Open Access
Add to BibSonomy
Abstract
An optical format interconversion scheme between on-off keying (OOK) and quadrature phase shift keying (QPSK) is proposed and verified in this paper. The conversion system mainly consists of a coherent vector combiner and a reconfigurable two-dimensional (2D) vector mover. As a key element of the proposed conversion system, the 2D vector mover is implemented by a non-degenerate phase-sensitive amplifier (PSA). The operating principle and theoretical derivations of the PSA-based 2D vector mover are fully introduced. The reconfigurable transfer characteristics of the vector mover are analyzed under different parameter settings to exhibit the flexible 2D moving function. The signal constellations, eye diagrams, spectrum, error vector magnitudes, and bit error ratios are estimated and depicted to validate the proposed idea. With the input signal-to-noise ratios of 20 dB and 25 dB, error-free conversions are achieved between 50G Baud OOK and QPSK. The scheme proposed in this paper fills the lack of the one-to-one interconversion between OOK and QPSK, and has potential applications in optical interconnect nodes, across-dimensional optical transmissions, and flexible optical transceivers.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The booming mobile communication networks continually spawn emerging high-traffic applications, such as autonomous driving, extended reality, and metaverse, while also promote optical communication networks to provide stronger transmission capabilities and more flexible network management and control [1]. Numerous research schemes have been proposed to improve the optical transmission capacity in terms of multiplexing technology, bandwidth and spectral efficiency [2], including the studies employing multi-core fibers [3,4], multi-band amplifiers [5,6], high-bandwidth devices [7–9], and advanced modulation formats [10,11], etc. The coexistence of various transmission technologies causes the difficulty in flexible interconnection between different optical networks, including transoceanic networks, backbone networks, metro networks, access networks, and data centers. In addition, each modulation format has its specific advantages and shortcomings, so that each optical network cannot rely on just one modulation format. Current transmission schemes can be divided into in-phase and quadrature modulation-coherent detection (IQ-coherent) and intense modulation-direct detection (IM-DD), which employs multi-dimensional and one-dimensional modulation formats, respectively. In response to these issues, optical format conversion technology has gradually attracted more attention [12] and many conversion schemes have been proposed, mainly including one-to-one conversion, aggregation, and de-aggregation, aiming to realize flexible translation between different transmission schemes.
Pulse amplitude modulation (PAM) and quadrature amplitude modulation (QAM) are essential formats in IM-DD and IQ-coherent transmission schemes, among which, on-off keying (OOK) and quadrature phase shift keying (QPSK) formats have been extensively studied in various transmission scenarios. OOK is mainly employed in short-reach scenarios, also in free-space optical [13,14] and underwater transmissions [15]. QPSK is the standard format of long-haul transmissions and has been tried for satellites [16] and submarine communications [17,18]. OOK has the advantages of system simplicity and low-cost, while QPSK can achieve longer transmission distance and higher spectral efficiency. Realizing flexible optical interconnection between OOK and QPSK can better utilize their respective strengths, and effectively promote the transmission performance of the optical networks employing these two formats. The proposed conversion schemes between OOK and QPSK mainly focus on the signal aggregations and de-aggregations, such as 2$\times$OOK-to-QPSK aggregation [19,20], 4$\times$OOK-to-dual polarization (DP) QPSK aggregation [21], and DP QPSK-to-4$\times$OOK de-aggregation [22], also some simultaneous aggregation and de-aggregation schemes, including OOK/binary phase shift keying (BPSK)-to-QPSK-to-OOK/BPSK conversion [23] and OOK/QPSK-to-8QAM-to-OOK/QPSK conversion [24,25]. Through analyzing the existing schemes, we can find that there is rarely no one-to-one conversion scheme between OOK and QPSK, which is a necessary conversion function. The main reason for this issue is that the QPSK and OOK have different modulation levels. Realizing the one-to-one conversion between the formats with different modulation levels is a challenging problem because the conversion system needs to deal with the different signal symbol rates while achieving the format conversion. Moreover, like the simultaneous aggregation and de-aggregation schemes, a comprehensive one-to-one conversion scheme consists of OOK-to-QPSK and QPSK-to-OOK conversions is more valuable and has more potential applications than a one-way conversion. Therefore, a comprehensive one-to-one interconversion between OOK and QPSK is an important and unsettled question.
In this paper, an optical interconversion scheme between OOK and QPSK under different signal quality is proposed. 10G/30G/50G Baud OOK-to-QPSK and QPSK-to-OOK conversions are implemented and the signal constellations, eye diagrams, spectrum, error vector magnitudes (EVMs), and bit error ratios (BERs) are calculated to measure the conversion system performances. The proposed system achieves the format conversions by employing a delay interferometer (DI)-based coherent vector combiner and a non-degenerate phase-sensitive amplifier (PSA)-based reconfigurable two-dimensional (2D) vector mover. The input OOK signal is firstly converted into a PAM3 signal through the vector combiner, and then moved into a QPSK signal by the vector mover. After being combined with amplifier spontaneous emission (ASE) noise, the QPSK signal is again converted into an OOK by another vector mover. The principle of reconfigurable 2D vector mover is introduced in detail and its power gain and phase transfer characteristics under different parameter settings are analyzed and depicted. With the input optical signal-to-noise ratio (OSNR) of 20dB, the 10G/30G/50G Baud QPSKs converted from OOKs have the receiver OSNR of 10.3 dB, 16 dB, and 19.7 dB, respectively, at the BER of ${10^{ - 3}}$. With the input OSNR of 20 dB, at the BER of ${10^{ - 3}}$, the receiver OSNR of the 10G/30G/50G Baud OOKs converted from QPSKs are 12.5 dB, 17.4 dB, and 20.2 dB, respectively. The proposed conversion scheme fills the lack of one-to-one conversion between OOK and QPSK, and can be applied in optical interconnect nodes, across-dimensional optical transmissions, and flexible optical transceivers.
2. Operation principle
The proposed interconversion scheme concept graph is shown in Fig. 1. The interconversion between OOK and QPSK is a typical function of the conversions between one-dimensional (1D) and 2D optical modulation, and also an important ability to flexible bridging the short-reach and long-haul transmission networks. For OOK-to-QPSK conversion, the 1D-to-2D conversion is implemented by coherent addition and 2D vector moving. The QPSK-to-OOK conversion is accomplished through 2D vector moving processing. The coherent addition and vector moving processing are provided by a vector combiner and a vector mover, and the detailed processing principles are given in following subsections.
2.1 Coherent vector combiner-based OOK-to-PAM3 conversion
To implement the OOK-to-QPSK conversion, the OOK signal needs to be converted into a PAM3 signal firstly, and the vector combiner-based conversion setup is shown in Fig. 2. After being injected into the DI, the input OOK signal is separated into OOK1 and OOK2 by a 50:50 power splitter. An n-bit time-delay (n is an integer) and a phase shifter are placed in the lower branch of DI to erase the correlation and add a $\pi /2$ relative phase shift between OOK2 and OOK1. Then OOK1 and OOK2 are combined by a 50:50 power splitter, the combining process can be expressed as:
2.2 Reconfigurable 2D vector mover-based conversions
The non-degenerate PSA is proposed and mainly employed in low-noise amplification [26,27] and 1D vector moving [28–30]. In this paper, a non-degenerate PSA-based 2D vector mover structure is employed to implement the PAM3-to-QPSK and QPSK-to-OOK conversions, and its system and spectrum setup are shown in Fig. 3(a). Two pumps (P1, P2) and the input signal (S) are injected into the non-degenerate PSA and occur four-wave mixing (FWM) effect. The PSA adds a coherent vector on the input signal to realize vector moving process, which can be a two-dimensional moving operation by control the amplitude and phase of the generated coherent vector. The PSA transfer function can be expressed as [27]:
Fig. 3. 2D vector mover-based conversion scheme. (a) Setup and spectrum of non-degenerate PSA; (b) and (c) are the constellations of PAM3-to-QPSK and QPSK-to-OOK conversions. HNLF, high nonlinear fiber; BPF, band pass filter.
The reconfigurable 2D vector mover is employed by this paper to implement PAM3-to-QPSK and QPSK-to-OOK conversions, as shown in Figs. 3(b) and (c). For PAM3-to-QPSK conversion, a vector with the angle of $5\pi /4 + 2n\pi$ is required to combine with PAM3 signal to get a QPSK signal. For QPSK-to-OOK conversion, a vector with the angle of $n\pi /2$ is required to make the QPSK I or Q component constellation points have the same sign, because both the QPSK I or Q components carry the complete information of the original input OOK. Through tuning the relative phase and relative power among input signals and input pumps, the 2D vector mover can realize the above moving process, so the interconversions between OOK and QPSK can be implemented by the proposed system. The corresponding relations in these conversions are listed in Table 1, we can see that the conversions are feasible in theory and information integrity can be guaranteed through the across-dimensional conversions process.
Table 1. Corresponding relations between OOK and QPSK conversions.a
The proposed scheme raises a preliminary solution for the conversion between the formats with different modulation levels. The converted QPSK carries one channel redundant information, while it also helps to achieve the one-to-one conversions of OOK-to-QPSK and QPSK-to-OOK. The conversion scheme does not improve the system spectral efficiency, but enables the transmission system to simultaneously achieve long transmission distance and low system cost and complexity, which can also be regarded as utilizing the advantages of OOK and QPSK. The realization of phase-sensitive amplification requires that the relative phase among input signal carrier and pumps is a constant value, which means the PSA input waves are phase-locked. The phase-locked harmonics generation is critical in PSA practical implementation, there are three main methods currently used: FWM-based local pump generation [31], cross-phase modulation-based local pump generation [32], and optical frequency comb (OFC)-based method [33]. Moreover, there will be slow phase distortion among the PSA input waves, caused by the acoustic and thermal effects, and it may affect the PSA performances. The coherence maintaining is also very important in the experimental process, which is usually achieved by employing active phase-locking loops [31,34] or on-chip schemes [35,36].
3. Verifications and discussions
The system verifications contain phase-locked harmonics generation, 2D vector mover transfer characteristics and format conversions performances. The detailed system setups, parameter settings and performance discussions are introduced in following subsections.
3.1 Phase-locked harmonics generation
The phase-locked harmonics generation in this paper is realized by a Mach-Zehnder modulator (MZM)-based OFC generation scheme. As shown in Fig. 4(a), a continuous wave (CW) laser (frequency: 193.1 THz, 100 mW) is injected into a MZM ( ${V_{pi - RF}}$: 8.8 volt, ${V_{pi - DC}}$: 8.8 volt). A radio frequency (RF) signal (frequency: 100 GHz, ${V_{RF}}$: 5 volt, ${V_{DC}}$: 0 volt) is employed to drive the MZM to generate an OFC. The electronic filed ${E_{OFC}}$ of the MZM-based generated OFC can be expressed as:
Fig. 4. (a) Setup of phase-locked harmonics generation scheme. (b) Spectrum of the phase-locked harmonics. CW, continuous wave; MZM, Mach-Zehnder modulator; RF, radio frequency signals; PF, programmable filter.
3.2 Reconfigurable 2D vector mover transfer characteristics
The PSA-based reconfigurable 2D vector mover setup is shown in Fig. 3(a), its power gain and phase transfer characteristics are estimated. The phase-locked pump1 (frequency: 193.0 THz, power: 40 mW), pump2 (frequency: 193.1 THz, power: 55 mW) and signal carrier (frequency: 193.2 THz) are injected into the vector mover. The vector mover contains a high nonlinear fiber (HNLF, length: 400 m, nonlinear coefficient: 13.2 ${W^{ - 1}}k{m^{ - 1}}$) and an optical band pass filter (BPF, center frequency: 193.2 THz, bandwidth: 20 GHz). The signal carrier is modulated to carry the phase information of $\left ( { - 1.5\pi \sim 3.5\pi } \right )$ to measure the vector mover power gain and output signal phase versus input signal phase characteristics. The signal input power is swept from 0.5 mW to 4 mW, and the pump2 is added variable phase shifts to depict the 2D vector moving processing.
Figures 5(a) and (b) show the constellations of input and output optical vectors processed by the vector mover. We can see that the input vectors experiences different vector moving processes because of the different signal powers and phase shifts added to pump2. With the signal power of 4 mW and 258 degree phase shift added to pump2, the input signal is added a 0 degree phase shift. With the signal power of 0.5 mW and 98 degree phase shift added to pump2, the input signal is added a 45 degree phase shift. Moreover, the signal with the power of 0.5 mW is moved to a farther position comparing with the signal with the power of 4 mW. This phenomenon is consistent with the description in Eq. (5), the $m\exp \left ( {j\theta } \right )$ item can realize two-dimensional vector moving process by control the relative power and relative phase among vector mover input waves, besides, the smaller input signal power makes the $m$ item larger, which leads the vector moving process has more effect on input signal. Figures 5(c) and (d) depict the vector mover power gain and output signal phase versus input signal phase characteristics, the signal with power of 0.5 mW has the power gain variation range of 43 dB and the output phase change variation of $0.9\pi$. While the signal with the power of 4 mW has the power gain variation range of 5.9 dB and the output phase variation range of $0.2\pi$. The significant differences among the signals with different input power in Figs. 5(c) and (d) are consistent with Eq. (6) and Eq. (7), which also tell that the PSA has more effect on the signal with smaller input power. Through the detailed analyses of transfer characteristics, it can conclude that the vector mover structure can provide flexible two-dimensional vector moving function by controlling the relative power and relative phase among input waves.
Fig. 5. 2D vector mover transfer characteristics. (a) and (b) are the input and output optical vectors constellations of the vector mover; (c) and (d) are the power gain and output signal phase versus input signal phase transfer characteristics, respectively.
3.3 Format conversions performances
The format conversion system setup is shown in Fig. 6, the signal carrier generated by OFC (frequency: 193.2 THz) is injected into a MZM (${V_{pi - RF}}$: 5 volt, ${V_{pi - DC}}$: 5 volt, ${V_{DC}}$: 2.5 volt). 10G/30G/50G Baud pseudo random bit sequences (PRBS) generated by a bit pattern generator are employed to drive the MZM to generate 10G/30G/50G Baud OOK signals. The input OOK is coupled with an ASE noise to control the OOK input OSNR. Then the OOKs are launched into a DI (1-bit delay and $\pi /2$ phase shift in lower branch) to implement OOK-to-PAM3 conversions. The PAM3-to-QPSK and QPSK-to-OOK conversions are realized by two identical vector movers, which both contain a HNLF (length: 400 m, nonlinear coefficient: 13.2 ${W^{ - 1}}k{m^{ - 1}}$) and a BPF (center frequency: 193.2 THz, bandwidth: 2 $\times$ symbol rate). The signals at the nodes of D and E in Fig. 6 are also combined with an ASE noise to control the signal OSNRs. The signals at the nodes of A, B, C, D and E are received and analyzed to measure the conversion performances. The signal constellations, eye diagrams, impact factors (IFs), and error vector magnitudes (EVMs) are calculated with ${2^{11}} = 2048$ signal symbols, and the BERs are calculated with ${2^{14}} = 16384$ signal symbols. The IF is defined as the ratio of the standard deviations between input and output signals.
Fig. 6. OOK-to-QPSK and QPSK-to-OOK conversion nodes setup. BPG, bit pattern generator; MZM, Mach-Zehnder modulator; ASE, amplified spontaneous emission noise; HNLF, high nonlinear fiber; BPF, band pass filter.
Figure 7 shows the constellations and eye diagrams of the format conversions process, the input OOKs and input QPSKs are both with the input OSNRs of 25 dB. The constellations and eye diagrams are depicted after amplitude normalization. The input 10G/30G/50G Baud OOK signals are firstly processed by the DI-based vector coherent combiner and converted into 10G/30G/50G Baud PAM3 signals, the PAM3 symbol rates can be observed from the eye diagrams. Then the converted PAM3 signals are launched into a non-degenerate PSA and moved into QPSK signals, the QPSK-C constellations in Fig. 7 can be viewed as normal and receivable QPSK constellations. The converted QPSK signals are combined with ASE noise and become the input QPSK signals for QPSK-to-OOK conversions. The 10G/30G/50G Baud input QPSK signals are processed by another PSA and moved into 10G/30G/50G Baud converted OOK signals. We can see from the OOK-C eye diagrams that the OOK-Cs have corresponding symbol rates and relatively high signal quality. The constellations and eye diagrams in Fig. 7 indicate that the proposed conversion system in this paper implements the OOK-to-QPSK and QPSK-to-OOK conversions.
Fig. 7. (a), (b) and (c) are the constellations and eye diagrams of the 10G/30G/50G Baud OOK-to-QPSK and QPSK-to-OOK conversions, respectively. PAM3-C, converted PAM3; OOK-C, converted OOK.
In order to quantitatively analyze the proposed conversion system performances, the signal EVMs and IFs are estimated and discussed with the input OSNR range of $\left ( {15dB \sim 30dB} \right )$. Figure 8(a) shows the signal EVM performances of OOK-to-QPSK conversions, we can see that the EVMs of input OOK and converted PAM3 signals are close with each other, because the OOK-to-PAM3 conversion process is linear. The converted QPSKs have higher EVMs than input OOKs which means the PSA causes extra noise while implementing conversions. For example, with the input OSNR of 22 dB, the EVMs of 10G/30G/50G Baud converted QPSKs are 2.3%, 1% and 2.1% higher than the corresponding input OOKs. Figure 8(b) depicts the amplitude and phase IFs of the OOK-to-QPSK conversion process, we can see that the signal amplitude and phase IFs are all negative within the input OSNR range of $\left ( {15dB \sim 30dB} \right )$. The negative IFs indicate that the signal amplitude and phase standard deviations get worse in the OOK-to-QPSK conversions, which is consistent with the EVM analyzes in Fig. 8(a). Figure 9(a) shows the EVM performances in QPSK-to-OOK conversions, with the input OSNR of 22 dB, the EVMs of 10G/30G/50G Baud converted OOKs are 1.4%, 3.3% and 3.8% lower than the input QPSKs. With the input OSNR range of $\left ( {15dB \sim 30dB} \right )$, the EVMs of the converted OOKs are all lower than input QPSKs, which is mainly caused by the reduction of modulation level. The signal amplitude and phase IFs in QPSK-to-OOK conversions are depicted in Fig. 9(b), the amplitude and phase IFs are all positive, which indicates that the signal amplitude and phase standard deviations get smaller in the QPSK-to-OOK conversions. The IF characteristics in Fig. 9(b) is also consistent with the EVM analyzes in Fig. 9(a).
Fig. 8. (a) EVM versus input OSNR curves of OOK-to-QPSK conversions. (b) Amplitude and phase impact factor versus input OSNR curves of OOK-to-QPSK conversions. PAM3-C, converted PAM3; QPSK-C, converted QPSK; IF, impact factor.
Fig. 9. (a) EVM versus input OSNR curves of QPSK-to-OOK conversions. (b) Amplitude and phase impact factor versus input OSNR curves of QPSK-to-OOK conversions. OOK-C, converted OOK; IF, impact factor.
The BERs of OOK-to-QPSK conversions are analyzed in Fig. 10, the BER versus receiver OSNR curves of the converted QPSKs are depicted to verify the conversion system performances. The BERs of back-to-back (B2B) OOKs and reference QPSKs are also estimated for comparisons and discussions. With the input OSNR of 20 dB, at the BER of ${10^{ - 3}}$, the receiver OSNR of 10G/30G/50G Baud converted QPSKs are 10.3 dB, 16 dB and 19.7 dB, respectively. Under the same situation, the receiver OSNR of 10G/30G/50G Baud B2B OOKs are 1.4 dB, 2 dB and 2.9 dB lower than converted QPSKs, and the reference QPSKs have similar BER performances with converted QPSKs. With the input OSNR of 25 dB, at the BER of ${10^{ - 3}}$, 10G/30G/50G Baud converted QPSKs have the receiver OSNR of 10.3 dB, 14.7 dB and 17.5 dB, respectively. Under the same situation, the receiver OSNR of 10G/30G/50G Baud B2B OOKs are 2.2 dB, 1.2 dB and 1.6 dB lower than converted QPSKs. The BER performances of converted QPSKs get worse comparing with the B2B OOKs mainly because of the increase of the modulation level and the generated noise in the conversion processes.
Fig. 10. BER versus receiver OSNR curves of OOK-to-QPSK conversions. OOK-B2B, back-to-back OOK; QPSK-Ref, QPSK for reference; QPSK-C, converted QPSK; Rec. OSNR, receiver OSNR.
The BERs of QPSK-to-OOK conversions are analyzed in Fig. 11, the system origin input OOK OSNRs are set to be 25 dB. The BERs of converted OOKs, reference OOKs and B2B QPSKs are estimated for the BERs performances discussions. With the input OSNR of 20 dB, 10G/30G/50G Baud converted OOKs show the receiver OSNRs of 12.5 dB, 17.4 dB and 20.2 dB, at the BER of ${10^{ - 3}}$. Under the same situation, the receiver OSNRs of 10G/30G/50G Baud B2B QPSKs are 1.6 dB, 0.9 dB and 0.2 dB lower than the converted OOKs, and the receiver OSNRs of 10G/30G/50G Baud reference OOKs are 4.1 dB, 2.7dB and 2.5 dB lower than converted OOKs. With the input OSNR of 25 dB, at the BER of ${10^{ - 3}}$, 10G/30G/50G Baud converted OOKs show the receiver OSNRs of 12 dB, 17 dB and 19 dB, respectively. The BER performances under the input OSNR of 25 dB are similar with the performances under the input OSNR of 20 dB. The BER performances of converted OOKs get worse comparing with the B2B QPSKs mainly because of the generated noise in the conversion processes. The BER performances analyzes in Fig. 10 and Fig. 11 indicate that the converted signals can be successfully received and the proposed scheme in this paper can implement the OOK-to-QPSK and QPSK-to-OOK conversions.
Fig. 11. BER versus receiver OSNR curves of QPSK-to-OOK conversions. QPSK-B2B, back-to-back QPSK; OOK-Ref, OOK for reference; OOK-C, converted OOK; Rec. OSNR, receiver OSNR.
4. Conclusion
This paper proposes an optical format interconversion scheme between OOK and QPSK signals. 10G/30G/50G Baud OOK-to-QPSK and QPSK-to-OOK conversions are simulated and analyzed. The signal constellations, eye diagrams, EVMs, IFs and BERs are estimated and depicted to verify the conversion performances. The proposed scheme mainly implements a coherent vector combiner-based OOK-to-PAM3 conversion, and 2D vector mover-based PAM3-to-QPSK and QPSK-to-OOK conversions. The theoretical derivations of the MZM-based OFC generation, DI-based coherent vector combining and PSA-based reconfigurable 2D vector moving are introduced in detail. Moreover, the reconfigurable transfer characteristics of the 2D vector mover are discussed and analyzed under different parameter settings. The results show that the conversion system is able to achieve 50G Baud error-free OOK-to-QPSK and QPSK-to-OOK conversions. The proposed system realizes the interconversions between OOK and QPSK, and has potential applications in optical interconnect nodes, across-dimensional optical transmissions, and flexible optical transceivers.
Funding
National Natural Science Foundation of China (62001046); National Key Research and Development Program of China (2019YFB2203104, 2020YFB2205801); Fundamental Research Funds for the Central Universities (2022RC02); State Key Laboratory of Information Photonics and Optical Communications (IPOC2022ZT08).
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. T. Xu, G. Liga, and P. Bayvel, “High-order modulation formats, constellation design, and digital signal processing for high-speed transmission systems,” in Optical Fiber Telecommunications VII (Elsevier, 2020), pp. 371–418.
2. P. J. Winzer, “Fibre-optic transmission systems from chip-to-chip interconnects to trans-oceanic cables,” in Handbook of Laser Technology and Applications (CRC Press, 2021), pp. 427–441.
3. M. Mazur, L. Dallachiesa, N. K. Fontaine, R. Ryf, E. Börjeson, H. Chen, H. Sakuma, T. Ohtsuka, T. Hayashi, T. Hasegawa, T. Hidehisa, D. T. Neilson, and P. Larsson-Edefors, “Real-time transmission over 2×55 km all 7-core coupled-core multi-core fiber link,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th4A–1.
4. R. S. Luis, B. J. Puttnam, G. Rademacher, Y. Awaji, and H. Furukawa, “372 tb/s unrepeatered 213 km transmission over a 125 μm cladding diameter, 4-core mcf,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper M4B–3.
5. T. Kobayashi, S. Shimizu, M. Nakamura, T. Umeki, T. Kazama, J. Yoshida, S. Takasaka, Y. Tatamida, H. Kawakami, F. Hamaoka, M. Nagatani, H. Yamazaki, K. Watanabe, T. Saida, and Y. Miyamoto, “50-tb/s (1 tb/s× 50 ch) wdm transmission on two 6.25-thz bands using hybrid inline repeater of ppln-based opas and incoherent-forward-pumped dra,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th4A–8.
6. A. Souza, N. Costa, J. Pedro, and J. Pires, “Raman amplification for simplified channel provisioning in wide-band optical networks,” in 2022 Optical Fiber Communications Conference and Exhibition (OFC) (IEEE, 2022), pp. 1–3.
7. H. Asakura, K. Nishimura, S. Yamauchi, K. Iwamoto, Y. Nakai, Y. Yamaguchi, K. Tani, R. Nakajima, and K. Naoe, “384-gb/s/lane pam8 operation using 76-ghz bandwidth ea-dfb laser at 50°c with 1.0-vpp swing over 2-km transmission,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th4C–4.
8. M. Xu, F. Pittalà, J. Tang, Y. Zhu, M. He, W. C. Ng, Z. Ruan, X. Tang, M. Kuschnerov, L. Liu, S. Yu, B. Zheng, and X. Cai, “Thin-film lithium niobate dp-iq modulator for driverless 130 gbaud 64 qam transmission,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th1J–2.
9. H. Ji, Z. Wang, X. Li, J. Li, R. R. Unnithan, W. Hu, Y. Su, and W. Shieh, “Silicon photonics 630-gb/s complementary polarization-diversity coherent receiver with 9-mrad/s polarization tracking speed for self-coherent homodyne detection,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th4B–7.
10. S. Xing, G. Li, J. Chen, J. Zhang, N. Chi, Z. He, and S. Yu, “First demonstration of ps-qam based flexible coherent pon in burst-mode with 300g peak-rate and record dynamic-range and net-rate product up to 7, 104 db·gbps,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th4A–4.
11. D. Che, “Ultra-low-complexity map demapper for bandwidth-limited pluggable coherent optics beyond 800g,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th2A–33.
12. A. E. Willner, A. Fallahpour, K. Zou, F. Alishahi, and H. Zhou, “Optical signal processing aided by optical frequency combs,” IEEE J. Sel. Top. Quantum Electron. 27(2), 1–16 (2021). [CrossRef]
13. V. Mai and H. Kim, “Variable focus lens-based beam steering and divergence control for wdm free-space optical communication,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper M1C–6.
14. M. N. Thi, V. Mai, and H. Kim, “Seven-aperture direct-detection receiver for free-space optical communication systems,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper Th2A–28.
15. T. Wang, B. Wang, L. Liu, R. Zhu, L. Wang, C. Tong, Y. Song, and P. Zhang, “15 mbps underwater wireless optical communications based on acousto-optic modulator and nrz-ook modulation,” Opt. Laser Technol. 150, 107943 (2022). [CrossRef]
16. Y.-J. Cai, S.-H. Yu, Z.-W. Huang, Y.-W. Wang, P.-H. Ting, Y.-P. Lan, C.-J. Fang, H.-C. Lin, B.-C. Chen, and C.-T. Lin, “10 gbps laser communication for low earth orbit satellites with volterra and machine learning nonlinear compensation providing link budget up to 74 db,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper M1C–2.
17. M. Mazur, J. C. Castellanos, R. Ryf, E. Börjeson, T. Chodkiewicz, V. Kamalov, S. Yin, N. K. Fontaine, H. Chen, L. Dallachiesa, S. Corteselli, P. Copping, J. Gripp, A. Mortelette, B. Kowalski, R. Dellinger, D. T. Neilson, and P. Larsson-Edefors, “Transoceanic phase and polarization fiber sensing using real-time coherent transceiver,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper M2F–2.
18. A. C. Meseguer, J.-C. Antona, J. U. Esparza, A. Calsat, P. Plantady, A. Quintana, and V. Letellier, “Experimental assessment of capacity prediction from g-snr measurements for submarine systems,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper W4G–3.
19. Y. Zhan, M. Zhang, M. Liu, L. Liu, and X. Chen, “All-optical format conversion of nrz-ook to qpsk and 16qam signals via xpm in a soa-mzi,” Chin. Opt. Lett. 11(3), 030604 (2013). [CrossRef]
20. Y. Mao, B. Liu, J. Ren, X. Xu, X. Wu, and L. Jiang, “All-optical ook -to-qpsk modulation format conversion with wavelength multicasting based on cascaded soa configuration,” IEEE Access 8, 77843–77849 (2020). [CrossRef]
21. Y.-H. Wen and K.-M. Feng, “A simple nrz-ook to pdm rz-qpsk optical modulation format conversion by bidirectional xpm,” IEEE Photonics Technol. Lett. 27(9), 935–938 (2015). [CrossRef]
22. K. Iwama, N. Goto, H. Kishikawa, and J. Fujikata, “All-optical modulation format conversion from dqpsk to ook using cross-polarization modulation,” in 2021 IEEE Photonics Conference (IPC) (IEEE, 2021), pp. 1–2.
23. H. Liu, H. Wang, Z. Xing, and Y. Ji, “Simultaneous all-optical channel aggregation and de-aggregation based on nonlinear effects for ook and mpsk formats in elastic optical networking,” Opt. Express 27(21), 30158–30171 (2019). [CrossRef]
24. H. Liu, H. Wang, and Y. Ji, “Simultaneous all-optical channel aggregation and de-aggregation for 8qam signal in elastic optical networking,” IEEE Photonics J. 11(1), 1–8 (2019). [CrossRef]
25. H. Kishikawa, M. Uetai, and N. Goto, “All-optical modulation format conversion between ook, qpsk, and 8qam,” J. Lightwave Technol. 37(16), 3925–3931 (2019). [CrossRef]
26. V. Torres-Company, Z. Ye, P. Zhao, M. Karlsson, and P. A. Andrekson, “Ultralow-loss silicon nitride waveguides for parametric amplification,” in Optical Fiber Communication Conference (Optica Publishing Group, 2022), paper W4J–3.
27. P. A. Andrekson and M. Karlsson, “Fiber-based phase-sensitive optical amplifiers and their applications,” Adv. Opt. Photonics 12(2), 367–428 (2020). [CrossRef]
28. K. Yu, W. Yang, and Y. Yu, “Wavelength preserved phase erasure and psk to conventional ook data format conversion based on phase sensitive amplification,” Opt. Commun. 377, 139–144 (2016). [CrossRef]
29. J. Cui, Y. Ji, G.-W. Lu, H. Wang, and M. Zhang, “Phase-sensitive amplifier-based optical conversion for direct detection of complex modulation format to bridge long-haul transmissions and short-reach interconnects,” Opt. Express 28(2), 2349–2365 (2020). [CrossRef]
30. J. Cui, Y. Ji, G.-W. Lu, H. Wang, and M. Zhang, “2d-to-1d constellation reforming using phase-sensitive amplifier-based constellation squeezing and shifting,” Opt. Express 29(3), 3724–3737 (2021). [CrossRef]
31. R. Slavík, F. Parmigiani, J. Kakande, C. Lundström, M. Sjödin, P. A. Andrekson, R. Weerasuriya, S. Sygletos, A. D. Ellis, L. Grüner-Nielsen, D. Jakobsen, S. Herstrøm, R. Phelan, J. O’Gorman, A. Bogris, D. Syvridis, S. Dasgupta, P. Petropoulos, and D. J. Richardson, “All-optical phase and amplitude regenerator for next-generation telecommunications systems,” Nat. Photonics 4(10), 690–695 (2010). [CrossRef]
32. P. Guan, F. Da Ros, M. Lillieholm, N.-K. Kjøller, H. Hu, K. M. Røge, M. Galili, T. Morioka, and L. K. Oxenløwe, “Scalable wdm phase regeneration in a single phase-sensitive amplifier through optical time lenses,” Nat. Commun. 9(1), 1049 (2018). [CrossRef]
33. K. Bottrill, G. Hesketh, L. Jones, F. Parmigiani, D. Richardson, and P. Petropoulos, “Full quadrature regeneration of qpsk signals using sequential phase sensitive amplification and parametric saturation,” Opt. Express 25(2), 696–705 (2017). [CrossRef]
34. J. Kakande, R. Slavík, F. Parmigiani, A. Bogris, D. Syvridis, L. Grüner-Nielsen, R. Phelan, P. Petropoulos, and D. J. Richardson, “Multilevel quantization of optical phase in a novel coherent parametric mixer architecture,” Nat. Photonics 5(12), 748–752 (2011). [CrossRef]
35. W. Li, M. Lu, A. Mecozzi, M. Vasilyev, S. Arafin, D. Dadic, L. A. Johansson, and L. A. Coldren, “First monolithically integrated dual-pumped phase-sensitive amplifier chip based on a saturated semiconductor optical amplifier,” IEEE J. Quantum Electron. 52(1), 1–12 (2016). [CrossRef]
36. L. Coldren, W. Li, A. Mecozzi, M. Lu, S. Arafin, M. Vasilyev, D. Dadic, and L. Johansson, “Single-chip dual-pumped soa-based phase-sensitive amplifier at 1550nm,” in 2015 IEEE Summer Topicals Meeting Series (SUM) (IEEE, 2015), pp. 88–89.
