Open Access
Add to BibSonomy
Abstract
This paper proposes a reconfigurable all-optical signal processing node architecture with selective all-optical signal processing in one simple configuration without optical-electrical-optical (O/E/O) conversion at the heterogeneous network connection node or repeater node. Proof-of concept experiments of encryption-key data superposition for binary phase shift keying and quadrature phase shift keying show the nonlinear optical effect generation within 1.5-dB and 3.5-dB characteristic degradation. Self-phase modulation and stimulated Brillouin scattering caused by increasing the highly nonlinear fiber input of pump light is the primary cause of signal degradation in the experiment.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
All-optical signal processing (AOSP) based on nonlinear optical effects has been studied for many years as a technology capable of high-speed and wide-band processing, and its applications are diverse [1–3]. Recent applications have increased demands for low-latency and wide-band transmission, and there are increasingly many practical applications of AOSP. Many optical devices capable of generating non-linear optical effects have been reported, such as semiconductor optical amplifiers [4], quantum dot semiconductor optical amplifiers [5], silicon photonics [6,7], and highly nonlinear fibers (HNLFs) [8,9]. The HNLF has a drawback in that the nonlinear efficiency per length is relatively low, but has the advantages of low connection loss, wideband, and high-speed response compared with the single-mode fiber (SMF) used in a standard transmission line. The optical wavelength conversion technology that does not depend on the modulation has been developed and adopted for practical use to increase the conversion efficiency in the HNLF [10].
The most significant feature of the AOSP, such as the elimination of O/E/O conversion, can be utilized for the relay processing of optical nodes. The application target of the AOSP is an optical relay node in an optical metro network that requires high-speed and wide-band processing [11,12]. All-optical wavelength conversion, all-optical phase sensitive amplifier [13], and all-optical 2R regeneration [14] were reported as AOSP technologies for optical relay nodes. Recently, applications that require low latency, such as the next-generation wireless communication standard, have appeared. In the AOSP, even with a standard unit of nonlinear generation consisting of an erbium-doped fiber amplifier (EDFA) and HNLF, the light wave propagation delay remains at about 0.15 µs/unit and 4.9 µs/km: therefore, optical conversion can provide ultra-low latency processing. Under the condition that the fiber length of HNLF is within 20 km, the processing time of OSP using HNLF is shorter than the processing time of O/E/O conversion, which requires a delay of 100 µs or more without forward error correction (FEC) [15]. To accommodate mobile devices and various IoT applications requiring low latency, the movement toward complete optical processing of optical access networks and optical metro networks is accelerating. The AOSP approach is attractive for nodes connected to optical edge networks where traffic delays can occur. The realization of ultra-low latency systems is essential for next-generation optical access/metro networks for applications like mobile broadband, automatic guided vehicles, and industrial automation. For ultra-low latency applications such as industrial automation and rigorous cloud computing, the latency requirements are stringent within one millisecond [16].
AOSP is also effective in heterogeneous gateway NW nodes (HNGNs) that connect optical access and optical metro networks. As an AOSP technology for HNGNs, all-optical modulation format conversion has been reported. AOSP-based conversions from intensity-modulated signals to in-phase (I) and orthogonal (Q) modulated signals use nonlinear optical effects such as mutual phase modulation (XPM) and four-wave mixing (FWM) [17–20].
Many of the applications for AOSP using nonlinear optical effects reported so far pertain to conversion and regeneration, as mentioned above. Recently, optical data superposition [21,22] and optical encryption key superposition [23] have been demonstrated as new applications. The conventional AOSP node architecture is problematic because it can be applied only to a specific application due to the fixed optical device arrangement. Therefore, a flexible OSP node architecture that does not depend on the allocation and wavelength arrangement of the probe and pump light is required.
This study aims to extend the scope of optical encryption-key superposition proposed in the preliminary study [23] and to discover the problems caused. In this paper, we propose a reconfigurable all-optical signal processing (RAOSP) node architecture that can flexibly respond to applications by switching each optical device and wavelength with an optical space switch and a wavelength selection switch (WSS). The RAOSP node architecture is independent of applicable nodes and can be used for both HNGNs and optical relay nodes in metro networks. Furthermore, we also propose optical encryption key superimposition using XPM and π/2 phase shift to suppress excessive non-linear effects and save power. To confirm the feasibility of a quadrature phase shift keying (QPSK) system with optical encryption key superimposition, we conducted experimental verification of the RAOSP node architecture. This study is the first full-paper report on data or optical encryption key superposition.
2. Reconfigurable all-optical signal processing node architecture
Figure 1 shows the basic configuration of an RAOSP node. When an optical signal is input into the node, it is branched by an optical coupler. One is input to the monitor, and the other is input to the 2×2 optical switch. After adjusting the timing based on the optical signal using the electric signal acquired by the monitor, the electric signal is input to the LD, and the directly modulated light is input to the 2×2 optical switch. The optical switch selects the input used as the pump light. The light selected for the pump light is input to the 2×1 optical coupler after EDFA adjusts the gain. Simultaneously, light other than pump light is input to the 2×1 optical coupler as it is. The light combined by the optical coupler generates a nonlinear optical effect in HNLF, and the desired light is selected in the wavelength domain by the WSS and output from the node.
Figure 2(a) shows the settings and input/output signals of each device when operating at full-light wavelength conversion. The optical signal input to the node is an M-PSK or PAM-M signal. Furthermore, the electrical signal in the node is continuous wave (CW). The 2×2 optical switch is in the bar state. The transmission wavelength of WSS is set to the newly generated λ3. The optical signal output from the node is the same as the input. Figure 2(b) shows the settings and input/output signals of each device when operating as a PAM-M to M-PSK conversion [24,25]. The optical signal input to the node is a PAM-M signal. Furthermore, the electrical signal in the node is CW. The 2×2 optical switch is in the cross-state. The transmission wavelength of WSS is the wavelength λ2 used for CW. The optical signal output from the node is an M-PSK signal. Figure 2(c) shows the settings and input/output signals of each device when implementing all-optical data/encryption key superposition. The signal input to the node is M-PSK. The electrical signal in the node is a PAM-M signal. The 2×2 optical switch is in the bar state. The transmission wavelength of WSS is the wavelength λ1 used for M’-PSK. The optical signal output from the node is an M + M’-PSK signal. Here, M and M’ are powers of 2.
Fig. 2. Settings of each device and input/output signals: (a) wavelength conversion case, (b) modulation format conversion case, and (c) data/key superposition case.
3. Principle of optical encryption key superposition
Figure 3(a) and 3(b) illustrate the operation principle of the optical encryption key superposition for QPSK. At the optical encoder in the repeater node, the phase shift amount is “π” in the case of the conventional method. The encryption key data is “1” and “0”, and the phase rotation of the symbol is “π/2” and “0”, respectively, in the case of the proposed method. In the digital decoder of the receiver, when the decoding key data is “1” and “0”, the phase shift amount is “π” and “0” in the conventional method, whereas it is “-π/2” and “0” in the proposed method. Ideally, the shape of the signal constellation does not change at any point.
Fig. 3. Operation principle of optical encoding and digital decoding: (a) conventional scheme, (b) proposed scheme.
Figure 4 shows a principle of optical encryption key superposition for QPSK system with the incorporation of a small amount of XPM effect. The serial data stream is split every two bits on the transmitter side, and the symbol mapper maps one symbol. In the AOSP-based analog encoder with an optical encryption key superimposed on an optical repeater, the HNLF-based on XPM generates a π/2 phase shift using EDFA according to the intensity of the PAM2 signal of the encryption key. A random binary data with a key length of 256 bits, which is the same as the high encryption standard, is used for the encryption keystream.
The amount of phase change of the probe light of each intensity of the PAM signal by XPM can be described as
where $\gamma$ and ${L_{eff}}$ are the nonlinear coefficient and the effective interaction length of the HNLF, respectively, with ${P_n}$ being the power of the n-th level of the PAM2 signal (n = 0,1).On the receiver side, the received symbol is decoded by a digital decoder that performs the opposite phase shift to the AOSP-based analog encoder. The timing of giving the received symbol a phase rotation based on the encryption key must be shared in advance by both the transmitter and receiver using the pilot signal. The output symbols from the digital decoder are converted to two-bit blocks and parallel-to-serial conversion to regenerate the original bit sequence. When the correct key decodes an AOSP-encrypted QPSK signal, it is close to the error rate characteristics of a standard QPSK signal. Furthermore, when the encrypted QPSK is decoded with the wrong key or not, the bit error rate (BER) characteristic is about 0.5, regardless of the amount of noise.
4. Experimental setup and results
Figure 5 shows an experimental setup of a coherent 10 GSymbol/s binary phase shift keying (BPSK)/QPSK system based on optical encryption key superposition. We experimentally compared the performances of standard BPSK/QPSK and BPSK/QPSK with superimposed optical encryption keys. We emulated an analog encoder and digital decoder using HNLF and a digital storage oscilloscope (DSO) to demonstrate the principles of optical encryption and digital decryption. The arbitrary waveform generator (AWG) and DSO sampling rates are 10 GSa/s and 20 GSa/s, respectively. Aliasing does not occur because the AWG band is limited to 7.5 GHz.
At the transmitter, a serial binary-data was divided into two bits and mapped to one of the symbols at the symbol mapper in an AWG. Each symbol phase was modulated by the IQM. A single tunable laser diode (LD) produced a continuous wave (CW) with a center wavelength of 1550 nm and a linewidth of 100 kHz. The CW output was launched to IQM and modulated into a BPSK/QPSK signal.
At the RAOSP node, an encrypted keystream signal consisting of serial binary data was modulated by an intensity modulator (IM). A single LD produced CW at the center wavelength of 1545 nm. The CW output is modulated by an IM to produce a PAM2 signal, as shown in Fig. 6. To monitor the amount of self-phase modulation (SPM) and stimulated Brillouin scattering (SBS) generated by PAM2 signal, the optical circulator, optical coupler, and optical spectrum analyzer (OSA) were placed before and after the HNLF. The XPM phase change for each symbol was generated by the encryption encoder as “π/2” or “π” when the encryption key data was “1” in HNLF. Moreover, when the encryption key data is “0” in HNLF, the symbol phase does not change due to XPM. The physical parameters of HNLF are summarized in Table 1. The encryption keystream uses a 256-bit pseudo-random bit sequence consisting of 28 binary data. An optical bandpass filter (OBPF) was placed after the HNLF to extract only the BPSK/QPSK signal with superimposed optical encryption keys. Figure 7 (a) and 7 (b) show the optical spectra of the BPSK/QPSK signal without and with superimposed optical encryption keys, respectively.
Table 1. HNLF Parameters
At the receiver, the optical encrypted BPSK/QPSK signal was input to the coherent receiver after adjusting the received power with a variable optical attenuator (VOA). We set the input power of the local oscillator (LO) to the coherent receiver to 10 dBm, which is required for balanced photodiodes. In the coherent receiver, the CW light that is polarized and input from the light source and the optical encrypted QPSK signal is input and divided into the I component and Q component of the optical domain. We implemented an offline DSP based on the output signal of DSO. The DSP consists of down-sampling (DS), frequency offset compensation (FOC), carrier-phase recovery (CPR), digital decoder, de-mapper, and BER measurement. These experiments assume that the recipient knows the timing of the encryption key in advance.
Figure 8 shows the experimental BER characteristics of the received signal for the received power. Here we compare the performance of BPSK and QPSK signals with and without optical encryption. Solid lines show the theoretical values of BPSK and QPSK calculated by numerical calculation. Figure 9(a) and 9(b) show the constellation of BPSK signal without and with superimposed optical encryption keys for the received power of -37 dBm. The constellation of QPSK signal without and with superimposed optical encryption keys for the received power of -37 dBm is shown in Fig. 10(a) and 10(b). Because it has converged precisely to four points, it can be confirmed that there is no problem with superimposed optical encryption keys of the optical QPSK signal itself. When the SPM-effected PAM2 signal is reflected by the SBS, a secondary XPM is generated between the reflected signal and the primary BPSK and QPSK signal. The power penalty within 1 dB from the theoretical characteristic at the time of high received power of the standard QPSK characteristic occurs due to the slight IQ imbalance seen in Fig. 10 (a). The power penalty is calculated based on the power difference at the FEC limit (BER = 3.8 × 10−3).
Figure 11(a) and 11(b) show the reflection and transmission spectra of the PAM2 signal. In the case of 36 mW, we can confirm the wavelength component of SBS at 0.08 nm higher than the primary peak signal. When the HNLF input power increases to 90 mW, the effects of SBS and SPM appear on the spectrum. Due to the slight SBS, the power penalties without and with superimposed optical encryption keys occurred at 1.5 dB and 3.5 dB in the case of BPSK and QPSK, respectively. For the constellation, BPSK with a large amount of phase change due to XPM, seems to deteriorate more than QPSK, but BER has less deterioration than QPSK because of the Euclidean distance of the signal point.
Fig. 11. Reflection and transmission spectra of PAM2 signal (a) 36 mW, (b) 90 mW HNLF input power case.
When the optical encrypted BPSK/QPSK signal was decoded with the wrong key or was not decoded, the BER characteristic was 0.5 regardless of the received power. Figure 12 shows the HNLF input power tolerance of the PAM2 signal for the BPSK and QPSK signals. The optimum HNLF input power was 35mW and 56mW for π/2 and π shift, respectively. By changing from the π shift to the π/2 shift method, the input power of the PAM2 signal equivalent to the encryption key to the HNLF can be saved by 21 mW.
5. Conclusions
We propose an RAOSP node architecture for low-latency application at both HNGNs and optical relay nodes in metro networks. A 10-GSymbol/s coherent BPSK/QPSK system with superimposed optical encryption keys at the repeater is successfully demonstrated. The required HNLF input power is moderated to 35 mW through the π/2 phase shift due to XPM, and the power penalty compared with standard BPSK and QPSK signal saves 1.5 dB and 3.5 dB at the FEC limit due to the SPM and SBS in BPSK and QPSK, respectively. To the best of our knowledge, there is no other report that applies optical encryption-key superposition techniques in the RAOSP node architecture. However, the proposed method is challenging to apply to systems with strict penalties because of the deterioration of characteristics. It is necessary to consider further suppressing the occurrence of induced Brillouin scattering in the future. Moreover, we plan to apply the encryption-key superposition technique to the high-order quadrature amplitude modulation, which enables large-capacity transmission by allocating information to both amplitude and phase to apply it to existing systems.
Funding
Japan Society for the Promotion of Science (JP22K04105).
Acknowledgments
We thank T. Miyazaki of the University of Yamanashi for their support in the experiments. We would like to thank Editage (www.editage.com) for English language editing.
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. A. E. Willner, S. Khaleghi, M. R. Chitgarha, and O. F. Yimaz, “All-optical signal processing,” J. Lightwave Technol. 32(4), 660–680 (2014). [CrossRef]
2. S. Webnitz and B. J. Eggleton, All-optical signal processing, Springer: Berlin, Germany, 2015.
3. M. Saruwatari, “All-optical signal processing for terabit/second optical transmission,” IEEE J. Sel. Top. Quantum Electron. 6(6), 1363–1374 (2000). [CrossRef]
4. S. Singh and Lovkesh, “Ultrahigh speed optical signal processing logic based on an SOA-MZI,” IEEE J. Sel. Top. Quantum Electron. 18(2), 970–977 (2012). [CrossRef]
5. D. Wang, M. Zhang, G. W. Lu, L. Qin, T. Sakamoto, K. Akahane, N. Yamamoto, T. Kawanishi, Z. Li, and Y. Cui, “Multifunctional all-optical signal processing scheme for simultaneous multichannel WDM multicast and XOR logoc gates based on FWM in QD-SOA,” Proc. Optical Fiber Communication Conference (OFC), Los Angeles, USA, Mar. 2015, Th2A.5.
6. D. J. Blumenthal, R. Heideman, D. Geuzebroek, A. Leinse, and C. Roeloffzen, “Silicon Nitride in silicon photonics,” Proc. IEEE 106(12), 2209–2231 (2018). [CrossRef]
7. J. Wang and Y. Long, “On-chip silicon photonic signaling and processing: a review,” Sci. Bull. 63(19), 1267–1310 (2018). [CrossRef]
8. S. Radic, “Parametric signal processing,” IEEE J. Sel. Top. Quantum Electron. 18(2), 670–680 (2012). [CrossRef]
9. K. Igarashi and K. Kikuchi, “Optical signal processing by phase modulation and subsequent spectral filtering aiming at applications to ultrafast optical communication system,” IEEE J. Sel. Top. Quantum Electron. 14(3), 551–565 (2008). [CrossRef]
10. T. Kato, S. Watanabe, T. Yamauchi, G. Nakagawa, H. Muranaka, Y. Tanaka, Y. Akiyama, and T. Hoshida, “Whole band wavelength conversion for wideband transmission,” Proc. Optical Fiber Communication Conference (OFC), San Diego, USA, Mar. 2021, F1B.1.
11. T. Kobayashi, S. Shimizu, M. Nakamura, T. Umeki, T. Kazama, R. Kasahara, F. Hamaoka, M. Nagatani, H. Yamazaki, H. Nosaka, and Y. Miyamoto, “Wide-band inline-amplified WDM transmission using PPLN-based optical parametric amplifier,” J. Lightwave Technol. 39(3), 787–794 (2021). [CrossRef]
12. H. Kawahara, M. Nakagawa, T. Seki, and T. Miyamura, “Experimental demonstration of wavelength-selective band/direction-switchable multi-band OXC using an inter-band all-optical wavelength converter,” Proc. European Conference on Optical Communication (ECOC), Sept. 2021.
13. Z. Tong, C. Lundstrom, P. A. Andrekson, C. J. Mckinstrie, M. Karlsson, D. J. Blessing, E. Tipsuwannakul, B. J. Puttnam, H. Toda, and L. G-Nielsen, “Towards ultrasensitive optical links enabled by low-noise phase sensitive amplifiers,” Nat. Photonics 5(7), 430–436 (2011). [CrossRef]
14. M. Rochette, L. Fu, V. Ta’eed, D. J. Moss, and B. J. Eggleton, “2R optical regeneration: an all-optical solution for BER improvement,” IEEE J. Sel. Top. Quantum Electron. 12(4), 736–744 (2006). [CrossRef]
15. V. Bobrovs, S. Spolitis, and G. Ivanovs, “Latency causes and reduction in optical metro networks,” Proc. SPIE 9008, 90080C (2014). [CrossRef]
16. G. Brown, “New transport network architecture for 5G RAN,” Fujitsu, Kanagawa, Japan, White Paper, 2018. [Online]. Available: https://networking.report/Resources/Whitepapers/34e518ed-c067-4579-adf1-38441f13bc66_New-Transport-Network-Architectures-for-5G-RAN.pdf
17. 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]
18. K. Mishina, S. Kitagawa, and A. Maruta, “All-optical modulation format conversion from on-off-keying to multiple-level phase-shift-keying based on nonlinearity in optical fiber,” Opt. Express 15(13), 8444–8453 (2007). [CrossRef]
19. G. W. Lu, E. Tipsuwannakul, T. Miyazaki, C. Lundstrom, M. Karlsson, and P. A. Andrekson, “Format conversion of optical multilevel signals using FWM-based optical phase erasure,” J. Lightwave Technol. 29(16), 2460–2466 (2011). [CrossRef]
20. W. Astar and G. M. Carter, “Mitigation of XPM penalty in 10-Gb/s OOK/DBPSK mixed modulation format 50-GHz-spaced WDM transmission by conversion of OOK to BPSK using an SOA,” IEEE Photonics Technol. Lett. 20(20), 1715–1717 (2008). [CrossRef]
21. T. Kodama and T. Miyazaki, “CW-probe-less OOK and BPSK to QPSK optical modulation format conversion and SSMF transmission,” Proc. Conference on Lasers and Electro-Optics (CLEO), San Jose, USA, May 2019, JTh2A.79.
22. Q. Li, X. Yang, and J. Yang, “All-optical aggregation and de-aggregation between 8QAM and BPSK signal based on nonlinear effects in HNLF,” J. Lightwave Technol. 39(17), 5432–5438 (2021). [CrossRef]
23. T. Kodama, “Photonic symbol masking using optical encryption key generator for secure optical regenerative repeaters,” Proc. Signal Processing in Photonic Communications (SPPCom), SpTu3I.5, Online, July 2020. [CrossRef]
24. T. Kodama, T. Miyazaki, and M. Hanawa, “Seamless PAM-4 to QPSK modulation format conversion at gateway for short-reach and long-haul integrated network,” Proc. The 39th European Conference on Optical Communications (ECOC), We3H.1, Rome, Italy, Sept. 2018.
25. T. Kodama and T. Miyazaki, “Scalable PAM8 to 8PSK optical modulation format conversion based on XPM in HNLF at gateway for short-reach and long-haul integrated network,” Proc. Asia Communications and Photonics Conference (ACP), S3A.5, Chengdu, China, Nov. 2019.
