlSi antenna and the waveguide. The buffer layer suppresses leakage currents and prevents the formation of a Schottky junction at the interface between AlSi and the waveguide.
2.2 MQW core layer
In the modulator, the InGaAs/InAlAs FACQW exhibiting a large Δn owing to the QCSE was adopted as a waveguide core layer. Figures 2(a) and (b) show a schematic view of the cross-sectional waveguide including the FACQW core layer and the band profile of the FACQW, respectively. The operational principle of the FACQW has been summarized in detail in Ref. 21. It has been reported that nonuniformity of the electric field in the FACQW occurs owing to residual carriers in the FACQW [22]. Since the FACQW is significantly sensitive to the applied electric field and its electrorefractive index change characteristics are affected, it is essential to design the structure of the core layer considering the nonuniformity to obtain a large Δn. Therefore, in the modulator, 12-period undoped In0.53Ga0.47As/In0.52Al0.48As FACQWs with six different structures were adopted as the waveguide core layer. In the combined structure, each FACQW is optimized to exhibit almost the largest Δn according to the electric field range at each position. The design concept has been summarized in Ref. 22.
Figure 3(a) shows the electrorefractive index change characteristics of the FACQW. The positive direction of the applied electric field F is defined as illustrated in Fig. 2(b). First, Δn of the FACQW as a function of voltage was obtained from the free spectrum range (FSR) of a ring resonator with the same waveguide as that shown in Fig. 2(a) under an applied voltage from 0 to −15 V. The derivation of Δn has been summarized in Ref. 23. Second, Δn was divided by the thickness of the undoped layer in the waveguide and it was converted into a function of F. The FACQW used in this research was designed considering the nonuniformity of the electric field in the core layer, and it exhibits a larger Δn than the FACQW designed without considering the nonuniformity of the electric field [23,24]. In addition, Δn of FACQWs varies almost linearly at an electric field close to 0 kV/cm, as shown in Fig. 3(a). In the antenna-coupled modulator, the applied electric field F was estimated to be at most ∓0.1 kV/cm in modulation measurement. Therefore, to calculate Δϕ from the electrorefractive index change characteristics of the FACQW and compare it with measurement results, the linear approximation of Δn of the FACQW in the electric field range from 0 to −26 kV/cm was used, as shown in Fig. 3(b). The coefficient of the electrorefractive sensitivity rFACQW is as large as 19.4 × 10−12 m/V, and Δn calculated from rFACQW is comparable to that of z-cut LN with the EO coefficient r33 = 31 × 10−12 m/V [25].
Fig. 3. (a) Electrorefractive index change Δn of FACQW and (b) linear approximation of Δn as a function of the applied electric field F for TE mode at the wavelength λ of 1550 nm.
2.3 Antenna design and simulated MMW characteristics
When the wireless MMW signals with x-polarization are incident on a modulator integrated with a gap-embedded antenna, a displacement current is induced across the gap owing to the continuity of the current flow [26]. Consequently, large electric fields in both positive and negative directions are localized in the gap. Considering the irradiation of wireless MMW signals at an incident angle θ, the RF electric field induced in the antenna can be expressed as
A schematic view of the InP-based antenna-coupled modulator is shown in Fig. 1. In the modulator, a gap-embedded patch antenna was adopted as an integrated planar antenna. The patch antenna has a simple structure and high gain, and antenna-coupled modulators integrated with the antenna structure have been reported [9,13,20]. The length of the patch antenna La is given by [27]
Figure 4 shows the calculated distribution of Ez in the cross-sectional waveguide base of the modulator. In the case of using the waveguide including a PIN structure, where a large Ez in both positive and negative directions was intensively and simultaneously applied to the FACQW core layers. Ez was also applied to the FACQW core layer of the base edges, and it was found that the magnitude of Ez applied to the waveguide core layer was reduced. On the other hand, the result shows that Ez is applied to multiple waveguides simultaneously, which indicates that the modulator is applicable to wavelength division multiplexing (WDM) communication technology.
Fig. 4. Calculated distribution of the z-component of the electric field Ez in the cross-sectional waveguide base of the modulator.
Figure 5(a) shows Ez/E0 and the peak frequency fp as functions of hbase. In the calculation, Wedge was set to 2 µm. The magnitude of Ez was calculated by averaging the z-component of the electric field applied to the entire FACQW core layer (1.45 µm × 0.3 µm × 700 µm). fp was defined as the point where Ez/E0 reaches the maximum. Ez/E0 tends to increase with increasing hbase, which results from an increase in the thickness of SU8, which is the effective antenna substrate, leading to improved antenna performance. Therefore, the larger the value of hbase, the higher Ez/E0 is, which provides a high modulation efficiency. In the case of using the fabrication process for deep InP etching discussed in Section 3, we can obtain hbase up to approximately 20 µm. fp tends to decrease with increasing hbase, since λRF increases owing to the reduction in ɛeff caused by the embedded SU8 in accordance with Eq. (10). Figure 5(b) shows the calculated Ez/E0 and fp as functions of Wedge. In the calculation, hbase was set to 17 µm. Ez/E0 significantly decreases with increasing Wedge. The base edge has the same cross-sectional structure as the waveguide, and the electric fields induced in the antenna are also applied to the base edge core layer, as shown in Fig. 4. The magnitude of Ez applied to the base edge core layer increases with increasing Wedge, that is, with increasing the contact area between the edge and the antenna, which causes the reduction in the magnitude of Ez applied to the core layer. Therefore, Wedge must be as small as possible to obtain a high Ez/E0. However, since the base edge is a by-product of the waveguide base, it cannot be removed as long as the above fabrication process is used. fp decrease with increasing Wedge, since ɛeff decreases owing to the reduction in the electric field induced in the antenna gap, and λRF increases in accordance with Eq. (10). In the modulator fabrication, hbase and Wedge were set to 17 µm and 5 µm, respectively. Other parameters for the modulator are shown in Fig. 1. In this research, the highest priority was given to the fabrication and verification of the InP-based antenna-coupled modulator; therefore, a device with a large margin was fabricated.
Fig. 5. Calculated electric field enhancement Ez/E0 and the peak frequency fp as functions of (a) the height of the waveguide base hbase and (b) the width of the base edge Wedge.
3. Device fabrication
A 2-inch n-InP wafer with the FACQW waveguide shown in Fig. 2(a) was cut to a chip with a size of 1×1 cm2. Then, waveguides were fabricated on the chip by electron-beam (EB) lithography and inductively coupled plasma (ICP) reactive ion etching (RIE), and were buried with BCB. Then, a deep trench was fabricated, as shown in Fig. 6. First, a SiO2 film with a thickness of approximately 1.25 µm was sputtered on the InP chip as a mask. A second EB lithography was then performed to form a trench-patterned EB resist on the chip, and the SiO2 mask was wet-etched with a buffered hydrofluoric (BHF) acid solution. Since the SiO2 mask was side-etched in the wet etching, the EB resist was patterned with a wide margin. InP was then deeply etched by RIE to fabricate the deep trench and the waveguide base. The waveguide base was shaped into a trapezoidal structure according to the side-etched SiO2 mask pattern, which showed that the SiO2 pattern was faithfully transferred to the InP chip. SU8 was uniformly embedded in the trench by spin coating and then etched by RIE to expose the top of the waveguide and flatten the chip surface. To pattern the EB resist for an antenna, a third EB lithography was performed and the antenna was formed using a general lift-off process. Finally, the processed InP chip was diced for each modulator and a glass substrate was attached using NOA81.
Figures 7(a) and (b) show micrographs of the fabricated modulator. The modulator with dimensions of 3×10 mm2 was fabricated almost as designed. The antenna gap was formed with a width of 4.3 µm, which is less than the designed gap width of 4.6 µm shown in Fig. 1(b). However, since an electric field induced in an antenna increases as the gap width decreases [9,16], the modulator performance was improved by this increase. In the modulation experiment described in Section 4.2, the modulator was measured and evaluated.
Figure 8 shows a cross-sectional scanning electron microscopy (SEM) image of the antenna-coupled modulator without the glass substrate, which was fabricated together in the same chip as the modulator shown in Fig. 7. The waveguide base with a height of 17 µm was formed with relatively high perpendicularity using the fabrication process shown in Fig. 6. The upper and lower widths of the base differ by approximately 6 µm owing to the side etching in the patterning of the SiO2 mask. The thickness of the SU8 embedded in the trench is approximately 14 µm, which is 3 µm smaller than that of the waveguide base. This decrease in thickness could be caused by overetching of the embedded SU8 owing to the variation of the etching rate with the position in the InP chip. Owing to the decrease, the contact length between the base edge and the antenna increased from the designed value of 5 µm to 7 µm, resulting in the increase in the contact area and the reduction in Ez/E0. In addition, the antenna position was shifted by approximately 0.4 µm from the designed position. However, its effect on the modulation characteristics is expected to be small since both antenna edges are placed on the waveguides and the induced electric field can be applied to the core layer. The CSR and Δϕ shown in Section 4.2 were calculated from the MMW characteristics analyzed using the FEM simulator with a 3D-CAD modulator model to which these results are fed back.
4. Measurement of modulation characteristics
We evaluated the phase modulation characteristics of the fabricated modulator by measuring its CSR under the irradiation of MMW signals.
4.1 Carrier-to-sideband ratio (CSR)
The modulation characteristics of an antenna-coupled phase modulator can be evaluated on the basis of CSR. Here, the phase-modulated optical signal fPM(t) is assumed to be [29]
4.2 Modulation characteristics
Figure 9 shows the experimental setup for measuring the modulation operation of the fabricated antenna-coupled modulator. In the electrical setup, the LW signal with the 60-GHz-band frequencies is generated by optical two-tone technique [30], and the LW signal is converted into the RF signal using a uni-traveling carrier photodiode (UTC-PD) (Finisar, XPDV2120R). The RF signal is then input to a horn antenna (Flann Microwave, 25240-20) with an antenna gain of ∼20 dBi via an amplifier (CERNEX, CBL57653055), and the wireless 60-GHz-band MMW signal is incident on the modulator placed on a metal pedestal. The vibration direction of the electric field of the MMW signal is shown in Fig. 9. The transmitter power of the horn antenna Pt is approximately 16 dBm. In the optical setup, the LW with the TE mode and an optical power Pin of 15.5 dBm from a tunable laser (Alnair Labs, TLG-200) is input to the modulator via a polarization controller through a polarization-maintaining (PM) fiber. The LW passing thorough the modulator is then coupled to an optical fiber and input to the optical spectrum analyzer (Yokogawa, AQ6370B). When the modulator is irradiated with the wireless 60-GHz-band MMW signal from the horn antenna, the output including sidebands can be observed.
Figure 10(a) shows the optical spectrum of the antenna-coupled modulator under the irradiation of the wireless MMW signal at a frequency of 57.5 GHz. The separation s between the horn antenna and the modulator and the wavelength of the input LW signal were set to 18 mm and 1550 nm, respectively. Under these conditions, the RF power density P of the MMW signal that is incident on the modulator was estimated as large as ∼77 W/m2. As shown in Fig. 10(a), sidebands were clearly confirmed and a CSR of 62.7 dB was observed, which corresponds to Δϕ of 1.46 mrad. The CSR obtained in the measurements is relatively large in comparison with those of LN- and EOP-based modulators integrated with gap-embedded patch antennas [9,13]. However, the result does not mean that the developed MQW-based modulator is inferior to the LN- and EOP-based modulators since the conductive n-InP substrate used in this research is not suitable for coupling antennas with the modulator. As discussed in Ref. 20, the MQW-based antenna-coupled optical modulator based on a semi-insulating Fe-InP substrate exhibits Ez/E0 of more than 800, which means CSR can be improved by more than 20 dB by using the Fe-InP substrate for the modulator. In addition, the FACQW used in this research is not optimized for driving in the electric field region close to 0 kV/cm. If an FACQW is designed so as to exhibit a large electric field induced change in refractive index in the low electric field region and adopted for the modulator, a smaller CSR can be achieved even with the same Ez/E0 as the developed modulator in this research. The intensities of the sidebands on the higher- and lower- frequency sides are almost the same, which results from a linear phase shift of the LW signal propagating the FACQW core layer. Namely, the induced RF electric fields with the same intensity in both positive and negative directions were applied to the core layer. Figure 10(b) shows the CSR as a function of the frequency of the wireless MMW signal. The smallest CSR, which means the highest m according to Eq. (15), was observed at the frequency of 57.5 GHz, which is almost the same as that 58 GHz for the calculated CSR. Note that the CSR was calculated from Γ=0.55 obtained by a beam-propagation method simulation, Ez/E0=54.3 obtained by the FEM simulation, and the effective refractive index of the FACQW of neff = 3.83. The measured CSR fluctuates even when changing the frequency by 0.5 GHz. Theoretically, the CSR varies continuously similar to the calculated CSR. However, in the modulation measurement, the distance between the horn antenna and the modulator was small, and the measurement might be performed under the near-field condition. Therefore, the reflection of the incident MMW signal by the metal pedestal of the modulator can significantly affect the CSR.
Fig. 10. (a) Optical spectrum of the modulator under irradiation of the wireless MMW signal at a frequency of 57.5 GHz. (b) CSR as a function of the frequency of the incident 60-GHz-band MMW signal.
Figure 11(a) shows the CSR and the transmittance of a carrier LW for the modulator as functions of the wavelength of the carrier LW. Note that s and the frequency of the wireless MMW were set to 18 mm and 57.5 GHz, respectively. The transmittance includes the insertion losses of the device and input and output waveguides including spot size converters, and fiber coupling losses at facets. The CSR tends to decrease with decreasing wavelength, which is consistent with the FACQW characteristic that Δn increases as the wavelength becomes closer to the absorption edge (∼1520 nm) [21]. In addition, according to Eq. (9), when the same electric field is applied to the FACQW core layer, Δϕ increases with decreasing the wavelength, resulting in a lower CSR (higher m). The difference between the maximal CSR at 1560 nm and the minimal CSR at 1530 nm is relatively small (∼1.2 dB) and not significant for the modulation operation. The transmittance also tends to decrease with decreasing wavelength since the absorption loss of the FACQW increases as the wavelength becomes closer to the absorption edge similar to Δn. However, the transmittance is large in the wavelength region of 1530–1560 nm and its effect on the modulation operation is very small. The result shows that the modulator can be operated in a broad range of wavelengths in the C band of optical communications. The insertion loss of the modulator was approximately 30 dB which includes the propagation loss in the FACQW waveguide of 5 dB (approximately 1.6 dB/mm [24]) and the coupling loss between the device and optical fibers of 13 dB/facet. Figure 11(b) shows CSR as a function of s. Note that the wavelength of the carrier LW and the frequency of the wireless MMW signal were set to 1550 nm and 57.5 GHz, respectively. The CSR increases almost linearly with increasing s; this result is very reasonable. Here, P from the horn antenna that is incident on the modulator can be expressed as [27]
where G is the gain of the horn antenna. According to Eq. (17), P is inversely proportional to the square of s. In addition, the square of m is proportional to P according to Eqs. (15) and (16). Therefore, m is inversely proportion to s, and CSR in dB is proportional to s in theory.Fig. 11. (a) CSR and transmittance as functions of the wavelength. (b) CSR as a function of separation s. Each broken line is a linear approximation drawn with reference to the plotted data.
Figure 12 shows Δϕ of the modulator as a function of P, which was estimated by using the designed parameters of the horn antenna and the results shown in Fig. 11(b). Δϕ due to the Pockels effect was calculated using the EO coefficient r41 = −1.69 × 10−12 m/V of the InGaAs/InAlAs MQW having the same material compositions as the FACQW [31]. Δϕ due to the QCSE and Pockels effect was calculated using rFACQW = 19.4 × 10−12 m/V from the linear approximation of Δn shown in Fig. 3(b). As shown in Fig. 12, Δϕ calculated from the CSR almost corresponds to that due to the QCSE and the Pockels effect of the FACQW. The result shows that the modulator is operated using Δn due to the QCSE. It is also obvious that the Δϕ calculated from the CSR is more than four times as large as that due to only the Pockels effect.
5. Conclusion
We demonstrated the fabrication of an InGaAs/InAlAs MQW optical phase modulator integrated with a planar antenna. Since the modulator can be fabricated monolithically with LDs and LEDs, low loss and high productivity of the modulator system can be achieved. In the core layer of the waveguide, an FACQW exhibiting a large Δn owing to its unique QCSE was used. The modulator exhibited a CSR of 62.7 dB under the irradiation of a wireless MMW signal with an RF power of 77 W/m2 at a frequency of 57.5 GHz. The CSR of 62.7 dB corresponds to Δϕ of 1.46 mrad. Although the CSR is relatively large, it can be reduced by more than 20 dB by using the semi-insulating substrate instead of the conductive substrate. The modulator can be operated over the C band of optical communications. The developed device fabrication process enables the design of an antenna-coupled optical modulator using various semiconductor substrates such as Si and GaAs. To improve the modulation efficiency, it is also effective to use an array of antennas [9,13,20]. Using an array antenna structure, we increased not only the accumulated magnitude of the electric field applied to the core layer by 3 dB, but also the signal-to-noise ratio (SNR) of the modulator because the LW that does not meet the phase-matching condition is canceled. To improve the SNR, it is also a promising approach to improve Pt of the horn antenna and Pin of the tunable laser used in the experimental setup. The modulation operation was attributed to the large Δn owing to the QCSE in the FACQW. This is the first case where the refractive index change of a semiconductor was used as an antenna-coupled optical modulator. We believe that the modulator is a promising candidate for MMW RoF technology in the future.
Funding
Ministry of Education, Culture, Sports, Science and Technology (18H01897).
Acknowledgments
The authors express sincere thanks to Professor Joo-Hyong Noh of Kanto Gakuin University for supporting the electromagnetic analysis using the FEM simulator. The authors also thank Advanced ICT Laboratory of the National Institute of Information and Communications Technology (NICT) for support in device fabrication. This work was partly supported by a Grant-in-Aid for Scientific Research B (No.18H01897) from the Ministry of Education, Culture, Sports, Science and Technology. A part of this work was conducted at Takeda Sentanchi Supercleanroom, The University of Tokyo, supported by “Nanotechnology Platform Program” of the MEXT, Japan, Grant Number JPMXP09F19UT0040.
Disclosures
The authors declare no conflicts of interest.
References
1. T. S. Rappaport, S. Sun, R. Mayzus, H. Zhao, Y. Azar, K. Wang, G. N. Wong, J. K. Schulz, M. Samimi, and F. Gutierrez, “Millimeter Wave Mobile Communications for 5G Cellular: It Will Work!” IEEE Access 1, 335–349 (2013). [CrossRef]
2. M. Agiwal, A. Roy, and N. Saxena, “Next Generation 5G Wireless Networks: A Comprehensive Survey,” IEEE Commun. Surv. Tutorials 18(3), 1617–1655 (2016). [CrossRef]
3. S. Li, L. D. Xu, and S. Zhao, “5G Internet of Things: A survey,” J. Ind. Inform. Integr. 10, 1–9 (2018). [CrossRef]
4. Y. Hu, Z. Zeng, J. Liu, J. Wang, and W. Zhang, “Design of a Distributed UHF Sensor Array System for PD Detection and Location in Substation,” IEEE Trans. Instrum. Meas. 68(6), 1844–1851, (2019). [CrossRef]
5. K. Ishihara, T. Murakami, H. Abeysekera, M. Akimoto, and Y. Takatori, “Distributed smart antenna system for high-density WLAN system,” Electron. Lett. 54(6), 336–338 (2018). [CrossRef]
6. G. H. Sim, A. Loch, A. Asadi, V. Mancuso, and J. Widmer, “5G Millimeter-Wave and D2D Symbiosis: 60 GHz for Proximity-Based Services,” IEEE Wireless Commun. 24(4), 140–145 (2017). [CrossRef]
7. Y. Ghasempour, C. R. C. M. da Silva, C. Cordeiro, and E. W. Knightly, “IEEE 802.11ay: Next-Generation 60 GHz Communication for 100 Gb/s Wi-Fi,” IEEE Commun. Mag. 55(12), 186–192 (2017). [CrossRef]
8. T. Meier, K. Kostrzcewa, B. Schuppert, and K. Petermann, “Electro-optical E-field sensor with optimised electrode structure,” Electron. Lett. 28(14), 1327–1329 (1992). [CrossRef]
9. Y. N. Wijayanto, H. Murata, and Y. Okamura, “Electrooptic Millimeter-Wave–Lightwave Signal Converters Suspended to Gap-Embedded Patch Antennas on Low-k Dielectric Materials,” IEEE J. Sel. Top. Quantum Electron. 19(6), 33–41 (2013). [CrossRef]
10. Y. Matsukawa, T. Inoue, H. Murata, and A. Sanada, “Millimeter-wave band optical single-sideband modulator using array-antenna-electrode and polarization-reversed structures with asymmetric Mach–Zehnder waveguide,” Jpn. J. Appl. Phys. 57(8S2), 08PB04 (2018). [CrossRef]
11. H. Murata, “Millimeter-Wave-Band Electro-Optic Modulators for Microwave Photonic Applications,” in 2019 MWP (2019), 1–4.
12. O. D. Herrera, K.-J. Kim, R. Voorakaranam, R. Himmelhuber, S. Wang, V. Demir, Q. Zhan, L. Li, R. A. Norwood, R. L. Nelson, J. Luo, A. K.-Y. Jen, and N. Peyghambarian, “Silica/Electro-Optic Polymer Optical Modulator With Integrated Antenna for Microwave Receiving,” J. Lightwave Technol. 32(20), 3861–3867 (2014). [CrossRef]
13. D. H. Park, V. R. Pagán, T. E. Murphy, J. Luo, A. K.-Y. Jen, and W. N. Herman, “Free space millimeter wave-coupled electro-optic high speed nonlinear polymer phase modulator with in-plane slotted patch antennas,” Opt. Express 23(7), 9464–9476 (2015). [CrossRef]
14. D. H. Park, V. R. Pagán, P. S. Cho, J. Luo, A. K.-Y. Jen, and P. V. Petruzzi, “RF photonic downconversion of vector modulated signals based on a millimeter-wave coupled electrooptic nonlinear polymer phase-modulator,” Opt. Express 25(24), 29885–29895 (2017). [CrossRef]
15. C.-J. Chung, X. Xu, Z. Pan, F. Mokhtari-Koushyar, R. Wang, H. Yan, H. Subbaraman, and R. T. Chen, “Silicon-Based Hybrid Integrated Photonic Chip for Ku band Electromagnetic Wave Sensing,” J. Lightwave Technol. 36(9), 1568–1575 (2018). [CrossRef]
16. Y. Salamin, W. Heni, C. Haffner, Y. Fedoryshyn, C. Hoessbacher, R. Bonjour, M. Zahner, D. Hillerkuss, P. Leuchtmann, D. L. Elder, L. R. Dalton, C. Hafner, and J. Leuthold, “Direct Conversion of Free Space Millimeter Waves to Optical Domain by Plasmonic Modulator Antenna,” Nano Lett. 15(12), 8342–8346 (2015). [CrossRef]
17. Y. Salamin, B. Baeuerle, W. Heni, F. C. Abrecht, A. Josten, Y. Fedoryshyn, C. Haffner, R. Bonjour, T. Watanabe, M. Burla, D. L. Elder, L. R. Dalton, and J. Leuthold, “Microwave plasmonic mixer in a transparent fibre–wireless link,” Nat. Photonics 12(12), 749–753 (2018). [CrossRef]
18. Y. Kawamura, K. Wakita, Y. Itaya, Y. Yoshikuni, and H. Asahi, “Monolithic integration of InGaAs/InP DFB lasers and InGaAs/InAlAs MQW optical modulators,” Electron. Lett. 22(5), 242 (1986). [CrossRef]
19. S. Lange, S. Wolf, J. Lutz, L. Altenhain, R. Schmid, R. Kaiser, M. Schell, C. Koos, and S. Randel, “100 GBd Intensity Modulation and Direct Detection With an InP-Based Monolithic DFB Laser Mach–Zehnder Modulator,” J. Lightwave Technol. 36(1), 97–102 (2018). [CrossRef]
20. Y. Miyazeki and T. Arakawa, “Proposal of InGaAs/InAlAs multiple quantum well Mach–Zehnder modulator integrated with array of planar antennas,” Jpn. J. Appl. Phys. 58(SJ), SJJE05 (2019). [CrossRef]
21. T. Arakawa, T. Toya, M. Ushigome, K. Yamaguchi, T. Ide, and K. Tada, “InGaAs/InAlAs Five-Layer Asymmetric Coupled Quantum Well Exhibiting Giant Electrorefractive Index Change,” Jpn. J. Appl. Phys. 50(3R), 032204 (2011). [CrossRef]
22. T. Arakawa, T. Hariki, Y. Amma, M. Fukuoka, M. Ushigome, and K. Tada, “Low-Voltage Mach–Zehnder Modulator with InGaAs/InAlAs Five-Layer Asymmetric Coupled Quantum Well,” Jpn. J. Appl. Phys. 51(4R), 042203 (2012). [CrossRef]
23. T. Makino, T. Gotoh, R. Hasegawa, T. Arakawa, and Y. Kokubun, “Microring Resonator Wavelength Tunable Filter Using Five-Layer Asymmetric Coupled Quantum Well,” J. Lightwave Technol. 29(16), 2387–2393 (2011). [CrossRef]
24. H. Kaneshige, R. Gautam, Y. Ueyama, R. Katouf, T. Arakawa, and Y. Kokubun, “Low-voltage quantum well microring-enhanced Mach-Zehnder modulator,” Opt. Express 21(14), 16888–16900 (2013). [CrossRef]
25. E. H. Turner, “High-frequency electro-optic coefficients of lithium niobate,” Appl. Phys. Lett. 8(11), 303–304 (1966). [CrossRef]
26. R. Rodriguez-Berral, F. Mesa, and D. R. Jackson, “Gap Discontinuity in Microstrip Lines: An Accurate Semianalytical Formulation,” IEEE Trans. Microwave Theory Tech. 59(6), 1441–1453 (2011). [CrossRef]
27. C. A. Balanis, Antenna Theory, 3rd ed. (Wiley-Interscience, 2005)
28. M. Born and E. Wolf, Principles of Optics, 7th ed. (Cambridge University, 1999).
29. J. D. Gibson, Principles of digital and analog communications, 2nd ed. (Macmillan, 1993).
30. D. E. Cooper and R. E. Warren, “Frequency modulation spectroscopy with lead–salt diode lasers: a comparison of single-tone and two-tone techniques,” Appl. Opt. 26(17), 3726–3732 (1987). [CrossRef]
31. S. Nishimura, H. Inoue, H. Sano, and K. Ishida, “Electrooptic effects in an InGaAs/InAlAs multiquantum well structure,” IEEE Photonics Technol. Lett. 4(10), 1123–1126 (1992). [CrossRef]
