Open Access
Add to BibSonomy
Abstract
We propose a system for the simultaneous measurements of Ka-band microwave angle of arrival (AOA) and Doppler frequency shift (DFS) based on a high-speed silicon dual-parallel Mach-Zehnder modulator (Si-DPMZM).. An echo signal drives a sub-MZM while the combination of a phase-delay echo signal and a transmitted signal drives the other sub-MZM. Two optical bandpass filters (OBPFs) are used to select the upper and lower sidebands of the Si-DPMZM output signal, detected by low-speed photodiodes, then generating two intermediate frequency (IF) signals. Thus, both AOA and DFS (with direction) can be obtained by comparing the powers, phases and frequencies of these IF signals. The estimation error of measured AOA is less than ±3° from 0 to ±90°. Meanwhile, the DFS at 30/40 GHz were measured with an estimated error of less than 9.8 × 10−10 Hz within ±1 MHz. In addition, the fluctuation of DFS measurement is less than 3 × 10−11 Hz within 120 minutes, indicating the high stability of the system.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Microwave angle of arrival (AOA) and Doppler frequency shift (DFS) are indispensable in navigation, wireless communication, radar and electronic warfare [1–3]. For instance, in radar system, identifying the position and speed of a target quickly and accurately is a fundamental task. The radar-detected speed can be calculated from the DFS of an echo signal, while the AOA is essential in obtaining the precise position of the target [3]. However, electrical approaches for measuring AOA and DFS are faced with bottlenecks of bandwidth, power consumption and electromagnetic interference [4,5]. Microwave photonics (MWP) is emerging as a promising and powerful solution to tackle these issues due to its advantages of ultra-broadband, low power consumption and strong anti-interference capability [6,7].
The primary method to obtain AOA based on MWP is to measure the phase difference [8–13] or time delay [14,15] between two received echo signals. For DFS measurement, the absolute value of DFS is the frequency of intermediate frequency (IF) signal generated by the mixing of an echo signal and a transmitted signal in the optical domain. The direction of DFS can be obtained by comparing the phase relationship of IF signals in two channels [16–20] or using reference signals [21–23]. The aforementioned schemes can only measure DFS or AOA. However, in most situation, it is highly desirable to have a single photonic system that can simultaneously measure the DFS and the AOA [24–27]. In [24], a dual-polarization 90° optical hybrid coupler is used to build two I/Q (in-phase/quadrature) channels for measuring AOA and DFS simultaneously. In [26], a dual-parallel dual-drive Mach-Zehnder modulator, with an additional phase shifter and a signal generator for generating multiple tones, is employed to achieve the same goal.
However, the above-mentioned systems are all built on LiNbO3 modulators, and an integrated solution is superior with compact size, low cost and flexibility. Compared with other integration methods, silicon photonics are more attractive due to their CMOS compatibility, high integration density [28] and potential for seamless integration with electronics [29,30]. Recently, a simultaneous measurement system of AOA and DFS was demonstrated by connecting two silicon dual-drive Mach-Zehnder modulator (DDMZM) in serial and inserting a micro-ring resonator (MRR) between them [31]. However, the coupling between the bus waveguide of the MRR needs to be precisely controlled for a desirable frequency-phase response, as well as a locking system of resonant wavelength for stabilization, which brings high cost and complexity to the measurement system. Besides, its operation frequency of 5-21.4 GHz is limited for future Ka-band wireless communications and radar systems [32,33].
In this work, a simultaneous measurement system for AOA and DFS based on a silicon dual-parallel Mach-Zehnder modulator (Si-DPMZM) is proposed for the first time. The bandwidth of designed silicon DPMZM is 41 GHz, larger than that in the published work [31], which is only 11.1 GHz limits for future Ka-band radar systems. Furthermore, there are no extra resonant device required for introducing a customized frequency-phase response in the silicon chip, which reduces the systematic complexity and improves the stability. Both these two sub-MZMs operate at the minimum point. An echo signal drives a sub-MZM while the combination of a phase-delay echo signal and a transmitted signal drives the other sub-MZM. Subsequently, by comparing the powers, phases and frequencies of the two IF signals, the AOA, DFS (with direction) can be obtained simultaneously. The measured results show that the estimation error of AOA is less than ±3° at 30 GHz. The DFS at 30/40 GHz is also measured whose estimation error is less than 9.8 × 10−10 Hz with the offset of ±1 MHz. In addition, the fluctuation of DFS is kept at ±3 × 10−11 Hz within 120 minutes, indicating the high stability of the system.
2. Device design and characterization
The schematic structure and the microscope image of the Si-DPMZM are shown in Figs. 1(a) and (b), respectively. It is fabricated on the silicon-on-insulator (SOI) wafer with a 220 nm thick silicon layer and a 3 µm thick buried oxide (BOX). The Si-DPMZM is composed of a 1 × 2 thermo-optical power splitter (TOPS) and two identical silicon carrier depletion traveling wave MZMs (MZM-1/2). The pitch between two sub-MZM is designed as 300 µm to avoid EO crosstalk [34]. The total fiber-to-fiber insertion loss is ∼17 dB, which is composed of a ∼1 dB loss of TOPS, a ∼5 dB loss of DPMZM and a ∼11 dB loss of the two grating couplers. Each sub-MZM is comprised of two 2 mm long active waveguides and ground-signal electrode. An on-chip terminator with TiN-based 50 Ω resistance is connected to the end of the RF electrode in each sub-MZM. The electrode has a width of 60 µm and a gap of 40 µm, which is fabricated in the top 2 µm thick metal layer defined as metal-1. The other metal layer below the metal-1 is metal-2, which is used for electrical connections. The width and slab height of the rib waveguides are 500 nm and 90 nm, respectively.
Fig. 1. (a) Configuration of silicon DPMZM. (b) The zoomed-in paragraph of the silicon chip. (c) Cross section of silicon-based PN junction and TiN heater. GC: grating coupler; MMI: multimode interferometer; HT: heater; TOPS: thermo-optical power splitter.
The doping concentration of the PN junction varies from 0.8 × 1018cm-3 to 1.5 × 1018cm-3 in vertical direction. The doping concentrations of intermediate P + and N + regions are 2 × 1018 cm-3 and added 400 nm away from the PN junction, thus achieving relatively low series resistance and optical insertion loss. The P++ and N++ regions with doping concentration of 1 × 1020cm-3 are 1150 nm away from the PN junction for ohmic contact. The two PN junctions are connected in series to form a push-pull configuration, further reducing its capacitance by half. Therefore, the bandwidth of such silicon MZM is normally larger than that of DDMZM [35].
As shown in Fig. 1(a), four TiN-based thin film heaters, HT-1, HT-2, HT-3 and HT-4, are integrated with the Si-DPMZM. HT-1 is utilized to tune the optical power ratio between two sub-MZMs, while HT-2/3 and HT-4 are for tuning the bias conditions of MZM-1/2 and parent MZM, respectively. The EO response of the PN junction is measured and as shown in Fig. 2(a), where the effective index variation Δneff and the propagation loss Rloss are plotted as a function of the reverse bias voltage. Measured discrete data points in Fig. 2(a) are fitted with fourth-order polynomials, i.e., and ${R_{loss}}(v )={-} 10\textrm{lg}[{{e^{ - \alpha (v )L}}} ]= \mathop \sum \nolimits_{n = 0}^4 {p_n}{v^n}$. Here, kn and pn are fitted polynomial coefficients whose values are listed in Fig. 2(a). v, α and L represent the reverse bias voltage, the optical absorption coefficient and the length of the PN junction, respectively.
Fig. 2. (a) Effective index variation and transmission loss vs. reverse bias voltage on the PN junction. (b) Frequency response of the sub-MZM at different reverse bias voltages.
As shown in Fig. 2(b), the EO frequency responses of sub-MZM at different reverse bias voltage are ${n_{eff}}(v )= \mathop \sum \nolimits_{n = 1}^4 {k_n}{v^n}$ tested with a lightwave component analyzer (LCA, Keysight N4373D, N5247A). The test results indicate that the bandwidth of Si-DPMZM is more than 41 GHz when the reverse bias voltage exceeds 2 V, which covers the whole Ka band. According to the measured EO response shown in Fig. 2(a), the half-wave voltage of the sub-MZM is calculated to be 10 to 12 V when reverse bias voltage varies from 2 to 6 V.
3. Operation principle
The schematic diagram of the measurements system for AOA and DFS is shown in Fig. 3. An optical carrier from the laser diode (LD), E(t)=E0exp(jω0 t), is injected into the Si-DPMZM and split into two sub-MZMs by the TOPS. E0 and ω0 are the amplitude and angular frequency of optical carrier, respectively. The echo signal received by the antenna element (AE-2) is directly applied to MZM-1. The echo signal received by AE-1 with a phase delay of φ, is combined with the transmitted signal through an electrical synthesizer (ES), and then applied to MZM-2. Subsequently, the output signals of MZM-1 and MZM-2 can be expressed as E1(t) = 1/20.5E0exp[jω0t + jmEcos(ωEt)+jφ1], E2(t)= 1/20.5E0exp[jω0t + jmEcos(ωEt) + jmTcos(ωTt)+jφ2]exp(jθ), respectively. Here, φ1/2 denotes the bias phase between the two arms of MZM-1/2 while θ denotes the bias phase of the parent MZM. ωE and ωT represent the angular frequencies of echo and transmitted signals, respectively. mE/T is the modulation index, which can be calculated as ${m_{E/T}} = 2\pi L/\left( {\mathop \sum \nolimits_{n = 1}^4 n{k_n}{v^{n - 1}}{v_{E/T}}} \right)$ by using the fitted coefficients in Fig. 2(a). λ is the wavelength of the optical carrier. vE and vT denote the amplitudes of echo and transmitted signals, respectively.
Fig. 3. Proposed DFS and AOA system. (a)–(e) are the diagrams of the optical spectra at different locations. LD: laser diode; AE: antenna element; DPMZM: dual-parallel Mach-Zehnder modulator; PD: photodiode; ES: electrical synthesizer; EDFA: erbium doped fiber amplifier; WDM: wavelength division multiplexer.
Subsequently, the output signal of the Si-DPMZM is separated into two channels by a wavelength division multiplexer (WDM). After the Jacobi-Anger expansion and ignoring the high-order components, the separated upper and lower sidebands can be expressed as:
Here, φ1 and φ2 are set to be π for carrier-suppressed double side modulation (CS-DSB). The IF signals in two channels are detected by two low-speed photodiodes (PDs) which can be expressed as:
Here, η is the responsivity of the PD. According to the measurement data in Figs. 2(a) and 2(b), the reverse bias voltages of MZM-1/2 is set to be 2 V for a larger modulation bandwidth. At this condition, J0(mT) approximately equals to 1. Here, assuming θ =π/6, thus, A, B, C, D can be expressed as A = -1-cos(φ+π/6), B = -sin(φ+π/6), C = -1-cos(φ-π/6); D = -sin(φ-π/6), respectively. Then the phase difference between IU(t) and IL(t) can be deduced to:
According to Eqs. (5) and (6), when φ is from 0 to 360°, the phase delay of the two IF signals is -30°/-210°, indicating that the DFS is in the positive direction (ωE > ωT), shown in Fig. 4(a). In contrast, when the phase delay of the two IF signals is 30°/210°, the direction of DFS is negative (ωE < ωT), shown in Fig. 4(b). Substituting the measured DFS (fDFS) and the transmitted signal (fT) into the equation of vd = 0.5cfDFS/fT, the speed of detected target is obtained. Here, c is the light speed in vacuum.
Fig. 4. The calculated phase difference of IF signals versus the phase difference of echo signals when DFS direction is (a) positive and (b) negative.
According to the theory of quadratic detection, the electrical power of two IF signals can be deduced as:
From Eqs. (7) and (8), the electrical power-phase mapping curves of IF signals in two channels can be obtained. By comparing the relative magnitudes of two IF signals, the value of φ can be determined. Subsequently, AOA of β can be measured without ambiguity in the range of 0 to 360° based on β=arcsin(φc/dωE). Here, d is the distance of two AEs, usually half the wavelength of the received signal. It is noted that, only first-order component is considered in Eqs. (7) and (8). However, the nonlinearities from MZM transfer function and free carrier dispersion effect, introduce high-order components in modulation. If the Bessel function is expanded to 3rd-order polynomial, only 1% variation is introduced to the optical power in two PDs. Therefore, the effect of these two nonlinearities on the accuracy of AOA can be ignored. The DFS is obtained by measuring the frequency of the desired IF signals, whose accuracy is merely depended on the performance of the measurement device [36].
4. Experimental results
To verify the proposed scheme, the experimental setup is built and shown in Fig. 5. A TE-polarized light at a wavelength of 1550 nm and a power of 14 dBm from a continuous wave (CW) tunable laser (Santec TSL-710) is coupled into and out of the chip by two fiber grating couplers. A RF signal generated from a microwave signal generator (MSG-1, Keysight E8267D) is split into two paths by an electric power divider (EPD), serving as two echo signals. In order to emulate the phase difference between the two received signals, an additional phase shift is introduced into one path via an electric phase shifter (EPS). The echo signal without/with phase delay is denoted as echo-1/2. Another microwave signal generator (MSG-2, Keysight E8257D) is employed to generate the transmitted signal. The two MSGs are synchronized by a 10 MHz signal.
Fig. 5. Proposed experimental setup of the approach for DFS and AOA. LD: laser diode; PC: polarization controller; EDFA: erbium doped fiber amplifier; OBPF: optical band-pass filter; OSA: optical spectrum analyzer; OC: optical coupler; PD: photodiode; MSG: microwave signal generator; EPD: electric power divider; ES: electrical synthesizer; EPS: electrical phase shifter; DC: direct current; OSC: oscilloscope; ESA: electrical spectrum analyzer.
Three two-channel direct current sources (DC-source1/2/3) are used to tune the Si-DPMZM. DC-source1 is employed to control the split ratio of the TOPS and the bias phase of the MZM-1, respectively. DC-source2 is used to control the bias phase of the MZM-2 and offers a reverse bias voltage for the PN junctions. In addition, DC-source3 is utilized to introduce a constant bias phase of the parent MZM and offers a reverse bias voltage for the PN junctions of the MZM-1, respectively. Both the MZM-1 and MZM-2 are set to the minimum point for CS-DSB modulation.
Reverse bias voltages of PN junctions on two sub-MZMs are all set to 2 V. The MZM-1 is directly driven by the echo-1 while the MZM-2 is driven by the combined signal of transmitted signal and echo-2. The output signal of the Si-DPMZM is amplified to be 10 dBm by an EDFA (Keopsy, CEFA-C-PB-HP), then divided into two paths by a 50/50 optical coupler (OC). Two optical band-pass filters (OBPF1 and OBPF2, EXFO, XTM-50) are employed to select the upper and lower sidebands of the optical signal, respectively. Here, two OBPFs are used to emulate the WDM. Subsequently, the two filtered optical signals are sent to two low-speed PDs for demodulating the IF signals, i.e. upper/lower sideband to PD1/2. After that, the IF signals are observed by an oscilloscope (OSC, Keysight MXR058A) and an electrical spectrum analyzer (ESA, Rohde & Schwarz FSWP).
The transmitted signal and the echo signals are set at 30 GHz and 30.001 GHz, respectively, while the electrical powers are both set at 15 dBm. In the practical applications, the received echo signal by an antenna is weak. Meanwhile, silicon modulators also introduce large insertion loss due to the carrier absorption effect [28]. To increase the sensitivity of the whole measurement system, a multi-stage low-noise RF amplifier is needed to amplify the echo signal firstly [37]. Here, the power of echo signal is set as 15 dBm to simulate this scenario. Then the filtered optical spectrums of upper and lower sidebands are captured by an optical spectrum analyzer (OSA, Yokogawa, AQ6374) with the resolution of 0.02 nm. As shown in Fig. 6(a), the sideband-to-carrier ratio is 33 dB, indicating that the OBPF brings a high suppression of optical carrier. Here, two sub-MZMs are not required to operate at the ideal CS-DSB point for the systematic stability. With the high-suppression OBPF, the sub-MZM optical sideband-to-carrier suppression ratio of 20 dB is enough for selecting the two optical sidebands. Since the frequency of transmitted signal is too close to that of the echo signal and the minimum resolution is merely 0.02 nm, they cannot be identified separately in the optical spectrum.
Fig. 6. (a) The measured optical spectrums of upper and lower sidebands after the OBPF at 30 GHz. (b) Measured electrical spectrum of DFS signal at 30 GHz.
As shown in Fig. 6(b), the ratio between the IF signal and its second-order harmonic distortion (HD2) captured by ESA is up to 46 dB. Thus, the HD2 exerts very little influence on DFS estimation. The normalized IF signal power versus phase difference between echo-1 and echo-2 is shown in Fig. 7. Since the two echo signals interfered in the optical domain, the phase delay between them causes the output power change of IF signal. As it can be seen that the IF signal power varies from 0 to -30 dB when the phase difference changes from 0° to 180°. Such significant change of IF signal power helps improve the ability to identify the phase difference. It is achieved by increasing the consistency of RF gains of two sub-MZMs, i.e., tuning the splitting ratio of the TOPS.
Fig. 7. The DFS signal results under the different phase differences of two echo signals at 30 GHz. (a) In phase. (b) Out of phase.
Figure 8(a) displays the normalized power of IF signal versus the phase delay of two echo signals from 0° to 360°. The curves and dots marked by blue/red denote the theoretical and experimental results of IF signal from the upper/lower sideband, respectively. As it is shown in Fig. 8(a), these two curves can be utilized to identify the AOA without ambiguity by comparing the absolute values and relative magnitudes of the two IF signals. As shown in Fig. 8(a), the phase distance between two curves is ∼70°, indicating that the bias phase of the parent-MZM is ∼35°. The estimation errors of the phase delay are shown in Fig. 8(b). It can be seen that the errors are less than 3° in the range of 360° which indicates this system has a good phase measurement performance. The estimation errors of AOA are mainly from the drifts of the bias phases in two sub-MZMs and parent MZM [38]. Based on our analysis model, 0.5/1° of estimation error is introduced to the AOA measurement when sub-MZMs/parent-MZM bias phase drifts 1°. Due to the thermal crosstalk of the TiN-based heaters for tuning the bias phases, the errors are deteriorated. This issue can be addressed by introducing deep-trench process in the next work. Assuming the distance of two AEs is the half of the microwave signal wavelength, the range of AOA can be measured from 0° to ±90° without ambiguity.
Fig. 8. (a) The measurement of AOA in which dots and curve represent the normalized experimental data and fitting curve at 30 GHz. (b) The measurement errors of AOA at 30 GHz.
Figure 9 shows the electrical waveforms of the IF signals demodulated by the PD1 (blue) and PD2 (red). According to Eqs. (5)-(6) and the theoretical analysis, the absolute value of IF signal is 1 MHz. As shown in Fig. 9(a), the waveform from the PD1 is ahead about 215° of that from the PD2, indicating the direction of DFS signal is positive and a target is moving towards a receiver. In contrast, as shown in Fig. 9(b), the waveform from the PD1 is behind about 215° of that from the PD2, denoting the direction of DFS signal is negative and the target is moving away from the receiver. The two experimental results verify the numerical calculation as shown in Fig. 4.
Fig. 9. The electrical waveforms of beating IF signal with 30 GHz transmitted signal from the PD1 and PD2 when the DFS signal is (a) + 1 MHz and (b) -1 MHz, respectively.
In order to investigate the frequency tunability and the accuracy of the measurement system, MSG-2 is fixed at 30/40 GHz and the frequency of the output signal of MSG-1 is adjusted according to the step steps of 100 kHz, 10 kHz and 1 kHz to keep it away from 30/40 GHz. The measured DFSs and the corresponding estimation errors are shown in Fig. 10(a)-(c). Significantly, the measured DFSs and estimation errors at 30 GHz are identical with that at 40 GHz. It confirms that the implementation of the proposed approach is independent of transmitted signal frequency. Here, the measured DFSs is just the specified frequency difference between the MSG-1 and MSG-2, i.e., the sources of the echo signal and the transmitted signal, respectively. Consequently, a subtraction between the frequency difference (MSG-1 and MSG-2) and the IF signal frequency measured by the ESA provides the estimation errors. As shown from Fig. 10(a) to (c), the estimation errors are calculated to be lower than 9.8 × 10−10, 7.2 × 10−10 and 1.8 × 10−10 Hz for the three offset steps, respectively. The measurement errors here are all positive, indicating inherent errors, and the actual DFS estimation error should be smaller. It is worth pointing out that the estimation errors are related to the performance of ESA [36] and can be further reduced by utilizing a higher-resolution ESA. However, in the practical applications, a better choice is to utilize digital signal processing (DSP) methods to obtain a DFS signal, whose estimation error is associated with the sampling rate of an analog-to-digital converter (ADC).
Fig. 10. The measured DFSs and estimation errors with the 30/40-GHz transmitted signal being tuned at steps of (a) 100 kHz, (b) 10 kHz, and (c) 1 kHz; (d)-(e) The measured actual speed and estimation errors based on the results of (a)-(c).
According to the measured DFSs and the estimation errors from Fig. 10(a) to (c), we calculate the speeds and the corresponding errors based on vd = 0.5cfDFS/fT, respectively. As shown from Fig. 10(d) to (e), the measurable ranges of speed at 30 GHz with the DFSs of ±1 MHz, ± 100 kHz and ±10 kHz are 5 km/s, 500 m/s and 50 m/s, respectively. By contrast, at 40 GHz, the measurable ranges of speed with the three DFSs are 3.75 km/s, 375 m/s and 37.5 m/s, respectively. On the other hand, the speed errors for the three offset steps are calculated to be less than 4.9 × 10−12/3.7 × 10−12, 3.6 × 10−12/2.7 × 10−12 and 0.9 × 10−12/0.7 × 10−12 m/s at 30/40 GHz, respectively. Obviously, there is a tradeoff between the measurement range and accuracy of target moving speed. Therefore, the frequency can be optimal selected based on the requirement of desired detected speed range and accuracy of a target.
In order to verify the system stability, the DFS measurement within 120 mins is carried out. The frequency fluctuation of the measured DFS, as shown in Fig. 11, is kept within ±3 × 10−11 Hz, which demonstrates the good system stability for DFS measurement.
5. Conclusion
In conclusion, a simultaneous Ka-band AOA and DFS measurement system based on the Si-DPMZM is proposed and demonstrated. The AOA and DFS can be obtained by processing two captured IF signals from the mixing of the transmitted signal and the echo signal. By comparing the relative amplitudes of the normalized power from the two IF signals, the AOA can be estimated without ambiguity from 0 to ±90°. The measurement results show that the estimation error is less than 3° at 30 GHz. The absolute value of DFS is the frequency of IF signal while the direction of DFS can be obtained by comparing the phase difference between the two IF signals. The estimation errors of DFSs at 30/40 GHz with the offsets of 1 MHz, 100 kHz and 10 kHz are less than 9.8 × 10−10, 7.2 × 10−10 and 1.8 × 10−10 Hz. Furthermore, the stability of the measurement system is demonstrated as well, the fluctuation of DFS is less than 3 × 10−11 Hz within 120 mins. To the best of our knowledge, it is the first measurement system for AOA and DFS at Ka band based on silicon DPMZM. In addition, the structure and performance of the system is quite simple and excellent. Furthermore, WDM, PDs can be integrated on a silicon photonic chip for miniaturizing the system. In the next work, the whole system is considered to be further optimized and simplified for higher robustness and lower cost, which might be a promising solution in high-frequency radar and communication applications.
Funding
National Key Research and Development Program of China (2018YFB2200602); Science and Technology Innovation 2025 Major Project of Ningbo (2020Z021); Zhejiang Lab Research Funds (2020LC0AD02); Science and Technology Plan Project of Zhejiang (2022C01108).
Acknowledgments
The authors Qiang Zhang and Sihan Chen contribute equally to this work. The authors would like to thank Dr. Bing Wei, Training Platform of Information and Microelectronic Engineering in Polytechnic Institute of Zhejiang University.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. X. Zou, B. Lu, W. Pan, L. Yan, A. Stohr, and J. Yao, “Photonics for microwave measurements,” Laser Photonics Rev. 10(5), 711–734 (2016). [CrossRef]
2. X. Zou, W. Bai, W. Chen, P. Li, B. Lu, G. Yu, W. Pan, B. Luo, L. Yan, and L. Shao, “Microwave photonics for featured applications in high-speed railways: communications, detection, and sensing,” J. Lightwave Technol. 36(19), 4337–4346 (2018). [CrossRef]
3. G. Li, X. Meng, L. Wang, M. Li, N. Zhu, and W. Li, “Simultaneous and unambiguous identification of DFS and AOA without dependence on echo signal power,” Opt. Lett. 48(4), 1028–1031 (2023). [CrossRef]
4. S. Dang, O. Amin, B. Shihada, and M. Alouini, “What should 6 G be,” Nat. Electron. 3(1), 20–29 (2020). [CrossRef]
5. X. Wang, F. Zhou, D. Gao, Y. Wei, X. Xiao, S. Yu, J. Dong, and X. Zhang, “Wideband adaptive microwave frequency identification using an integrated silicon photonics scanning filter,” Photonics Res. 7(2), 172–181 (2019). [CrossRef]
6. J. Yao, “Microwave Photonics,” J. Lightwave Technol. 27(3), 314–335 (2009). [CrossRef]
7. J. Capmany, J. Mora, I. Gasulla, J. Sancho, J. Lloret, and S. Sales, “Microwave Photonic Signal Processing,” J. Lightwave Technol. 31(4), 571–586 (2012). [CrossRef]
8. X. Zou, W. Li, W. Pan, B. Luo, L. Yan, and J. Yao, “Photonic approach to the measurement of time-difference-of-arrival and angle-of-arrival of a microwave signal,” Opt. Lett. 37(4), 755–757 (2012). [CrossRef]
9. Z. Cao, Q. Wang, R. Lu, H. P. A. van den Boom, E. Tangdiongga, and A. M. J. Koonen, “Phase modulation parallel optical delay detector for microwave angle-of-arrival measurement with accuracy monitored,” Opt. Lett. 39(6), 1497–1500 (2014). [CrossRef]
10. H. Zhuo, A. Wen, and Y. Wang, “Photonic angle-of-arrival measurement without direction ambiguity based on a dual-parallel Mach–Zehnder modulator,” Opt. Commun. 451, 286–289 (2019). [CrossRef]
11. H. Chen and E. H. W. Chan, “Simple approach to measure angle of arrival of a microwave signal,” IEEE Photonics Technol. Lett. 31(22), 1795–1798 (2019). [CrossRef]
12. T. Lin, Z. Zhang, Y. Wang, Z. Zhu, S. Zhao, J. Liu, W. Jiang, and G. Wu, “Photonic 2-D angle-of-arrival estimation based on an L-shaped antenna array for an early radar warning receiver,” Opt. Express 28(26), 38960–38972 (2020). [CrossRef]
13. P. Li, L. Yan, J. Ye, X. Feng, X. Zou, B. Luo, W. Pan, T. Zhou, and Z. Chen, “Angle-of-arrival estimation of microwave signals based on optical phase scanning,” J. Lightwave Technol. 37(24), 6048–6053 (2019). [CrossRef]
14. Z. Tu, A. Wen, Z. Xiu, W. Zhang, and M. Chen, “Angle-of-arrival estimation of broadband microwave signals based on microwave photonic filtering,” IEEE Photonics J. 9(5), 1–8 (2017). [CrossRef]
15. H. Chen and E. H. W. Chan, “Angle-of-arrival measurement system using double RF modulation technique,” IEEE Photonics J. 11(1), 1–10 (2019). [CrossRef]
16. W. Chen, A. Wen, X. Li, Y. Gao, Y. Wang, S. Xiang, H. He, and H. Zheng, “Wideband Doppler frequency shift measurement and direction discrimination based on a DPMZM,” IEEE Photonics J. 9(2), 1–8 (2017). [CrossRef]
17. C. Yi, H. Chi, and T. Jin, “A photonic approach for Doppler frequency shift measurement with dispersion medium,” IEEE Photonics J. 12(5), 1–8 (2020). [CrossRef]
18. F. Zhang, J. Shi, and S. Pan, “Photonics-based wideband Doppler frequency shift measurement by in-phase and quadrature detection,” Electron. Lett. 54(11), 708–710 (2018). [CrossRef]
19. B. Lu, W. Pan, X. Zou, Y. Pan, X. Liu, L. Yan, and B. Luo, “Wideband microwave Doppler frequency shift measurement and direction discrimination using photonic I/Q detection,” J. Lightwave Technol. 34(20), 4639–4645 (2016). [CrossRef]
20. X. Li, A. Wen, W. Chen, Y. Gao, S. Xiang, H. Zhang, and X. Ma, “Photonic Doppler frequency shift measurement based on a dual-polarization modulator,” Appl. Opt. 56(8), 2084–2089 (2017). [CrossRef]
21. J. Li, W. Yu, Z. Zhang, T. Lin, S. Dai, and J. Liu, “Simple approach for Doppler frequency shift estimation based on a dual-polarization quadrature phase shift keying (DP-QPSK) modulator,” Appl. Opt. 59(7), 2114–2120 (2020). [CrossRef]
22. P. Zuo and Y. Chen, “Photonic-assisted filter-free microwave Doppler frequency shift measurement using a fixed low-frequency reference signal,” J. Lightwave Technol. 38(16), 4333–4340 (2020). [CrossRef]
23. Y. Chen, P. Zuo, T. Shi, and Y. Chen, “Photonic-enabled Doppler frequency shift measurement for weak echo signals based on optical single sideband mixing using a fixed low-frequency reference,” J. Lightwave Technol. 39(10), 3121–3129 (2021). [CrossRef]
24. Z. Tang and S. Pan, “Simultaneous measurement of Doppler-frequencyshift and angle-of-arrival of microwave signals for automotive radars,” in Proc. Int. Topical Meeting Microw. Photon., 2019, pp. 1–4.
25. P. Li, L. Yan, J. Ye, X. Feng, W. Pan, B. Luo, X. Zou, T. Zhou, and Z. Chen, “Photonic approach for simultaneous measurements of Doppler-frequency-shift and angle-of-arrival of microwave signals,” Opt. Express 27(6), 8709–8716 (2019). [CrossRef]
26. C. Huang, H. Chen, and E. H. W. Chan, “Simple photonics-based system for Doppler frequency shift and angle of arrival measurement,” Opt. Express 28(9), 14028–14037 (2020). [CrossRef]
27. J. Zhao, Z. Tang, and S. Pan, “Photonic approach for simultaneous measurement of microwave DFS and AOA,” Appl. Opt. 60(16), 4622–4626 (2021). [CrossRef]
28. B. Jalali and S. Fathpour, “Silicon Photonics,” J. Lightwave Technol. 24(12), 4600–4615 (2006). [CrossRef]
29. A. H. Atabaki, S. Moazeni, F. Pavanello, H. Gevorgyan, J. Notaros, L. Alloatti, M. T. Wade, C. Sun, S. A. Kruger, H. Meng, K. A. Qubaisi, I. Wang, B. Zhang, A. Khilo, C. V. Baiocco, M. A. Popovic, V. M. Stojanovic, and R. J. Ram, “Integrating photonics with silicon nanoelectronics for the next generation of systems on a chip,” Nature 556(7701), 349–354 (2018). [CrossRef]
30. C. Rogers, A. Y. Piggott, D. J. Thomson, R. F. Wiser, I. E. Opris, S. A. Fortune, A. J. Compston, A. Gondarenko, F. Meng, X. Chen, G. T. Reed, and R. Nicolaescu, “A universal 3D imaging sensor on a silicon photonics platform,” Nature 590(7845), 256–261 (2021). [CrossRef]
31. L. Tang, Z. Tang, S. Li, S. Liu, and S. Pan, “Simultaneous Measurement of Microwave Doppler Frequency Shift and Angle of Arrival Based on a Silicon Integrated Chip,” IEEE Trans. Microwave Theory Techn. 1, 1 (2022). [CrossRef]
32. Y. Song, Y. Wu, J. Yang, Y. Tian, W. Tong, Y. Chen, C. Wang, X. Tang, J. Benedikt, and K. Kang, “A Compact Ka-Band Active Integrated Antenna With a GaAs Amplifier in a Ceramic Package,” Antennas Wirel. Propag. Lett. 16, 2416–2419 (2017). [CrossRef]
33. W. Deng, R. Wu, Z. Chen, M. Ding, H. Jia, and B. Chi, “A 35-GHz TX and RX Front End With High TX Output Power for Ka-Band FMCW Phased-Array Radar Transceivers in CMOS Technology,” IEEE Trans. VLSI Syst. 28(10), 2089–2098 (2020). [CrossRef]
34. P. Xia, H. Yu, M. Yang, N. Cui, H. Liao, Q. Zhang, Z. Fu, Q. Huang, N. Ning, Z. Wei, X. Jiang, and J. Yang, “High sideband suppression silicon single sideband modulator integrated with a radio frequency branch line coupler,” Photonics Res. 11(2), 329–336 (2023). [CrossRef]
35. M. Li, L. Wang, X. Li, X. Xiao, and S. Yu, “Silicon intensity Mach–Zehnder modulator for single lane 100 Gb/s applications,” Photonics Res. 6(2), 109–116 (2018). [CrossRef]
36. X. Zou, W. Li, B. Lu, W. Pan, L. Yan, and L. Shao, “Photonic Approach to Wide-Frequency-Range High-Resolustion Microwave/Millimeter-Wave Doppler Frequency Shift Estimation,” IEEE Trans. Microwave Theory Techn. 63(4), 1421–1430 (2015). [CrossRef]
37. Q. Zhang, J. Ji, Q. Cheng, Y. Duan, J. Zang, J. Yang, H. Yu, and X. Zhang, “Two-dimensional phased-array receiver based on integrated silicon true time delay lines,” IEEE Trans. Microwave Theory Techn. 71(3), 1251–1261 (2022). [CrossRef]
38. H. Zhuo and A. Wen, “A Photonic Approach for Doppler-Frequency-Shift and Angle-of-Arrival Measurement Without Direction Ambiguity,” J. Lightwave Technol. 39(6), 1688–1695 (2021). [CrossRef]
