Phase to intensity modulation and fading compensation using a single ring resonator for radio-over-fiber links

Zhi Lai; Zhi Lai; Zhenjiang Zhang; Teng Shi; Liang Guo; Liang Guo

doi:10.1364/OE.511966

1. Introduction

Radio-over-Fiber (RoF) constitutes a pivotal component of broadband wireless access services within the burgeoning realm of optical-wireless networks. This technology serves as a potent solution for augmenting both capacity and mobility, while concurrently reducing costs. Its applicability extends to diverse environments, including but not limited to conference centers, airports, hotels, and, ultimately, residential spaces and small offices [1].

These systems facilitate the integration of radio access control and signal processing functions at a centralized control station (CS), allowing for the seamless transmission of the radio signal to simplified antenna sites (AS) through optical fiber. There is an anticipation that the millimeter-wave (mm-wave) bands will be employed to fulfill the need for higher signal bandwidth and mitigate frequency congestion in forthcoming ROF-based optical-wireless access networks [2]. Despite being recognized as a highly promising candidate in radar communication, ROF confronts a spectrum of both opportunities and challenges. A notable concern is the predominant emphasis on standard-reach (∼20 km) ROF access networks, with relatively limited attention directed towards long-reach (100 km) ROF systems. A primary factor contributing to this discrepancy is the substantial chromatic dispersion induced by long-haul fibers, leading to disparate delays between the sidebands and the carrier. To address this issue, Optical Carrier Suppression (OCS) modulation is employed as a strategy for mitigation [3]. Nevertheless, it is essential to note that OCS modulation, while serving as a solution, can introduce significant time shifting [4] of the data due to chromatic dispersion induced by the two tones. This, in turn, leads to an unacceptable error ratio during the transmission process [5]. An alternative approach to mitigate the power fading effect and time shifting is the utilization of a Single-Sideband (SSB) ROF system. However, it is important to note that SSB necessitates precision in optical notch filtering to effectively separate the carrier and side tone. In response to these challenges, certain researchers propose compensating for chromatic dispersion (CD) by employing a Dispersive Compensation Fiber (DCF) within the optical domain. Nevertheless, it's worth noting that DCF often entails high levels of loss and cost. Additionally, the dispersion compensation provided by DCF tends to be unadjustable.

Another notable concern revolves around the substantial demand for ROF, necessitating the use of low-cost and low-complexity link components to facilitate mass production. A ROF link is comprised of several key components, including optical downlink transmission utilizing the optical frequency multiplication principle, the generation of a remote local oscillator (LO), the remote down-conversion of radio-frequency uplink signals, and optical uplink transmission employing intensity modulation-direct detection. The MZMs are commonly utilized by researchers to achieve the hybrid integration of ROF and fiber-to-the-home systems [6].

Nevertheless, a limitation arises regarding the maximum modulating microwave signal power. This limitation stems from the requirement for an external modulator to operate within its linear region, necessitating weak modulation of the optical carrier by microwave signals. Conversely, an excessive increase in the power of the optical carrier relative to the sidebands leads to a diminished dynamic range and the potential for damage to components, such as photodiodes. Furthermore, the direct detection of the optical carrier results in a DC photocurrent, which contributes directly to the shot noise at the photoreceiver. The Phase Modulator (PM), considered as an alternative candidate, exhibits superior performance compared to the MZM. The primary advantage of employing an optical PM in a microwave photonic (MWP) system is the absence of biasing, thereby eliminating the bias drifting issue inherent in an MZM-based MWP system [7]. To alleviate these effects and enhance link modulation efficiency simultaneously, one approach is to remove a portion of the optical carrier immediately after modulation. Historically, reported methods for reducing the optical carrier have involved various techniques, including the utilization of Brillouin scattering [8,9], and the incorporation of external optical filters such as silica delay lines [10], Fabry–Perot filters [11], fiber Bragg gratings [12], and arrayed waveguide gratings (AWG) [13]. With the exception of the Fiber Bragg Grating (FBG) approach, these schemes can pose challenges in implementation or may not function effectively across a broad spectrum of modulation depths, frequencies, or formats. Alternatively, other methods either necessitate large and costly pumps or involve non-integrated devices, resulting in instability and a lack of functional extensibility.

The ring resonators (RRs) have the advantages of smaller size and lower complexity, they can be easily incorporated with tuning elements for design flexibility, thus it shows great application in the ROF systems [14,15]. The concept of Phase Modulation to Intensity Modulation (PM-IM) conversion utilizing a ring resonator has been introduced in Ref. [16]. Nevertheless, while the modulation transformer based on the RR has been proposed, it has not been subjected to a detailed analysis. In the present study, we employ a phase modulation to Ring Resonator (PM-RR) combination as a substitute for MZMs. This configuration offers a compelling solution for transforming phase-modulated optical signals into intensity-modulated versions with minimal conversion of laser phase noise to intensity noise, presenting an intriguing advancement. During the optimization of ring parameters, the PM-RR combination exhibits superior signal quality and lower nonlinearity compared to traditional MZMs. Moreover, the RR effectively combats fading effects by adjusting the phase response of the RR to match the phase changes induced by CD. Furthermore, the proposed system can be integrated on chip without discrete bulk devices or benchtop instruments and hence are plagued by the limitations in system cost, space, and power consumption.

2. Device and system principles

2.1 Ring resonator performance

The configuration of the RR is illustrated in Fig. 1. The implementation of the designed filter response can be achieved through the utilization of a RR. The transfer function and the resultant frequency responses of a RR, specifically the amplitude and phase responses, can be formulated as per the expression provided in Ref. [17]

(1)$${H_{RR}}(f) = \frac{{\sqrt {1 - k} - \sqrt \alpha {e^{ - j(f + \phi )}}}}{{1 - \sqrt {\alpha (1 - k)} {e^{ - j(f + \phi )}}}}$$

Here, f = 2πF/f_FSR denotes the angular frequency normalized to the free spectral range (FSR) f_FSR of the RR. The parameters k, φ, and α represent the power coupler coupling coefficient, an additional roundtrip optical phase shift of the ring loop, and the roundtrip power transmission coefficient, respectively. In practical application, the parameters k and φ can be realized as tuning elements. Specifically, k governs the form of the frequency responses, while φ regulates the frequency shift in the response. In practical, the coupling coefficient k can be controlled precisely by controlling the heater on the arm of the Mach-Zehnder (MZ) couplers in the MZ coupler-assisted ring resonator [18].

Fig. 1. A schematic of a RR and an illustration of phase and amplitude configurations.

Download Full Size | PDF

The ROF link, based on PM-RR, consists of a Continuous Wave (CW) laser, PM, RR, and a photodiode employed for direct signal detection, as illustrated in Fig. 1(a). Figure 1(b) displays the amplitude and phase response characteristics of the RR. The notch depth can be adjusted by manipulating k, while the frequency shift in the responses of RR can be accomplished by φ. Leveraging the tuning properties of the device allows for the attainment of various configurations in the phase response. Evidently, the targeted phase jump of π as illustrated by the red line in Fig. 1(b) can be approximated by setting k to 0.87 and φ=-π/2, alternatively, the jump phase of π/2 can be achieved by setting k to 0.96 and φ=-π/2 as illustrated by the blue dash line. In contrast to the ideal abrupt-phase jumps anticipated in the desired phase responses, the RR introduces phase errors attributable to the intrinsic shape of its phase transition, thereby impacting the 3-dB bandwidth of the device. The phase to intensity modulation can be proved in Fig. 1(c), which can also be integrated using silicon photonics for more compact, and stable operation [19–22].

2.2 System performance

Assuming that the RF bandwidth of the PM- Microwave Photonic Link (MPL) is determined by the suggested modulation transformer, the calculation of the equivalent RF transfer function of the RR-based modulation transformer can be performed using the transfer response function of the RR, as expressed in Eq. (1), This calculation is formulated by

(2)$$\begin{array}{l} {H_{RF}}({f_{RF}}) = {T_U} + {T_L}\\ = \frac{1}{2}real[{H_{RR}^{MT}{{( - \Delta f)}^\ast }H_{RR}^{MT}({f_{RF}} - \Delta f)} ]+ \frac{1}{2}real[{ - H_{RR}^{MT}{{( - {f_{RF}} - \Delta f)}^\ast }H_{RR}^{MT}( - \Delta f)} ]\end{array}$$

where T_U and T_L indicate the beat product contributions from the upper and lower optical sidebands, respectively, f_RF= 2πΔf_RF/f_FSR signifies the RF frequency normalized to the free spectral range (FSR), f_FSR, of the RR, and Δf = 2πΔf/f_FSR = |φ| denotes the normalized frequency spacing between RR resonance and the optical carrier. Figure 2 depicts the profile of RF power transmission across a RF frequency spacing within the RR frequency range, equivalent to the FSR of the RR for various values of k and Δf, with the loss α set to 0.96. In Fig. 2(a), where k = 0.87, it is observed that as the Δf decreases, the transmission power initially rises before subsequently declining, while the bandwidth consistently expands. Conversely, in Fig. 2(b), with Δf set to 0.5π, reducing the value of k results in an initial increase in power, followed by the emergence of a notch in the middle of the transmission profile. To enhance the system's assessment, two parameters are defined. The first is the 3 dB average power of the RR-based modulation transformer, signifying the average power within the range of 3 dB of RF transmission power. In cases where the maximum power falls below 3 dB, the 3 dB average RF transmission power is redefined as the average power between the maximum power and the maximum-3 dB power. The second parameter is the 3 dB transmission bandwidth, representing the range between the maximum RF bandwidth and the RF bandwidth at 3 dB below the maximum. In cases where the maximum power is below 3 dB, the 3 dB average RF transmission bandwidth is redefined as the bandwidth between the maximum power and maximum-3 dB power.

Fig. 2. Simulated RF dispersion of the RR-based modulation transformer for different values of (a) normalized RF frequency (k = 0.87) and (b) k (Δf = 0.5π).

Download Full Size | PDF

3. System optimization

3.1 Bandwidth and average power

The outcomes for the 3 dB average power of the RR-based modulation transformer, the 3 dB transmission bandwidth, and dispersion for various values of Δf, k are depicted in Fig. 3. During this evaluation, the parameterα is maintained at a value of 0.96. In Fig. 3(a), when k approaches 1, the average RF transmission profile exhibits a parabolic shape. In contrast, a noticeable notch appears in the middle of the average RF transmission figure. A selection of k = 0.724 and Δf = 0.26π is made to achieve a relative high power and minimize variation when altering the values of k and Δf. In Fig. 3(b), it is evident that, regardless of the value of k the 3 dB RF bandwidth decreases with the increase of the Δf. Furthermore, when k is less than 0.3, a pronounced notch appears in the graph, indicating a substantial decrease in bandwidth within this range. Under these conditions, the selected parameters are k = 0.164, Δf = 0.05π to attain the maximum bandwidth.

Fig. 3. (a) Simulated RF average 3 dB power transmission (b) Normalized 3-dB transmission bandwidth (c) dispersion with k = 0.87 and (d) dispersion with the Δf = 0.5π of the RR-based modulation transformer for different values of k and normalized RF frequency. (α=0.96).

Download Full Size | PDF

During the optimization of the dispersion for the RR-based modulation transformer, it is observed that the dispersion variation increases as the Δf deviates from 0.5π as depicted in Fig. 3(c). Figure 3(d) indicates that the dispersion experiences a substantial increase with a reduction in the value of k. To provide a numerical illustration, when k = 0.87, Δf = 0.5π, the maximum of dispersion is 6 dB. In contrast, for k = 0.27, Δf = 0.5π, the resulting dispersion surges to 3000 (1000 T/f_FSR). The unit of dispersion is normalized based on the time delay T and the f_FSR of the RR, with 1000 representing the number of points chosen to determine the deviation in transmission of the RR-based modulation transformer.

3.2 Nonlinearity

The generation of the third-order intermodulation distortion (IMD) is characterized by a spur in the radiofrequency (RF) output, emanating from the identical mechanism observed in Mach-Zehnder Modulator-Based Photonic Links (MZM-MPLs). In this context, the inclusion of ring resonators (RRs) within the optical paths gives rise to an additional modification in the optical signal spectrum. This alteration subsequently engenders changes in the linearity performance of the overall system. In order to assess and juxtapose the performances of the proposed ring resonator (RR) and conventional Mach-Zehnder Modulator-Based Photonic Links (MZM-MPLs), we introduce the ratio of the resulting third-order input interception points for the proposed PM-MPL to that of the MZM. Through a comparative analysis of the third-order input interception point (IIP3) obtained in the proposed PM-MPL with the corresponding metric in the Mach-Zehnder Modulator-Based Photonic Links (MZM-MPL) under identical modulation indices (IIP3MZM), the discernible influence of the ring resonators (RRs) on the system's linearity can be effectively characterized. For a more comprehensive performance evaluation between our designed PM- Microwave Photonic Link (PM-MPL) and conventional Mach-Zehnder Modulator-Based Photonic Links (MZM), we introduce the concept of relative third-order input interception point (RIP3). Specifically, RIP3_PM-RR is defined as the ratio of the third-order input interception point in the PM-MPL with ring resonators (IIP3_PM-RR) to that in the MZM (IIP3_MZM).

The IIP3 of the MZM is (4/RL) × (V_πMZM/π)2, and then the IIP3 of the RR can be written as (4/RL) × (V_πMZM/π)2×RIP3_PM-RR. Assuming a two-tone transmission with radiofrequency (RF) frequencies denoted as f_RF1 and f_RF2 (f_RF1 > f_RF2), the complex signal amplitude subsequent to sequential passage through a modulator transformer can be articulated as follows [16]:

(3)$$\begin{aligned} & E_{{out}}(t)=E_{{in}} \sqrt{L} \exp \left[-j m_{R F 1} \cos \left(2 \pi f_{R F 1}\right)-j m_{R F 2} \cos \left(2 \pi f_{R F 2}+\varphi\right)\right] \otimes h_{R R}^{M T}(t) \\ & =E_{{in}} \sqrt{L} \sum_{p=-\infty}^{\infty} \sum_{q=-\infty}^{\infty} j^{p+q} J_p\left(m_{R F 1}\right) J_q\left(m_{R F 2}\right) \exp \left[j 2 \pi\left(p f_{R F 1}+q f_{R F 2}\right) t+j q \varphi\right] H_{R R}^{M T} \\ & =E_{{in}} \sqrt{L} \sum_{p=-\infty}^{\infty} \sum_{q=-\infty}^{\infty} \Gamma_{p, q} \end{aligned}$$

Here, Γ_p,q represents normalized signal amplitude of the modulation transformer, φ represents the initial phase difference between the two RF tones. p, q represents the order of expansion of Eq. (3) using a Bessel function. The interaction among the optical components outlined in Eq. (3) results in the production of various orders of electrical harmonics and intermodulation distortion (IMD) products. Specifically, a photocurrent component featuring a frequency of f_RF1+ f_RF2 can be formulated as:

(4)$${I_{{n_1},{n_2}}}(t) = r{|{{E_{in}}} |^2}L\textrm{Re} \left\{ {\sum\limits_{p ={-} \infty }^\infty {\sum\limits_{q ={-} \infty }^\infty {{\Gamma _{p,q}}\Gamma _{p,q}^\ast } } } \right\} = A \cdot \textrm{Re} \left\{ {\sum\limits_{p ={-} \infty }^\infty {\sum\limits_{q ={-} \infty }^\infty {{\Gamma _{p,q}}\Gamma _{p,q}^\ast } } } \right\}$$

where r represents the responsivity of the detector and Re{…} signifies the real part of the complex signal. For the sake of simplicity, we make the assumption that m_RF1= m_RF2= m and a small signal approximation is deemed applicable, wherein only Γ_p,q with a frequency ≤2f_RF2 are considered for contribution in Eq. (4). Subsequently, the amplitudes of the RF fundamental and IMD3 components can be expressed as functions of the modulation index m, allowing for the determination of the the 3_rd-order input interception point (IIP3) for the proposed PM-MPL. In principle, the PM-MPL with two RRs and a balanced photodetector is free of even-order intermodulation distortions (IMDs) [16]. For simplicity, here we employ the PM-RR link with a phase modulator, a single ring resonator, and a PD to proof the concept.

The outcomes of RIP3_PM-RR calculation are presented in Fig. 4, taking into account the parameter α = 0.97. To comprehensively analyze the performance, we extend the consideration to p, q = 10. In Fig. 4(a), for k = 0.97, the RIP3 exhibits a parabolic curve shape, reaching its minimum value when f_RF is 1π. As k decreases, the RIP3 assumes an irregular graphical pattern. Furthermore, irrespective of the values of k and Δf, the RIP3 consistently attains zero when f_RF equals 1π. Additionally, the RIP3 graph exhibits symmetry with respect to f_RF. Therefore, in subsequent analyses, it suffices to consider only half of the f_RF. In Fig. 4(b), where the value of Δf is varied, a phenomenon analogous to that in Fig. 4(a) is observed: the RIP3 consistently reaches zero when f_RF equals to1π.

Fig. 4. Calculated RIP3 of PM-RR as a function of the normalized RF frequency with (a) different values of k (Δf = 0.4π) and (b) different values of Δf (k = 0.87) (α=0.96).

Download Full Size | PDF

Figure 5 illustrates the calculation of the RIP3 for the PM-RR system across various values of k and Δf, considering different f_RF and parameter α. By appropriately configuring the parameters of the ring resonator, it is feasible to achieve superior nonlinearity tolerance in the PM-RR system compared to that of MZM. In Fig. 5(b) and (e), a specific observation emerges: when f_RF equals 1π, the RIP3 consistently attains zero, regardless of the values of k and Δf. Moreover, it is noticeable that the RIP3 at f_RF equals 0.4 π consistently surpasses that at f_RF = 1π. It's notable that when α changes, the shape of the RIP3 remains consistent.

Fig. 5. Calculated RIP3 respect to k and RF frequency shift Δf for different α and RF frequency cases.

Download Full Size | PDF

4. Transmission system demonstration

Figure 6 delineates the simplified model under consideration, wherein an optical up-conversion scheme involving a phase modulator and ring is employed to generate an optical signal. The inset spectrum are the frequency spectra in the electrical domain at different positions in the system. The phase modulator and ring combination are biased at V_π/2, where V_π represents the modulator switching voltage. The system is driven by an RF oscillator. The 2f_RF frequency undergoes filtration through two low pass filters (LPFs) successively. Following the demodulation, the system's performance is evaluated. The inset spectrum illustrates the received power spectral density (PSD) of an RF signal with a frequency of 25 GHz considering a DC of 0.5 V.

Fig. 6. Schematic model of PM-RR system.

Download Full Size | PDF

By implementing the system in depicted Fig. 6, we validate that the proposed PM-RR can modulate QPSK, 16QAM and 16QAM_OFDM signal using VPI. Figure 7 shows the output eye diagram and constellation when setting the OSNR as 30 dB. The eye diagrams exhibit significant openness, particularly in the QPSK data transmission system, with the Quality Factor (Q) reaches 28.6 dB as illustrated in Fig. 7(b). Here, the chosen RF frequency f_RF is 25 GHz.

Fig. 7. Eye diagram and constellation of (a) 16QAM (b) QPSK (c) 16QAM_OFDM transmission system.

Download Full Size | PDF

In Fig. 8, we individually transmit QPSK, 16QAM and 16QAM_OFDM signal into the designed systems. Subsequently, we compare the Q factor of the PM-RR and MZM transmission systems. Figure 8 illustrates that the performance of the PM-RR as an electrical-optical convertor surpasses that of MZM system. By optimizing the parameters of the ring, superior results can be obtained using PM-RR compared to MZM. In this case, we choose the relatively high RIP3 scenario, specifically with k = 0.976 and Δf = 0.71π, to mitigate the nonlinearity effect. In Fig. 8(a), the Optical Signal-to-Noise Ratio (OSNR) is set to 20 dB. For the OFDM simulation, the configuration includes 194 subcarriers with 60 OFDM symbols. Among these, the central 154 subcarriers are loaded with 16 QAM modulated data, incorporating an 8-point cyclic prefix (CP). The optical carrier wavelength is set at 1550 nm and the bandwidth of OFDM signal is 10 GHz, carrying a data rate of 20 Gbit/s for one polarization. For the transmission of QPSK and 16QAM signals, the DAC rate is set at 10 GHz. When varying the RF frequency, the performance of MZM remains stable at 19.53 dB for QPSK transmission, 18.6 dB for 16QAM transmission and 17.52 dB for 16QAM_OFDM transmission. In the PM-RR system, the Q decreases as the RF frequency increases, reaching the minimum value of Q when f_RF = 1π. This is attributed to the fact that the nonlinearity of the PM-RR system attains its maximum value at this particular RF frequency, as depicted in Fig. 5. As the RF frequency varies within the range of 50% to 16% of the f_FSR, the Q of PM-RR system demonstrates an improvement of nearly 0.25 dB. In Fig. 8(b), the spectral noise density is maintained at 4 × 10⁻⁷ A/Hz^(1/2), and the RR is configured in the case of maximum average 3 dB transmission power. Specifically, k is set to 0.724, and Δf is set to 0.26π. Q increases with the rise in RF frequency in this configuration. A notable observation in Fig. 8 is that the performance of the PM-RR system consistently outperforms that of MZM when the right parameters of RR are selected.

Fig. 8. Compare the Q factor between MZM and PM-RR (a) OSNR is set as 20 dB (b) noise spectral noise density as 4 × 10⁻⁷ A/Hz^(1/2) for QPSK, 16 QAM and 16QAM-OFDM transmission (DACrate is 5e9 bit/s).

Download Full Size | PDF

Figure 9 illustrates the Q factor with respect to the DAC rate for both the PM-RR and MZM systems. In the case where the RF frequency is 25 GHz. The Q decreased with the increasing of DAC rate for MZM system. Conversely, for the PM-RR system, the Q initially increases, and then drops as the DAC rate ranges from 5e9 bit/s to 20e9 bit/s.

Fig. 9. Q factor as a function of DAC rate for PM-RR and MZM systems.

Download Full Size | PDF

The most notable advantage of the PM-RR system lies in its controllable phase response . This characteristic can be effectively utilized for compensating the CD in single mode fiber (SMF). By adjusting the coupling coefficient k, the phase jump can be altered, offering an effective method to control the phase response, as depicted in Fig. 1(b). This adjustability proves valuable in compensating for the dispersion of the Single Mode Fiber (SMF), assuming a dispersion value of 16 ps/nm/km. For instance, when the fiber is 10 km, the time delay caused by the dispersion is 32 ps, namely the phase variation is 1.6π considering the RF frequency Δf_RF is 25 GHz. In this case, the k of the ring is configured to be 0.5, resulting in an approximate phase jump of 1.6π. This adjustment is tailored to align with the phase alterations induced by the dispersion within the SMF. The selected values for the coupling coefficient k in the simulation, varying with the length of the fiber, are depicted in the illustration. φ is consistently set at -π/2 in all cases to achieve the different phase jumps in two sidebands. Figure 10 illustrates the Q factor of the PM-RR and MZM systems relative to the maximum Q, considering QPSK, 16QAM and 16QAM_ OFDM signals. The Q of the MZM system exhibits a damped sine curve pattern with the escalation of fiber length. Conversely, the Q of the PM-RR system decreases slowly because of the internal bandwidth dispersion caused by phase change in the high-frequency and low-frequency components within the signal bandwidth. Remarkably, the Q of the PM-RR system surpasses that of MZM system, with the advantage becoming more pronounced as the fiber length increases. When the fiber length reaches 100 km, the Q in the proposed PM-RR system is 40% better than that in MZM system. In practical, the fiber length in the system can be further increased when the phase jump in ring resonator is changed accordingly to compensate the accumulated CD in the fiber.

Fig. 10. Relative Q factor of PM-RR and MZM system as a function of fiber length (RF frequency is 5e9 bit/s).

Download Full Size | PDF

5. Conclusions

In this study, a thorough investigation and demonstration have been conducted on an efficient RR-based modulation transformer. The primary objective is to achieve the conversion from phase modulation to intensity modulation. The RR is systematically optimized by considering the power coupler coupling coefficient k and roundtrip optical phase shift φ. To attain the maximum 3-dB average power, the selected parameters include k = 0.724, Δf = 0.26π. Additionally, for achieving the maximum 3-dB bandwidth, the chosen parameters are k = 0.164, Δf = 0.05 π. To minimize dispersion, the simulation employs k = 0.89 and Δf = 1π for the RR. For maximizing the RIP3, k is set to 0.976, and Δf is configured as 0.71π. Through meticulous tuning of the phase response, the Q of the PM-RR system demonstrates an enhancement of over 40% when the fiber length extends to 100 km, in comparison to the MZM system. The simulation results unequivocally highlight that the proposed PM-RR device exhibits a straightforward architecture, stable transformation, effective carrier suppression and commendable overall performance.

Funding

111 Project (B17035); Aviation Industry Corporation of China (20180881005); Research Plan Project of National University of Defense Technology (China) (ZK18-01-02); National Major in High Resolution Earth Observation (China) (11-H37B02-9001-19/22, 30-H30C01-9004-19/21, GFZX0403260313); National Key Research and Development Program of China (2018YFA0701903).

Acknowledgments

We thank VPI photonics for the use of their simulator, VPI transmission Maker WDM V9.1.

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. C. Lim, A. Nirmalathas, M. Bakaul M, et al., “Fiber-wireless networks and subsystem technologies,” J. Lightwave Technol. 28(4), 390–405 (2010). [CrossRef]

2. G. Chang, J. Yu, Z. Jia, et al., “Novel optical-wireless access network architecture for simultaneously providing broadband wireless and wired services,” in Proc. OFC (2006) Paper OFM1.

3. S. Weber, J. G. Andrews, and N. Jindal, “The effect of fading, channel inversion, and threshold scheduling on ad hoc networks,” IEEE Trans. Inf. Theory 53(11), 4127–4149 (2007). [CrossRef]

4. C. Lin, C. Wei, and M. Chao, “Phase noise suppression of optical OFDM signals in 60 GHz RoF transmission system,” Opt. Express 19(11), 10423–10428 (2011). [CrossRef]

5. K. Zhu, H. Ou, Y. Hu, et al., “Tunable single-bandpass microwave photonic filters with high Q factor or flat-top shape based on cascaded optical structures,” in Proceedings of Optical Fiber Communication & Optoelectronic Exposition & Conference (IEEE, 2008), pp. 1–3.

6. C. Lin, J. Chen, P. Peng, et al., “Hybrid optical access network integrating fiber-to-the-home and radio-over-fiber systems,” IEEE Photonics Technol. Lett. 19(8), 610–612 (2007). [CrossRef]

7. M. J. LaGasse and S. Thaniyavaru, “Bias-free high-dynamic-range phase-modulated fiber-optic link,” IEEE Photonics Technol. Lett. 9(5), 681–683 (1997). [CrossRef]

8. K. Williams and R. D. Esman, “Stimulated Brillouin scattering for improvement of microwave fibre-optic link efficiency,” Electron. Lett. 30(23), 1965–1966 (1994). [CrossRef]

9. S. Tonda-Goldstein, D. Dolfi, J. Huignard, et al., “Stimulated Brillouin scattering for microwave signal modulation depth increase in optical links,” Electron. Lett. 36(11), 944–946 (2000). [CrossRef]

10. M. Lagasse, W. Charczenko, M. Hamilton, et al., “Optical carrier filtering for high dynamic range fibre optic links,” Electron. Lett. 30(25), 2157–2158 (1994). [CrossRef]

11. R. D. Esman and K. J. Williams, “Wideband efficiency improvement of fiber optic systems by carrier subtraction,” IEEE Photonics Technol. Lett. 7(2), 218–220 (1995). [CrossRef]

12. F. Zeng and J. Yao, “Ultrawideband impulse radio signal generation using a high-speed electrooptic phase modulator and a fiber-Bragg-grating-based frequency discriminator,” IEEE Photonics Technol. Lett. 18(19), 2062–2064 (2006). [CrossRef]

13. H. Toda, T. Yamashita, T. Kuri, et al., “25-GHz channel spacing DWDM multiplexing using an arrayed waveguide grating for 60-GHz band radio-on-fiber systems,” International Topical Meeting on Microwave Photonics (IEEE, 2003), pp. 287–290.

14. K. V. Baliž, A. Debevc, J. Krč, et al., “Power-fading-mitigation approach in an intensity-modulated radio-over-fiber link using a single integrated micro-ring resonator,” Opt. Fiber Technol. 73, 103008 (2022). [CrossRef]

15. Y. Chen, K. Chen, J. Yan, et al., “Integrated Tunable Sideband Suppressor for Optical Carrier Reuse in Millimeter-Wave Radio-Over-Fiber Transmission,” J. of Lightwave Technol. 40(9), 2696–2702 (2022). [CrossRef]

16. L. Zhuang, C. G. H. Roeloffzen, R. G. Heideman, et al., “Ring resonator-based on-chip modulation transformer for high-performance phase-modulated microwave photonic links,” Opt. Express 21(22), 25999–26013 (2013). [CrossRef]

17. C. K. Madsen and J. Zhao, Optical Filter Design and Analysis: A Signal Processing Approach (Wiley-Interscience, 1999).

18. L. Zhuang, C. Zhu, B. Corcoran, et al., “Sub-GHz-resolution C-band Nyquist-filtering interleaver on a high-index-contrast photonic integrated circuit,” Opt. Express 24(6), 5715–5727 (2016). [CrossRef]

19. Y. Tao, H. Shu, X. Wang, et al., “Hybrid-integrated high-performance microwave photonic filter with switchable response,” Photonics Res. 9(8), 1569 (2021). [CrossRef]

20. W. Zhang and J. Yao, “On-chip silicon photonic integrated frequency-tunable bandpass microwave photonic filter,” Opt. Lett. 43(15), 3622–3625 (2018). [CrossRef]

21. Y. Yao, Y. Zhao, J. Dong, et al., “Silicon integrated frequency-tunable microwave photonic bandpass filter,” IET Optoelectron. 17(2-3), 70–76 (2023). [CrossRef]

22. H. Yan, Y. Xie, L. Zhang, et al., “Wideband-tunable on-chip microwave photonic filter with ultra-high-Q U-bend-Mach-Zehnder-interferometer-coupled microring resonators,” Laser Photonics Reviews 17(11), 2300347 (2023). [CrossRef]

Phase to intensity modulation and fading compensation using a single ring resonator for radio-over-fiber links

Abstract

1. Introduction

2. Device and system principles

2.1 Ring resonator performance

2.2 System performance

3. System optimization

3.1 Bandwidth and average power

3.2 Nonlinearity

4. Transmission system demonstration

5. Conclusions

Funding

Acknowledgments

Disclosures

Data availability

References

Data availability

Cited By

Figures (10)

Equations (4)

Optics Express