We introduce and experimentally demonstrate a flexible temporal illusion at telecommunication data rate in optical fiber communication system. The temporal illusion cannot only transform an event to another event as expected, but also mask the event with high-level signal, providing a novel method to conceal the confidential information. We successfully transform the output temporal waveforms of a return-to-zero (RZ), dark RZ and nonreturn-to-zero (NRZ) event into that of any above modulation format event and high-level signal at different illusion bits and mosaic bits at a data rate of 5 Gb/s, respectively. Our works offer us new perspectives on illusion optics for falsifying event rather than object, which has potential applications in secure communication, data encryption and other military applications.
© 2017 Optical Society of America
The invisibility cloak, as ultimate illusion, has recently attracted great interests in many fields due to its mysterious, fascinating and exciting characteristics . According to the space-time duality, the invisibility cloak can be classified into two types: spatial cloak and temporal cloak. The spatial cloak is used to hide the object by excluding electromagnetic waves in certain regions or by designing the complementary media within a certain distance outside the objects [2–12]. Based on the idea of complementary media, the concept of spatial cloak is extended to the spatial illusion aiming to transform the scattered signals of the remote object into that of the pre-designed targets [13–18]. As shown in Figs. 1(a) and 1(b), a general spatial illusion is generated such that a red apple appears to be a yellow pear in the observer view. Different from the global spatial cloak and illusion that make the whole object completely invisible and artificial, the localized spatial cloak (spatial mosaic) and illusion will hide or change the scattered signals of certain parts of the real object and retain the signals of remaining parts . Figures 1(c) and 1(d) show that a spatial mosaic device (solid, gray) can cause the observer outside the solid curve to see the image of a red apple with five white square boxes instead of a full red apple. The spatial cloaks and illusions can be realized by using transformation optics, conformal mapping and Fourier analysis [20–24]. Meanwhile, by extending the concept of the spatial cloak to the temporal domain, researches have recently described a temporal cloak with or without temporal gap which can hide some events without detection from the observer [25–32]. Many approaches have been proposed to accomplish the temporal cloak, such as using four-wave mixing, Talbot effect and Fourier analysis [33–38].
In principle, the concept of the spatial illusion and mosaic can also be extended to the temporal domain. However, temporal illusion and mosaic have not been reported yet to our knowledge. Obviously, the temporal illusion can be depicted as a function which transforms a real event to a pre-designed event. For example, as Figs. 1(e) and 1(f) show, when a temporal illusion is created, a military command, ‘Attack at 11:55 am’, will appears to be another totally different command, ‘Attack at 10:15 pm’. Similarly, a temporal mosaic can block some parts of a real event with high or low level signals and keep remaining parts of the real event unchanged. Figs. 1(g) and 1(h) show that a temporal mosaic device (solid, gray) causes the observer outside the solid curve to receive the command, ‘Attack at 1_:5_ am’, instead of ‘Attack at 11:55 am’. Therefore, temporal illusion and mosaic can be used to mislead the enemy, and conceal the real events.
In this paper, we introduce and experimentally demonstrate a temporal illusion and mosaic scheme for the first time to our knowledge. The temporal illusion can transform the output temporal waveforms of a return-to-zero (RZ), dark RZ and nonreturn-to-zero (NRZ) event into that of any other event at different illusion bits at a data rate of 5 Gb/s. The temporal mosaic can block some target bits with high-level signal. Our scheme can be widely applied in secure communication, data encryption and other military applications.
2. Principle and experiment
Figure 2(a) presents the schematic diagram of the temporal illusion device, with two modes: falsifying mode and mosaicking mode. A continuous-wave (CW) probe beam is injected into an optical switch (OS), and the probe beam is divided into two parts (the upper and lower beams) by the OS. For simplicity, we assume that the upper beam (real channel) is used to probe the real event from the first intensity modulator (IM1), and the illusion event is modulated by the second intensity modulator (IM2) in the lower beam (non-real channel). Later, the two beams are combined by an optical coupler (OC). Finally, a photodetector (PD) acts as an observer to perceive the output probe beam. It should be noted that the synchronization and incoherency of the two arm beams are required. The optical tunable delay line (OTDL) in the upper arm (or lower arm) is used to realize the synchronization of two arms, and the interference will not occur after the OC by adjusting the polarization and synchronization of the two arm beams.
The temporal illusion is summarized in Fig. 2(b). Without loss of generality, we set four pulses as real event (blue). The real event is loaded by IM1. When the OS is switched to the upper arm all the time, the temporal illusion will be turned off. In this case, the CW beam will be injected into real channel to probe the real event, and the output waveform with four pulses is detected by the PD. When the OS is driven by the control logic signal, the temporal illusion will be turned on. The light distribution (orange) of real and non-real channel is shown in Fig. 2(b). In the falsifying mode, five pulses (dash, pink) are recorded by the PD, including two pulses (dash, red) acting as illusion event and three bits from real event (dash, blue). Obviously, the output waveform is different from the real event. When we replace the illusion event with direct current (DC, solid, red), the temporal illusion could be switched to the mosaicking mode. On the mosaicking mode, three real pulses and five high-level signals (solid, pink) are detected by PD. Clearly, the recorded signal is also different from the real event. Some bits are occupied by the high-level so that one of real pulses is masked, while all other bits remain unchanged. It should be noted that the useless pulses are allowed, such as one pulse in the real event and one in the illusion event.
The experimental configuration for temporal illusion is presented in Fig. 2(c). This setup consists of four parts as follows: (1) an OS (orange dashed box), (2) a real signal emulator (blue dashed box), (3) a mosaic/illusion event emulator (red dashed box), and (4) an OC. In the first part, a CW probe light emitted by a tunable laser source (TLS), is injected into the first optical coupler (OC1). The light is divided into two parts equally, and they propagate through intensity modulators (IM3 and IM4) synchronously driven by a bit pattern generator (BPG) with control logic and inverse control logic, respectively. Then the probe beam will be switched into one of the two channels (real and non-real channel) by the pre-designed switch logic. In the second and third parts, IM1 and IM2 are used to act as the real signal emulator and the mosaic/illusion event emulator, respectively. Finally, the fourth part only contains an OC, which is used to combine the upper and lower arm beams into one before digital communication analyzer (DCA). The four test points (A, B, C, D) are used to measure the OS control signals (set by BPG) and the real event (all ‘1’ is modulated by IM3) and illusion event (all ‘1’ is modulated by IM4), respectively. In addition, all erbium-doped fiber amplifiers are used to compensate the system loss; the electrical amplifiers are used to provide the drive signal; the polarization controllers are used to adjust the polarization state and the OTDLs are used to provide the synchronization for the communication link. It should be noted that the equal optical intensity and incoherence of the upper and lower beams before the OC2 are necessary. The two arms must be incoherent due to two factors. One is that the OS may have a finite switch extinction ratio and introduce crosstalk in the two arms. Another factor is that the finite rise and fall time of the switch will cause waveform overlap when combining the optical signals of the two arms.
3. Results and discussion
3.1 Falsifying mode
Firstly, the central wavelength and the output optical power of a CW light emitted from a TLS is fixed at 1550 nm and 10 dBm, respectively. At the same time, IM3 is driven by a periodic logic signal ‘0011’ at 10 Gb/s while IM4 is driven by the inversed logic signal ‘1100’ provided by the BPG, and the output waveform (green) of IM3 is recorded, as shown in Fig. 3(a). Then the two adjacent illusion bits (green label) are created at a repetition rate of 2.5 GHz. Secondly, a periodic RZ signal with sequence of ‘1011’ is modulated by IM1 as the real event, and the waveform of the real event (blue) is shown in Fig. 3(d). Meanwhile, an illusion event (red), a periodic RZ signal with sequence of ‘0011’, is also loaded by IM2, shown in Fig. 3(g). Figure 3(j) shows that the output waveform is detected by DCA, and the output data (black) is the same as the illusion event and is different from the real event. The localized temporal illusion with falsifying mode succeeds in changing the real event into the illusion event with the same data pattern. Now the real event is fixed and the illusion event (see Figs. 3(m) and 3(s)) is replaced by the periodic ‘1111’ (dark RZ) and ‘1000’ (NRZ), respectively. We measure the output waveforms (black) which is different from not only the real event but also the illusion event, as shown in Figs. 3(p) and 3(v), because the output waveforms are hybrid of these two modulation formats. Moreover, we further tamper the real event with other modulation format (dark RZ and NRZ) using our temporal illusion. For simplicity, the OS logic signal is fixed (see Figs. 3(b) and 3(c)). The real event with dark RZ and NRZ is shown in Figs. 3(e) and 3(f), respectively. When the RZ event acts as the illusion event (see Figs. 3(h) and 3(i)), the output waveforms with two formats are also different form the real and illusion event, as shown in Figs. 3(k) and 3(l), respectively. Then we change the illusion event (see Figs. 3(n) and 3(o), dark RZ), and a high-level signal (see Fig. 3(q)) and a hybrid format signal (see Fig. 3(r)) are measured, and they are all different form the real and illusion event. Finally, we use two NRZ signals (see Figs. 3(t) and 3(u)) acting the illusion event, and the output waveforms are shown in Figs. 3(w) and 3(x), respectively. Particularly, Fig. 3(w) shows an output data that is the same as the illusion event, although the real and illusion events have different modulation formats. The ripple factor, defined as the ratio of the ripple power to the average power at high-level, is ~34% and the extinction ratio of the output waveform is ~8.33 dB, respectively.
To flexibly modify different illusion bits, we change the OS logic signal by resetting the driven signal of IM3 and IM4. The driven signal of IM3 is reset with periodic bit sequences of ‘0111’, ‘1010’, and ‘0100’, as shown in Figs. 4(a)-4(c), respectively. And the driven signal of IM4 are simultaneously changed into inversed data of IM3. Then the real event (RZ, ‘1000’, blue, see Figs. 4(d)-4(f)) and illusion event (RZ, ‘1011’, red, see Figs. 4(g)-4(i)) are provided by BPG. The output RZ signal (black) are observed by DCA as shown in Figs. 4(j)-4(l), respectively. Comparing the above three cases, one can see that the output waveforms are different, although real event and illusion are the same. It indicates that the OS logic signal also plays a decisive role in output result. Particularly, the output signal and the illusion event are the same, when the driven signal is set to ‘0100’. In addition, the output signal is the same as the real event which is different from the illusion event, while the driven signal is set to ‘0111’. Note that no matter what the output is, the illusion bits and remaining bits of output signal must be the same as the corresponding bits of illusion event and real event, respectively.
3.2 Mosaicking mode
Figure 5 shows the experiment results of temporal mosaic. A temporal mosaic is created by only changing the illusion event into DC event. First of all, we fix the output data of IM3 (‘0111’, periodic, green), as shown in Figs. 5(a) and 5(b), respectively. And the output signal of IM4 is simultaneously changed into inversed data of IM3. Secondly, the two periodic data sequences (‘0011’ and ‘1011’) with three modulation formats (RZ, dark RZ and NRZ) act as real event which is driven by IM1, while the DC event is driven by IM2. All the real events (blue) are shown in Fig. 5. Then, the mosaicked signals (black) are recorded by DCA. The mosaic bits are masked by the gray labels. We can see that the output signal at the mosaic bits are always setting at high-level whatever the real event signal at mosaic bits is. Then the real event is mosaicked at specified bits. We should note that the ripple on the output signal (see Fig. 5 (j)) is produced since the incoherence of the two arms is not completely satisfied.
To obtain different mosaic bits, we also change the OS logic signal. The CW probe beam (see Fig. 6(a), cyan) is modulated by the periodic driven signal of IM3 (‘0101’ and ‘0100’), and the output waveform (green) of IM3 is shown in Figs. 6(b) and 6(c), respectively. And the output signal of IM4 is simultaneously changed into inversed data of IM3. Then more mosaic bits are created. The real events (blue) with three data patterns ‘1011’ (RZ, dark RZ and NRZ) are shown in Figs. 6(d), 6(g) and 6(j), respectively. Figures 6(e)-6(f), 6(h)-6(i) and 6(k)-6(l) show the output waveforms of the temporal mosaic with different OS control signals. Obviously, all the mosaic bits are also set at high-level and the initial bits are erased. Then the real event is mosaicked at specified bits according to our choice by changing the OS logic signal.
In conclusion, we have proposed and experimentally demonstrated two modes for a general temporal illusion in high-speed telecommunications. Falsifying mode allows us to exactly transform some special bits into the pre-designed bits. Mosaicking mode allows us to mask some secret bits with high-level signal at any mosaic bits. All these modes extend the concept of illusion optics from falsifying object to event, and greatly widen the range of possible applications in military applications.
National Natural Science Foundation of China (NSFC) (61475052 and 61622502).
References and links
2. D. Schurig, J. J. Mock, B. J. Justice, S. A. Cummer, J. B. Pendry, A. F. Starr, and D. R. Smith, “Metamaterial electromagnetic cloak at microwave frequencies,” Science 314(5801), 977–980 (2006). [CrossRef] [PubMed]
7. Y. Lai, H. Chen, Z.-Q. Zhang, and C. T. Chan, “Complementary media invisibility cloak that cloaks objects at a distance outside the cloaking shell,” Phys. Rev. Lett. 102(9), 093901 (2009). [CrossRef] [PubMed]
8. L. H. Gabrielli, J. Cardenas, C. B. Poitras, and M. Lipson, “Silicon nanostructure cloak operating at optical frequencies,” Nat. Photonics 3(8), 461–463 (2009). [CrossRef]
12. J. S. Choi and J. C. Howell, “Digital integral cloaking,” Optica 3(5), 536–540 (2016). [CrossRef]
13. Y. Lai, J. Ng, H. Chen, D. Han, J. Xiao, Z.-Q. Zhang, and C. T. Chan, “Illusion optics: the optical transformation of an object into another object,” Phys. Rev. Lett. 102(25), 253902 (2009). [CrossRef] [PubMed]
15. C. Li, X. Meng, X. Liu, F. Li, G. Fang, H. Chen, and C. T. Chan, “Experimental realization of a circuit-based broadband illusion-optics analogue,” Phys. Rev. Lett. 105(23), 233906 (2010). [CrossRef] [PubMed]
16. W. X. Jiang, H. F. Ma, Q. Cheng, and T. J. Cui, “Illusion media: Generating virtual objects using realizable metamaterials,” Appl. Phys. Lett. 96(12), 121910 (2010). [CrossRef]
17. M. Liu, Z. Lei Mei, X. Ma, and T. J. Cui, “Dc illusion and its experimental verification,” Appl. Phys. Lett. 101(5), 051905 (2012). [CrossRef]
19. W. Xiang Jiang, S. Ge, C. Luo, and T. Jun Cui, “Localized transformation optics devices,” Appl. Phys. Lett. 103(21), 214104 (2013). [CrossRef]
21. Z. Mei, J. Bai, and T. Cui, “Illusion devices with quasi-conformal mapping,” J. Electromagn. Waves Appl. 24(17-18), 2561–2573 (2010). [CrossRef]
25. M. W. McCall, A. Favaro, P. Kinsler, and A. Boardman, “A spacetime cloak, or a history editor,” J. Opt. 13(2), 024003 (2011). [CrossRef]
26. P. Y. Bony, M. Guasoni, P. Morin, D. Sugny, A. Picozzi, H. R. Jauslin, S. Pitois, and J. Fatome, “Temporal spying and concealing process in fibre-optic data transmission systems through polarization bypass,” Nat. Commun. 5, 4678 (2014). [CrossRef] [PubMed]
28. G. Li, Y. Chen, H. Jiang, Y. Liu, X. Liu, and X. Chen, “Tunable temporal gap based on simultaneous fast and slow light in electro-optic photonic crystals,” Opt. Express 23(14), 18345–18350 (2015). [CrossRef] [PubMed]
29. R. B. Li, L. Deng, E. W. Hagley, J. C. Bienfang, M. G. Payne, and M. L. Ge, “Effect of atomic coherence on temporal cloaking in atomic vapors,” Phys. Rev. A 87(2), 023839 (2013). [CrossRef]
30. M. S. A. Jabar, B. A. Bacha, and I. Ahmad, “Temporal cloak via Doppler broadening,” Laser Phys. 25(6), 065405 (2015). [CrossRef]
31. F. Zhou, J. Dong, S. Yan, and T. Yang, “Temporal cloak with large fractional hiding window at telecommunication data rate,” Opt. Commun. 388, 77–83 (2017). [CrossRef]
34. S. Arnon and M. Fridman, “Data Center Switch Based on Temporal Cloaking,” J. Lightwave Technol. 30(21), 3427–3433 (2012). [CrossRef]
36. J. M. Lukens, A. J. Metcalf, D. E. Leaird, and A. M. Weiner, “Temporal cloaking for data suppression and retrieval,” Optica 1(6), 372–375 (2014). [CrossRef]