Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Application of a novel SSFBG-based single-input multiple-output encoder/decoder to optical packet switching

Open Access Open Access

Abstract

We propose a novel superstructured fiber Bragg grating (SSFBG) based single-input multiple-output (SIMO) en/decoder, which can simultaneously process a group of independent optical codes with a specific permutation and combination of code patterns and spatially separate encoded and decoded signals into different optical paths. The number of optical codes processed by the SIMO en/decoder can be flexibly changed. We theoretically and experimentally investigate the coding performance of the SIMO en/decoder and discuss the unique features of the proposed device. In the experiment, we employ the SIMO en/decoder for optical label processing in an optical-code based optical packet switching system with the data rate of 10 Gbit/s and the packet rate of 312.5 MHz. The four-SSFBG based SIMO en/decoder are capable of simultaneously generating/recognizing four 31-chip 640 Gchip/s optical code based labels and distribute the labels into four designated destinations.

©2014 Optical Society of America

1. Introduction

Coherent optical code processing techniques are regarded as promising techniques to manipulate broadband optical signals at an ultra-high speed. The techniques are widely used in the many aspects, including optical code-division multiple access (OCDMA) systems, optical secure communication systems and optical packet switching (OPS) networks [16]. In the OPS networks, an optical code can be used as a unique address for labeling an optical packet because an encoded optical signal has unique physical characteristics and can be recognized only by a corresponding decoder. Since a single-bit encoded signal contains a large amount of code information, the optical-code based label can be very short and the length of packet header can be significantly shortened, which is desirable for nowaday busy networks. Some coherent-optical-code-based optical packet routers have been demonstrated in the previous work. Photonic routers using optical-code-based label switch were demonstrated to switch packets with a large throughput of 1.28 Tbit/s [5]. An optical-code-based multi-wavelength optical packet router accomplished optical packet switching in the 40 wavelength channels [6].

A variety of coherent optical code processing techniques were realized by means of different devices including arrayed waveguide grating (AWG), fiber Bragg grating (FBG), micro-ring reflector (MRR) and planar lightwave circuit (PLC) [714]. Among various en/decoding devices, AWG-based multiport en/decoder can simultaneously process a group of optical codes. However, the optical codes generated by the AWG-based multiport encoder do not have independent physical characteristics and the autocorrelation function of the codes is triangular shape, which results in severe beat noise [7]. In order to reduce the beat noise and enhance the cascadability of the en/decoders, a modified configuration was theoretically proposed by inserting a splitter and a set of phase shifters [8]. The en/decoding devices become more complex and practically the precise control of each phase shifter is very complicated. By contrast with the coding properties of AWG-based en/decoders, superstructured FBG (SSFBG) en/decoders are able to generate optical codes with independent characteristics. The autocorrelation function of the codes is well-defined needle-shape and the ratio between auto-correlation and cross-correlation functions is very high [10]. Besides, SSFBG en/decoders are able to process ultra-long optical codes (1023-chip) at an ultra-high chip rate (640 Gchip/s) with a very low cost.

In this paper, we propose a novel SSFBG-based single-input multiple-output (SIMO) en/decoder which is able to simultaneously process a group of independent optical codes with desired coding performance. We also apply the SIMO en/decoder in the OPS network for optical label processing. The paper is organized as follows. In Section 2, we introduce the structure of the SIMO en/decoder and analyze the coding performances. We also discuss the unique features of the SIMO en/decoder. In Section 3, we present an optical-code based OPS network and demonstrate an OPS network using the SIMO en/decoder for optical label processing. We summarize the paper in Section 4.

2. SSFBG-based SIMO en/decoder

2.1 Structure of the SIMO en/decoder

Figure 1 shows the block diagrams and schematic diagrams of the SIMO en/decoder. The SIMO en/decoder consists of an optical circulator array, a group of SSFBG en/decoders and a set of optical tunable delay lines (OTDL) and variable optical attenuators (VOA). All the circulators have three ports and are placed in order. Port 1 of the first circulator is used as the input port. All Port 3′s of the circulators are connected to the OTDLs and VOAs for the outputs. The SSFBG en/decoders and the circulators are connected serially. One side of the SSFBG is connected to Port 2 of the circulator and the other side is connected to Port 1 of the next-stage circulator. Since the SSFBGs are designed with relatively low reflectivity to avoid multi-reflection of the light inside the gratings, the transmission of the SSFBGs en/decoders have very low loss and slight distortion. Figure 2 shows the calculated results of the comparison of an input Gaussian-shaped 2 ps (FWHM) optical pulse and a transmitted output after 8 SSFBGs. The shape of the optical pulse remains the same. Almost no difference can be observed between the input and output pulses. Thus, a signal passing through an SSFBG en/decoder experiences negligible distortion and can be re-used in the next-stage en/decoding.

 figure: Fig. 1

Fig. 1 (a) Block diagrams of the SIMO en/decoder. (b) Schematic diagrams of the SIMO en/decoder. E and D stands for encoder and decoder. The first inferior, 1, 2 … M, is marshalling sequence. The second inferior, C1, C2 … CN, is code number.

Download Full Size | PDF

 figure: Fig. 2

Fig. 2 Comparison of the input optical pulse and the transmitted pulse after 8 SSFBGs.

Download Full Size | PDF

In the encoding, an optical pulse is input into the input port of the SIMO encoder. The pulse passes through circulators and SSFBGs sequentially. The encoded signals are reflected from the SSFBGs into Port 2’s of the circulators and output from Port 3′s of the circulators. The TODLs and VOAs are used to adjust the temporal delay and balance the power. Consequently, the encoded signals are simultaneously generated from the output ports of the SIMO encoder.

In the decoding, an encoded signal is input into the SIMO decoder and passes through all the circulators and SSFBGs. The SSFBGs used in the SIMO decoder have spatially reversed structures of those used in the SIMO encoder to form pairs of matched en/decoders, which can be easily realized by spatially reversely writing the gratings using the same mask. Only when the encoded signal passes through a matched decoder, a needle-shape optical pulse is generated. Otherwise, very low noise-like signals are generated. The decoded signals are output simultaneously from the SIMO decoder and the needle-shape decoded signal can be obtained from the corresponding output port.

2.2 Coding performance of the SIMO en/decoder

First of all, we mathematically investigated the coding performance of the SIMO en/decoder. In the calculation, we designed 16 pairs of SSFBGs with 31-chip Gold codes and 16 pairs of SSFBGs with 63-chip Gold codes. We calculated autocorrelation and cross-correlation functions of any two pairs of SSFBGs when they were placed in the different stages of the SIMO en/decoder. To figure out the potential performance of the SIMO en/decoder, we assumed there were 16 stages in total. Figure 3 shows the normalized peak power of the decoded signals and (P/C), which is a key parameter to evaluate the coding performance, with the change of the position of the SSFBG decoder. When the SSFBG is moved from the first stage (conventional back-to-back en/decoding) to the sixteenth stage in the SIMO decoder, the coding performance (P/C) slightly degrades. The average degradation of 8 stages is only 5.3% for 31-chip SSFBG and 7.1% for 63-chip SSFBG, while the average degradation of 16 stages is 17.6% and 19.4% for 31-chip and 63-chip SSFBGs respectively. Although the coding performance degrades, there is no difficulty to distinguish the autocorrelation signal from the cross-correlation noises. The waveforms of the decoded signals are shown in Fig. 4. Needle-shape optical pulses are obtained when the SSFBG decoder matches the SSFBG encoder wherever the decoder is placed and low-intensity noises are generated when the decoder does not match the encoder. To achieve a balanced coding performance, the inverse sequence of the SSFBG decoders is preferred for first-encoding-last-decoding, as shown in Fig. 1.

 figure: Fig. 3

Fig. 3 Normalized power of the decoded signals and the ratio of the autocorrelation and cross-correlation functions with the change of the positions of the SSFBG in the SIMO decoder. SSFBGs with (a) 31-chip Gold codes and (b) 63-chip Gold codes

Download Full Size | PDF

 figure: Fig. 4

Fig. 4 Waveforms of the decoded signals when the SSFBG are placed in the different stages of the SIMO decoder. (a) Autocorrelation signals when the decoders match the encoders. (b) Cross-correlation signals when the decoders do not match the encoders.

Download Full Size | PDF

We carried out experiments to verify the feasibility of the proposed SIMO en/decoder. In the experiment, we used four pairs of 31-chip 640 Gchip/s SSFBGs in the SIMO en/decoders. Before investigating the coding performance of the SIMO en/decoder, we analyzed the transmission properties of the SSFBGs. The transmission and reflection spectra of two different pairs of SSFBGs are shown in Fig. 5. Each pair of SSFBGs in the SIMO en/decoder has same spectra. The average transmission loss of the SSFBGs is only –1 dB. The average chromatic dispersion of the four serially connected SSFBGs is 0.02 ps/nm.

 figure: Fig. 5

Fig. 5 (a) and (b) Transmission and reflection spectra of two SSFBGs. Green solid line: measured transmission spectrum. Blue solid line: measured reflection spectrum. Red dashed line: calculated spectra.

Download Full Size | PDF

We installed the four pairs of SSFBGs in first four stages of the SIMO en/decoder and measured encoded and decoded waveforms. A sequence of 10 GHz optical pulses with pulse width of 2 ps (FWHM) was generated by a mode-locked laser diode (MLLD) and injected into the SIMO encoder. Four encoded signals were obtained simultaneously from the first four output ports of the SIMO encoder, as shown in Fig. 6(a). The encoded waveforms are noise-like signals. Then the each encoded signal was input into the SIMO decoder separately. The decoded signals were output simultaneously from the first four output ports of the SIMO decoder. The sixteen measured and calculated decoded signals are illustrated in Fig. 6(b). Optical pulses with high-intensity peaks were obtained only from the output ports where SSFBGs match. From other output ports, we only obtained very low-intensity noise-like signals. The good coding performance indicates that the SIMO en/decoder is feasible to generate and recognize a group of independent optical codes simultaneously.

 figure: Fig. 6

Fig. 6 (a) Encoded and (b) decoded waveforms. Blue lines: Calculated results.

Download Full Size | PDF

2.3 Power budget of the SIMO en/decoder

The SSFBGs used in the SIMO en/decoder are designed with low reflectivity to avoid multi-reflection of the light inside the gratings and guarantee the coding performance. The low reflection of the SSFBG leads to a loss (LR) of the SIMO en/decoder. Besides, the insertion loss of the optical circulators (LC) and transmission loss of the SSFBGs (LT) also contribute to the insertion loss of the SIMO en/decoder. For an output on the Mth stage, the total loss is LR + MLC + (M–1)LT. To compensate the insertion loss, a booster amplifier with gain of GB before the SIMO en/decoder and several in-line amplifiers with gain of GI within the SIMO en/decoder can be used. The amplification of the booster amplifier is to compensate the loss of the reflection of the SSFBG, i.e. GB≥LR. The in-line amplifiers can be inserted in the SIMO en/decoder every N stages, where N = (GI + LT)/(LT + LC), to compensate the insertion loss of the optical circulators and the transmission loss of the SSFBGs.

2.4 Features of the SIMO en/decoder

The SIMO en/decoder are able to simultaneously process a group of optical codes and spatially separate encoded and decoded signals into different optical paths. Since every SSFBG in the SIMO en/decoder is individually selected, the SIMO en/decoder can independently process the optical codes. The code processing of the SIMO en/decoder is much flexible by organizing the permutation and combination of SSFBGs and arranging the type of SSFBGs. The number of SSFBG can be flexibly changed. The serial-connection structure is simple and it makes the SIMO device have the potential to realize in-line signal processing, such as in-line signal amplification and in-line signal regeneration. Besides, the SSFBG based SIMO en/decoder have the advantages of high compactness, polarization insensitivity and low cost.

3. The SIMO en/decoder in the OPS networks

Since the SIMO en/decoder can process multiple optical codes at the same time and the optical codes can be used as optical labels in the OPS networks, the SIMO en/decoder are suitable devices in the OPS networks for optical label processing.

3.1 Architecture of an optical-code based OPS network

Figure 7 depicts the architecture of an optical-code based OPS network. In the network, many routers switch optical packets to different destinations. An optical packet consists of an optical label and an optical payload. Optical codes are used as the optical labels. In this network there are three kinds of routers. Ingress routers generate and assign an optical label to the optical packet entering the core network and pass the optical packet to the core network by a certain route. Routers inside the core network separate the optical label and the optical payload. Based on the information of the label, the routers generate and assign a new label to the payload to form an optical packet. Then, the optical packet is switched to another router by a designated route. When an optical packet leaves the core network, it passes through an egress router. The egress router separates the label and payload and selects a certain route to transmit the payload.

 figure: Fig. 7

Fig. 7 Architecture of optical-code based OPS network and the functions of different routers.

Download Full Size | PDF

3.2 Demonstration of the OPS system

Figure 8 shows the experimental setup of the OPS system. We demonstrated the key functions of the routers including payload generation, label generation, label writing, label and payload separation, label recognition and packet selection and switching.

 figure: Fig. 8

Fig. 8 Experimental setup of the OPS system.

Download Full Size | PDF

In this experiment, four payloads are generated by intensity modulation of a continuous-wave (CW) light with a 10 Gbit/s NRZ signal. The payload duration is 3 ns and there is a time slot of 200 ps between the two adjacent payloads for inserting optical labels. The data rate and the length of payloads can be much higher and longer and it does not affect the label processing. The waveform of Payload 3 is shown in Fig. 9(a). To generate optical labels, including headers and trailers, a 10 Gbit/s optical pulse train is modulated by an intensity modulator with a specified data pattern and input into the SIMO encoder. The four optical labels are generated simultaneously from the SIMO encoder and inserted into the time slots of the payloads, as shown in Fig. 9(b). The optical labels only have two data bits, but still contain sufficient routing information. The packet rate is 312.5 MHz. The spectrum of the optical packet is shown in Fig. 10(a). The spectra of the payloads and labels are overlapped together to efficiently use the optical spectrum.

 figure: Fig. 9

Fig. 9 Waveforms of (a) the payloads, (b) insertion of the optical labels into the payloads, (c) the separated optical labels, (d) the generated electrical time gate and the separated payloads, (e) the recognized optical labels, and (f) the selected payload.

Download Full Size | PDF

 figure: Fig. 10

Fig. 10 (a) The spectrum of the optical packet and (b) the spectra of the separated payloads and labels after the notch filter.

Download Full Size | PDF

To separate the labels and the payloads, a spectral notch filter with 3 dB bandwidth of 0.6 nm is used. The waveforms and spectra of the separated payloads and labels are illustrated in Figs. 9(c), 9(d) and 10(b). The separated optical labels are input into the SIMO decoder for label recognition. The recognized labels are output simultaneously from the four output ports of the SIMO decoder. In the each output of the SIMO decoder, only the target label is recovered into a high-intensity optical pulse, as shown in Fig. 9(e). Then, the recognized labels are converted into electrical signals to generate a time gate by using a T flip-flop. The generated time gate is used to drive an intensity modulator to select the target payload from the separated payloads. The selected payload is shown in Fig. 9(f). The payload selection can also be realized all-optically by using optical time gating techniques so as to achieve all-optical packet switching.

The BER performance of the optical packet switching system is illustrated in Fig. 11. All selected payloads are error-free. Compared with the original generated payload, the selected payload has only about 2 dB degradation.

 figure: Fig. 11

Fig. 11 The BER performance of the OPS system.

Download Full Size | PDF

4. Conclusion

We have proposed an SSFBG based SIMO en/decoder and demonstrated en/decoding functions of the device. The SIMO en/decoder have simple structures and are able to simultaneously process a group of independent optical codes with a guaranteed coding performance and spatially separate the codes into the designated optical paths. The ratios of the autocorrelation and cross-correlation functions are larger than 11 and 15 for a 16-stage 31-chip-SSFBG based SIMO en/decoder and a 16-stage 63-chip-SSFBG SIMO en/decoder, respectively. It is easy to distinguish the autocorrelation signal from low-intensity cross-correlation noises. By organizing the permutation and combination of SSFBGs, it is possible to generate optical codes with a specific code pattern, which makes the code processing very flexible. Furthermore, the number of optical codes processed by the SIMO en/decoder can be flexibly changed.

In the experiment, we have applied the four-SSFBG based SIMO en/decoder to the optical-code based OPS system with the packet rate of 312.5 MHz and the data rate of 10 Gbit/s, and demonstrated pivotal functions of routers including payload generation, label generation, label writing, label and payload separation, label recognition and packet selection and switching. The SIMO en/decoder generates and recognizes four 31-chip 640 Gchip/s optical code based labels at the same time and distributes the labels into four different optical paths. The labels containing address information only have two bits, significantly reducing the redundancy of packets. The compactness of the SIMO en/decoder simplifies the OPS system.

The unique optical code processing capabilities make the SIMO en/decoder a promising device for the applications in the fields of optical secure communication, optical packet switching and optical signal processing.

Acknowledgments

This work was partially supported by National Natural Science Foundation of China (61378060); Dawn Program of Shanghai Education Commission (11SG44); the Research Fund for the Doctoral Program of Higher Education of China (20123120130001). The authors would like to thank Mr. H. Sumimoto of NICT for his technical support.

References and links

1. S. J. B. Yoo, J. P. Heritage, V. J. Hernandez, R. P. Scott, W. Cong, N. K. Fontaine, R. G. Broeke, J. Cao, S.-W. Seo, J.-H. Baek, F. M. Soares, Y. Du, C. Yang, W. Jiang, K. Aihara, Z. Ding, B. H. Kolner, S. Anh-Vu Pham, S. Lin, F. Olsson, S. Lourdudoss, K. Y. Liou, S. N. Chu, R. A. Hamm, B. Patel, W. S. Hobson, J. R. Lothian, S. Vatanapradit, L. A. Gruezke, W. T. Tsang, M. Shearn, and A. Scherer, “Spectral phase encoded time spread optical code division multiple access technology for next generation communication networks,” J. Opt. Commun. Networking 6, 1210–1227 (2007).

2. J. A. Salehi, “Emerging OCDMA communication systems and data networks,” J. Opt. Networking 6(9), 1138–1178 (2007). [CrossRef]  

3. Z. Gao, B. Dai, X. Wang, N. Kataoka, and N. Wada, “40 Gb/s, secure optical communication based upon fast reconfigurable time domain spectral phase en/decoding with 40 Gchip/s optical code and symbol overlapping,” Opt. Lett. 36, 4326–4328 (2011). [CrossRef]   [PubMed]  

4. X. Wang and N. Wada, “Experimental demonstration of OCDMA traffic over optical packet switching network with hybrid PLC and SSFBG en/decoders,” J. Lightwave Technol. 24(8), 3012–3020 (2006). [CrossRef]  

5. N. Wada, W. Chujo, and K. Kitayama, “1.28 Tbit/s (160 Gbit/s x 8 wavelengths) throughput variable length packet switching using optical code based label switch,” in Proc. 27th European Conf. on Opt. Commun., Amsterdam, The Netherlands, PD-A-1–9 (2001).

6. P. C. Teh, B. C. Thomsen, M. Ibsen, and D. J. Richardson, “Multi-wavelength (40 WDMx10 Gbit/s) optical packet router based on superstructure fibre Bragg gratings,” Trans. Comm. E 86B, 1487–1492 (2003).

7. G. Cincotti, N. Wada, and K. Kitayama, “Characterization of a full encoder/decoder in the AWG configuration for code-based photonic routers-part I: modeling and design,” J. Lightwave Technol. 24(1), 103–112 (2006). [CrossRef]  

8. G. Cincotti, G. Manzacca, X. Wang, T. Miyazaki, N. Wada, and K. Kitayama, “Reconfigurable multi-port optical encoder/decoder with enhanced auto-correlation,” IEEE Photonics Technol. Lett. 20(2), 168–170 (2008). [CrossRef]  

9. P. C. Teh, P. Petropoulos, M. Ibsen, and D. J. Richardson, “A comparative study of the performance of seven-and 63-chip optical code-division multiple-access encoders and decoders based on superstructured fiber Bragg gratings,” J. Lightwave Technol. 19(9), 1352–1365 (2001). [CrossRef]  

10. X. Wang, K. Matsushima, A. Nishiki, N. Wada, and K. Kitayama, “High reflectivity superstructured FBG for coherent optical code generation and recognition,” Opt. Express 12(22), 5457–5468 (2004). [CrossRef]   [PubMed]  

11. Y. Dai, X. Chen, Y. Zhang, J. Sun, and S. Xie, “Phase-error-free, 1023-chip OCDMA En/de-coders Based on Reconstruction equivalent chirp Technology and Error-correction Method,” in Proc. Opt. Fiber Commun. Conf. (OFC’ 06), Anaheim, USA (2006), paper JWA28. [CrossRef]  

12. B. Dai, Z. Gao, X. Wang, N. Kataoka, and N. Wada, “Performance comparison of 0/π- and ± π/2-phase-shifted superstructured Fiber Bragg grating en/decoder,” Opt. Express 19(13), 12248–12260 (2011). [CrossRef]   [PubMed]  

13. X. Wang and Z. Gao, “Novel reconfigurable 2-dimensional coherent optical en/decoder based on coupled micro-ring reflector,” IEEE Photonics Technol. Lett 23(9), 591–593 (2011). [CrossRef]  

14. K. Takiguchi, T. Shibata, and M. Itoh, “Encoder/decoder on planar lightwave circuit for time-spreading/wavelength-hopping optical CDMA,” Electron. Lett. 38(10), 469–470 (2002). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (11)

Fig. 1
Fig. 1 (a) Block diagrams of the SIMO en/decoder. (b) Schematic diagrams of the SIMO en/decoder. E and D stands for encoder and decoder. The first inferior, 1, 2 … M, is marshalling sequence. The second inferior, C1, C2 … CN, is code number.
Fig. 2
Fig. 2 Comparison of the input optical pulse and the transmitted pulse after 8 SSFBGs.
Fig. 3
Fig. 3 Normalized power of the decoded signals and the ratio of the autocorrelation and cross-correlation functions with the change of the positions of the SSFBG in the SIMO decoder. SSFBGs with (a) 31-chip Gold codes and (b) 63-chip Gold codes
Fig. 4
Fig. 4 Waveforms of the decoded signals when the SSFBG are placed in the different stages of the SIMO decoder. (a) Autocorrelation signals when the decoders match the encoders. (b) Cross-correlation signals when the decoders do not match the encoders.
Fig. 5
Fig. 5 (a) and (b) Transmission and reflection spectra of two SSFBGs. Green solid line: measured transmission spectrum. Blue solid line: measured reflection spectrum. Red dashed line: calculated spectra.
Fig. 6
Fig. 6 (a) Encoded and (b) decoded waveforms. Blue lines: Calculated results.
Fig. 7
Fig. 7 Architecture of optical-code based OPS network and the functions of different routers.
Fig. 8
Fig. 8 Experimental setup of the OPS system.
Fig. 9
Fig. 9 Waveforms of (a) the payloads, (b) insertion of the optical labels into the payloads, (c) the separated optical labels, (d) the generated electrical time gate and the separated payloads, (e) the recognized optical labels, and (f) the selected payload.
Fig. 10
Fig. 10 (a) The spectrum of the optical packet and (b) the spectra of the separated payloads and labels after the notch filter.
Fig. 11
Fig. 11 The BER performance of the OPS system.
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.