Information Exchange In Free Space Using Spectral Switching Of

Transcript

2952 J. Opt. Soc. Am. A / Vol. 25, No. 12 / December 2008 Yadav et al. Information exchange in free space using spectral switching of diffracted polychromatic light: possibilities and limitations Bharat Kumar Yadav,* Swati Raman, and Hem Chandra Kandpal Optical Radiation Standards, National Physical Laboratory, Dr. K.S. Krishnan Road, New Delhi, India *Corresponding author: [email protected] Received August 14, 2008; accepted September 17, 2008; posted September 25, 2008 (Doc. ID 100135); published November 11, 2008 Spectral switching is now a well-known phenomenon. Several research groups have studied it theoretically and experimentally for different optical systems. Recently, its potential applications have been demonstrated for information encoding and transmission in free space. In this paper we report experimental observations in the far field for spectral switching of polychromatic light passing through an astigmatic aperture lens. An information encoding scheme, a free-space optical link, and related boundary conditions are studied in detail. In addition, on the basis of experimental and theoretical studies carried out so far for spectral switching, we explore possibilities of information transmission in free space and analyze experimental limitations of spectralswitching-based free-space optical communication. © 2008 Optical Society of America OCIS codes: 260.6042, 200.2605, 060.4510, 200.3050, 200.4650. 1. INTRODUCTION The phenomenon of spectral switching [1] has been studied both theoretically [2–5] and experimentally [6–10] for different optical systems, and it is now a well-known phenomenon. It has been shown that the phenomenon has a natural interpretation in the framework of singular optics with polychromatic light [5]. It might have a number of potential applications, namely, free-space optical communication, optical computing, optical signal processing, and the design of spectrum-selective optical interconnects. Recently, the application of spectral switching to information encoding and transmission has been explored [11–13]. The studies show that if the redshift and blueshift of diffracted polychromatic light could be associated with information bits 1 and 0, the spectral flipping of diffracted light from lower frequency (redshift) to higher frequency (blueshift) or vice versa may be utilized for information encoding and information transmission in free space. This paper contains experimental results for far-field spectral switching and in-depth analysis of possibilities and limitations of spectral-switching-based free-space optical communication. We describe experimental results for spectral switching of polychromatic light passing through an astigmatic aperture lens. For the first time the spectral anomalies of diffracted polychromatic light are experimentally investigated for larger distances. Because of experimental limitations, our study is confined to a few meters only, but it shows the promise for future applications. Section 2 describes the experimental setup and theoretical background. Section 3 elaborates on the anomalous behavior of diffracted light in the far field. It also describes the spectral shifting regions for redshifts and blueshifts, where any value of u (the relative transversal 1084-7529/08/122952-8/$15.00 coordinate in the x direction, u = x / w0, where w0 denotes the waist width of the beam) may be used to associate an information bit with its corresponding spectral shift. Section 4 describes the free-space optical links in the phasesingularity region for spectral-switching-based optical communication. Section 5 is dedicated to proposed applications on the basis of the experimental and theoretical work carried out; it deals with an information encoding scheme, a general optical communication model and its operation, the advantages of spectral-switching-based free-space optical communication, and limitations of the proposed schemes. 2. EXPERIMENTAL SETUP AND THEORETICAL BACKGROUND The theory of spectral switching was developed principally for the ideal cases assuming that optical systems were aberration-free. However, the almost inevitable aberrations in optical systems may result in discrepancies between the theoretically predicted results and the experimentally measured observations. To study more general cases, in the last few years, the effect of optical aberrations on spectral switching has been investigated [14–16]. Very recently, experimental verification of the theoretical predictions [14] has also been reported [17]. The results of the experimental study [17] were quite interesting, but the distance between aperture lens and observation plane was limited to a few centimeters. The observations were taken at the geometrical focal plane (i.e., f = 20 cm兲 of the lens; thus ⌬z = 0, where ⌬z is the relative propagation distance ⌬z = 共z − ƒ兲 / ƒ. Here, z is the distance between aperture lens and observation plane, and f is the focal length of the lens. The experimental setup was arranged to study the second case of [14]. It be© 2008 Optical Society of America Yadav et al. came evident that one could use a lens of much larger focal length to increase the distance between aperture lens and observation plane, but at the expense of more aberrations. To make a comprehensive study, we take a rather general case in which the focal length was f = 20 cm but z varied a few meters. The experimental setup is shown in Fig. 1. It was built on a vibration-isolation table with environmental temperature fluctuating within only ±1 ° C to ensure that temperature variation did not affect the spectrum of the radiation. The primary source used was a 1000 W tungsten halogen lamp that was operated with a highly stabilized (1 part in 104) DC power supply. A diffuser D was placed between the source and the circular aperture A so that a secondary source of uniform brightness at the aperture could be produced. A broadband filter F having a Gaussian spectral distribution and peak at 628 nm was placed to illuminate the aperture A. The aperture A of radius 0.8 mm was made on a photographic film of thickness 0.1 mm. It acted as a quasi-homogenous secondary source that produced a partially coherent optical field. The van Cittert–Zernike theorem was used to generate spatially fully coherent light over the aperture lens. The coherence length 共lc = 0.16␭R / r兲 developed at the square aperture plane for ␭ = 628 nm (the peak wavelength of the source spectrum) at a distance R equal to 3.5 m from the circular aperture A was calculated to be 0.440 mm. The spectral distribution of the secondary source was assumed to be the same at all source points; hence it obeyed the so-called scaling law [18]. The aperture As was a square one having full width 共2a兲 of 0.8 mm and was placed close to the astigmatic lens L of focal length 20 cm. The on-axis spectrum S共␻兲 was obtained using a SPEX 1404 double-grating monochromator with a holographic grating of 1200 lines mm−1 blazed at 500 nm and coupled with a Peltier-cooled photomultiplier having GaAs as a cathode surface. The output of the photomultiplier was coupled to a data processing system. The distance z between the spectral switching generator plane (SSGP) (Fig. 1) and monochromator was varied by folding the light beam using mirrors (good quality mirrors with high reflectance). After passing through the astigmatic aperture lens, the modified spectrum [14] of the diffracted Fig. 1. (Color online) Schematic diagram of experimental setup. Vol. 25, No. 12 / December 2008 / J. Opt. Soc. Am. A 2953 polychromatic light expected theoretically at the observation plane in the far field is expressed as S共u,v,⌬z, ␻兲 = S共0兲共␻兲M共u,v,⌬z, ␻兲, 共1兲 where M共u,v,⌬z, ␻兲 = M1共u,⌬z, ␻兲M2共v,⌬z, ␻兲, 冋 S共0兲␻ = exp − 共 ␻ − ␻ 0兲 2 2␴02 册 . 共2兲 Here ␻0 is the central frequency and ␴0 is the bandwidth. M共u , v , ⌬z , ␻兲 is the spectral modifier that has been defined in [14] with all related parameters. 3. ANOMALOUS BEHAVIOR OF DIFFRACTED LIGHT IN FAR FIELD AND THE BOUNDARY CONDITIONS To describe the spectral anomalies of diffracted polychromatic light, we give experimental observations in Subsection A. Boundary conditions are discussed separately in Subsection B with numerically simulated results for experimental parameters. The purpose of simulation for this analysis is to investigate spectral evolution in a very small area of the dark ring where the spectral anomalies take place. A. Experimental Observations The effect of astigmatism on the spectral switching of a polychromatic Gaussian beam is illustrated in Fig. 2. Observations were taken at different distances by varying z. The observation plane is shown in Fig. 1 by O for z = 1 m and O⬘ for z = 5 m. For the sake of brevity, here we are giving only the observations for z = 5 m. In our experimental arrangement beyond this distance we were getting a poor signal-to-noise ratio and it was quite difficult to observe and record the spectral changes at the observation plane. All these observations were made on-axis. The spectrum of the radiation after transmission through the diffuser and the filter was recorded without the square aperture As and the lens L. The circles in Fig. 2 show the normalized spectrum obtained experimentally. In the next step, the square aperture As and the lens L were placed at a distance of about 3.5 m from the circular aperture A. It was observed that the spectrum of the diffracted light was not the same as recorded previously. It was found that for the relative transversal position in the x direction of the square aperture u = 0.5408, the spectral profile of the source spectrum split into two halves. This position is denoted as the critical position uc; on either side of the critical position we observed the spectrum to flip either to the lower frequency (redshift) or to the higher frequency (blueshift). The blueshift was observed at the relative transversal position of the aperture u = 0.5369, and the redshift was observed at the relative transversal position of the aperture u = 0.5441. The experimentally recorded curves blueshift, two-peak spectrum, and redshift are shown in Figs. 2(a)–2(c), respectively, by solid diamonds. In Fig. 2, the dashed (source spectral profile) and solid (modified spectrum) curves show theoretically expected curves for experimental pa- 2954 J. Opt. Soc. Am. A / Vol. 25, No. 12 / December 2008 Yadav et al. rameters using Eq. (1). The spectral changes occur in the vicinity of the first dark fringe of the diffraction pattern and in a very narrow range of ⌬u from the critical position. It was observed that a very small change in the position of the aperture leads to drastic spectral changes at the observation plane. Fig. 2. Experimentally observed spectral switching at z = 5 m: (a) blueshift, (b) TPD spectrum in critical direction, and (c) redshift. B. Spectral Evolution and Boundary Conditions Theoretically expected curves using Eq. (1) for experimental parameters are given in Fig. 3. To generate the source spectrum, the angular frequency ␻ is varied from ␻ = 2.7 ⫻ 1015 to 3.3⫻ 1015 s−1. Here the central frequency is ␻0 = 3.0⫻ 1015 s−1, C6 = 0 mm−1, and the bandwidth is ␴0 = 6.0⫻ 1013 s−1. For blueshifts [Fig. 3(a)], u (transversal coordinate in the x direction) is varied from 0.5358 (solid curve) to 0.5379 (short-dashed curve). Experimentally, by changing the position of the square aperture As in the x direction with a precise digital micrometer one can change the value of u. For the sake of convenience and to determine boundary conditions, we may assume that if the height of the second peak (SP) of the modified spectrum (after diffraction) rises from 0 to 90% (0.9) of the normalized source spectrum (reference spectrum) by changing u from 0.5358 to 0.5379, we may consider it a blueshift region. The first peak (FP) and SP of the modified spectrum are shown in Fig. 3. The blueshift region may be considered the maximum spectral shift toward higher frequencies with respect to the reference spectrum (source spectrum). In Fig. 3(a), the blueshift region is indicated by two vertical parallel dotted–dashed lines with upward arrows. Rf denotes the central reference line on the reference spectrum (source spectrum), while BS indicates the central line for maximum (acceptable) spectral shift (blueshift), i.e., for u = 0.5379. WBSR represents the width of the blueshift region. Beyond this limit, the blueshifted spectrum heads toward the two-equal-peaks spectrum (see below). Any spectrum falling in this region for a particular value of u can be treated as a blueshifted spectrum and may be used to designate information bit 0, for example. In the same manner, we may assign a redshift region [Fig. 3(c)] from u = 0.5430 (short dashed curve) to u = 0.5451 (solid curve). Again, any spectrum from this region for a particular value of u may be chosen for designating information bit 1. Here, WRSR denotes the width of the redshift region, while RS and Rf represent redshift boundary line and the reference line, respectively. WRSR is the maximum spectral shift toward lower frequencies. Some spectra for different values of u for blueshift and redshift in each region are shown in Fig. 3(a) and 3(c), respectively. In these figures, Hv denotes the height variation zone for the second peak of the modified spectrum (blueshift or redshift). Figure 3(b) shows a special condition for a critical value of uc = 0.5408, where the source spectrum splits into two halves. This spectrum may be termed the “two-equalpeaks” spectrum, where both peaks of the modified spectrum are nearly equal in height. Experimentally, it is almost impossible to get two peaks exactly equal in height for this spectrum; hence, we may accept a 10% uncertainty. If the heights of the two peaks of the two-equalpeaks spectrum are no further apart than 10% of the Yadav et al. Vol. 25, No. 12 / December 2008 / J. Opt. Soc. Am. A 2955 higher peak value [within what we call the uncertainty region (UR)], we may consider them two equal peaks for our purposes here. 4. FREE-SPACE OPTICAL LINKS FOR SPECTRAL-SWITCHING-BASED OPTICAL COMMUNICATION Foley and Wolf have shown that the spectral switching phenomenon is closely related to phase singularities [5], and Gbur et al. that the drastic spectral changes take place in the vicinity of an intensity zero in the focal region of a polychromatic, spatially coherent, converging spherical wave [19]. Simultaneously, it has been demonstrated that the spectral changes occur in the neighborhood of some special directions—which may be called “critical directions”—associated with the dark rings of the Airy pattern in the Fraunhofer region of a field produced by diffraction of a spatially coherent, polychromatic wave at an aperture [20]. The critical directions along with the dark fringes of diffraction patterns may be used to form optical links. The self-similarity of the far-zone spectrum may make it possible to transmit optical information over appreciable distances. A. Free-Space Optical Links in Fraunhofer Diffraction Free-space optical links in Fraunhofer diffraction are shown in Fig. 4. The figure also shows propagation of the diffraction pattern along with the critical directions in the vicinity of dark rings. Here O1, O2, and O3 show the first, second, and third critical directions, respectively. As the diffracted light field propagates in free space, the effective region (ER) (around the singular point) where the blueshift, two-equal-peaks spectrum, and the redshift take place expands to a very small extent around the critical direction. To obtain a better understanding of the internal structure of the diffracted light field, let us hypothetically take a longitudinal and a transverse cross section of the upper hemisphere of the first dark ring. The schematic diagram is shown in Fig. 5. RSD and BSD indicate the directions where the redshift and the blueshift occur after diffraction of polychromatic light through the aperture, while TDP (between them) shows the critical direction where the source spectrum splits into two halves. In this figure, S is the polychromatic light source, and ORSR shows the direction of propagation (z axis). Fig. 3. Simulated curves for experimental parameters to understand the spectral evolution and to determine boundary conditions for (a) blueshift, (b) TPD spectrum, (c) and redshift. Fig. 4. Optical field propagation in far field after diffraction pattern. 2956 J. Opt. Soc. Am. A / Vol. 25, No. 12 / December 2008 Yadav et al. Fig. 6. Free-space optical link through HGB between transmitter and receiver. Fig. 5. Schematic diagram to illustrate the internal structure of a free-space optical link. It shows longitudinal and transverse cross sections of the upper part of the first dark ring. O indicates the observation plane. All the directions are imaginary, but the investigation indicates that the spectral changes (as discussed in Section 3) take place around these directions. The ER is a very small region (almost a point) in the near field but it spreads to a very small extent with distance during propagation along the z axis (the direction of diffraction field propagation). Because of this spreading the blueshift region 共WBSR兲, two-equal-peaks region, and redshift region 共WRSR兲 also spread and become sufficiently broadened in the far field to make it easy to resolve the spectral changes. To illustrate the concept, the redshift, two-equalpeaks region, and blueshift regions are shown by subtended areas in Fig. 5 from point OR to the RSD (cross hatching), TPD (dark gray), and BSD (vertical hatching), respectively. The imaginary direction along with the subtended area (where the spectral anomalies occur) of each direction may be treated as the free-space optical link. Apart from the demonstration of the free-space optical link, Fig. 5 also gives a glimpse of the complex structure of the light radiation near a phase singularity [19] and its propagation in free space. It has already been shown that the spectral anomalies also take place in a small circular loop around the TPD. To show the concept a small circle with broad arrow band is shown behind the dark ring on the critical direction (TPD). The spectral changes also occur by changing the polar angle ␾ associated with this circle. The spectral changes with ␾ [12] and the relation between the diffraction angle ␪ (not shown here) and polar angle ␾ have previously been demonstrated [20]. It should be noted that for the experimental observations discussed in Subsection 3.A and [17], the spectral changes and construction of the free-space optical link would be considered in transverse cross section (Fig. 5) in the related dark ring, because in these observations we are changing the position coordinate u to produce spectral switching. In other typical experimental observations where spectral switching is controlled by changing spatial coherence [7,8] and diffraction angle [12], a longitudinal cross section will be effective to observe spectral anomalies and to form the free-space optical link. B. Free-Space Optical Link in Hollow Gaussian Beam Recently, it has been demonstrated theoretically [21] that spectral switching can be observed in a focused hollow Gaussian beam (HGB). The HGB may be ideal to create an optical link due to its interesting properties. Figure 6 depicts a schematic diagram of a free-space optical link via HGB. In this scheme the optical link lies in the center of the HGB. Again the redshift, two-equal-peaks region, and the blueshift region are shown by cross hatching, dark gray shade, and vertical hatching, respectively, along with the critical direction. The most interesting aspect of the HGB is the light intensity distribution around the free-space optical link. It may act as a shield for the free-space optical link against atmospheric turbulence. The propagation properties of the HGB have already been studied theoretically, and it has been shown that the HGB has good propagation stability in the near zone. With further increase of propagation distance z the intensity distribution diverges and the dark region across the HGB decreases. In the far field, the dark region disappears and the on-axis intensity becomes maximum [22]. The above investigations reveal that despite the HGB not being entirely suitable for larger distances, a freespace optical link may be created for distances of a few meters. Future technological advancement may increase this possibility multifold. Experimental work on this topic is underway and will soon be reported elsewhere. While analyzing spectral-switching-based free-space optical communication from a few centimeters to a few meters (maximum up to 5 m), we have not taken atmospheric turbulence into account, as for such shorter distances, the atmospheric turbulence will be almost negligible. It has already been demonstrated that for larger distances atmospheric turbulence affects the light beam during propagation in free space [23]. 5. SPECTRAL-SWITCHING-BASED INFORMATION ENCODING AND FREESPACE OPTICAL COMMUNICATION Recent research shows that spectral switching may be exploited for information encoding and transmission [11–13]. Information encoding schemes have already been discussed in detail but spectral-switching-based information transmission has not yet been treated in detail. In this section, we discuss the possibilities and experimental limitations for spectral-switching-based free-space optical communication. The analysis is based on the experimental and theoretical investigations carried out to date. The investigation may be quite significant as spectral- Yadav et al. Vol. 25, No. 12 / December 2008 / J. Opt. Soc. Am. A switching-based optical communication may be exploited to design free-space optical interconnects [24] and even free-space optical communication [25]. We give a general analysis that may be useful for both types of applications. Despite some typical experimental limitations, spectral-switching-based optical communication has some interesting advantages over other optical communication systems. It may provide a new way of communication. To explore possibilities, a general model for spectral-switching-based free-space optical communication is proposed and discussed. To ensure consistency with experimental observations, the experimental data are used for the information encoding and processing scheme. A. Free-Space Optical Communication Model A block diagram of the proposed free-space optical communication system in a laboratory frame is depicted in Fig. 7. In this model, the transmitter consists of a polychromatic light source (PLS), a spectral switching generator (SSG), and a spectral switching control mechanism (SSCM). Any optical setup that can produce spectral switching may be utilized as the SSG. The configuration of the SSCM will depend on the optical setup used for the SSG. For example, supposing we want to use the optical setup described in Section 2, we would need a precise position control system to change the value u. On the other hand, if we use the optical setup that was discussed in [7], the SSCM might be replaced by a coherence modulator. The receiver comprises a high-scanning speed, highresolution monochromator (HSHRM) with a CCD detection system. To enable two-way communication, we may construct a complete unit with one transmitter and one receiver for a particular communication point. Each unit consists of a PLS, SSG, SSCM, HSHRM, and a computer having special algorithms and protocols for free-space communication. The architecture of the system is shown by block diagram in Fig. 7. These units (Unit-I and UnitII) may be placed at suitable distances apart. The distance between the two units will depend strongly on the light source used in the optical setups, the effectiveness of the SSCM, and the suitability of the HSHRM for this communication model. Fig. 7. 2957 B. Information Encoding Scheme To illustrate the concept, a simple example of an information encoding scheme with spectral shift is shown in Fig. 8. Here R and B represent redshift 共u = 0.5441兲, i.e., 1, and blueshift 共u = 0.5369兲, i.e., 0, respectively. The decimal number 10, which is equivalent to the binary value 1010 is encoded with the combination of redshift and blueshift, i.e., RBRB. Figure 8 also shows the association of spectral shifts with the information bits. The critical value of u 共uc = 0.5408兲, where the source spectrum splits into the TPD spectrum may be used as an initial position. C. Working of the System and Internal Processing To understand the working of the system, let us take an example. Suppose we replace the SSG component of Fig. 8 with our current experimental setup (Fig. 1). In this case the SSCM would be a very precise position-changing mechanism for changing the value of u. We may use highspeed computerized position-control (position changer) systems that are commercially available. Before going into the communication process, let us list some basic assumptions for the proposed communication system. • At the beginning of the process, the system is initialized and the source spectrum has a Gaussian profile. • The receiver is at a distance where a good signal-tonoise ratio could be obtained in the far field. • The transmitter and receiver are aligned properly so that the dark ring of the diffraction pattern falls on the entrance slit of the monochromator to connect the transmitter and the receiver through a free-space optical link. • All values of related parameters (for example, u, uc for current experimental study) to generate the spectral shifts and the TPD spectrum are available in the buffer or database with the computer at the transmitter. The values of u and uc may be determined with the help of experimental observations within the uncertainty limits. Suppose a user wants to transmit a message, say, decimal number 10. The user will simply feed the number to the computer that has a program based on a special algorithm to control all the functions of the transmitter. The system will convert 10 into its binary equivalent, i.e., (Color online) General communication model for spectral-switching-based free-space optical communication. 2958 J. Opt. Soc. Am. A / Vol. 25, No. 12 / December 2008 Yadav et al. of the symmetry of the spectral shifts in the far field along the axis, the information can be transmitted to two symmetrical points at which the amount of spectral shifts is the same [11]. 2. Apart from the information bits’ association with redshift and blueshift one can entangle information bits (for example, 10) using the TPD spectrum [12]. The entangled bits provide an additional degree of freedom to encode information (1, 0, and 10). This may certainly enhance the throughput of the information transmission process. 3. Entangled bits may also be used as a special condition to conceal the information [13]. Fig. 8. (Color online) Concept of information encoding with spectral switching. 共10兲2 = 1010 and associated information bits as discussed in Subsection 5.B. Now the system is ready to process the information optically using the free-space optical link in the far field. For convenience, let us designate the plane formed by the aperture As with lens L to generate spectral switching by the SSGP. A 3-D schematic of the SSGP is shown in the inset of Fig. 1. The system starts processing (left to right) and takes the first spectral shift R from the message string 共10 = RBRB兲 that has corresponding value u = 0.5441 in the buffer of the system. Immediately, a signal goes to the SSCM to adjust the position of the SSGP for the given value of u. As a result of the repositioning of the SSGP, the source spectrum will be modified and the receiver will read the redshifted spectrum. In the next step, the SSGP will automatically readjust itself for the initial position, i.e., uc = 0.5408. It should be noted here that we are changing the position (very precisely) of square aperture As while the lens L is fixed at the SSGP. The same process of value selection and spectral shift generation will continue for each value of u until the transmitter transmits the whole string (message) to the receiver over the free-space optical link. Simultaneously, the spectra reaching the receiver plane will be recorded by means of the HSHRM. A sophisticated decoding algorithm may be used here to process the recorded spectra. It is the function of the receiver to determine the nature of each spectral shift (whether the recorded spectrum is redshifted or blueshifted). The receiver records the spectral shifts and simultaneously converts them into their corresponding information bits (redshift= 1 and blueshift= 0). The converted sequences of information bits are finally used to regenerate the original message. D. Advantages of Spectral-Switching-Based Optical Communication Information encoding and transmission in free space through spectral switching has some advantages over other optical communication systems. Some significant advantages are listed below. 1. In the proposed scheme, the information is encoded by spectral shifts; any fluctuations in the light intensity do not cause errors in bit transmission. Moreover, because E. Limitations of the Proposed Scheme The spectral-switching-based optical communication scheme is at an experimental stage and it has some typical experimental constraints that may be addressed in the near future. Some important limitations are listed below. • Although the diffraction pattern maintains its selfsimilarity and the information could be sent over appreciable distances, the signal-to-noise ratio at the far zone will affect the quality of transmission beyond certain distances. • A good broadband light source (like a so-called white light laser) is the prime requirement of the communication system to send information for longer distances. Such a broadband source may be commercially available in the near future. • Atmospheric turbulence will be a serious threat when we consider optical communication for larger distances. • In the present scenario, speed may be a major issue for such systems. On the transmitter side, if we use the optical setup discussed in [7], fast coherence modulation is required to generate spectral switching with high speed. For this system, the maximum coherence modulation speed may be 1 ␮s. This is the best speed attained so far for the coherence modulation technique [26]. On the receiver side, on the other hand, the situation is not very encouraging as the best scanning speed (demonstrated so far) is only 1000 nm/ s 共60,000 nm/ min兲. Because of these constraints, at this stage the proposed scheme cannot be compared speedwise with modern ultrafast communication schemes. 6. CONCLUSION In this paper we have discussed experimental results for spectral switching of polychromatic light in the far zone. Different optical links in the vicinity of dark rings, along with the critical direction near a phase singularity, have been discussed in detail. The results of the study indicate that if we have a broadband light source with high intensity, the spectral switching could be controlled with high speed and accuracy in a predefined manner within experimental uncertainty, and if we have a very high-resolution monochromator with high scanning speed, the information encoding and transmission using spectral switching Yadav et al. may be realizable for the near field as well as for the far field. Unfortunately, this is not so easy at this stage, but it has great potential for implementation. Realization of the proposed scheme may take the passage of time, but the study may be quite significant as these techniques may provide a new way for free-space optical communication. Vol. 25, No. 12 / December 2008 / J. Opt. Soc. Am. A 10. 11. 12. 13. ACKNOWLEDGMENTS The authors thank the Director, National Physical Laboratory, New Delhi, India for permission to publish this paper. S. Raman is also thankful to Council for Scientific and Industrial Research (CSIR) for financial support in the form of a Senior Research Fellowship. 14. 15. 16. REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. J. Pu, H. Zhang, and S. Nemoto, “Spectral shifts and spectral switches of partially coherent light passing through an aperture,” Opt. Commun. 162, 57–63 (1999). J. Pu and S. Nemoto, “Spectral shifts and spectral switches in diffraction of partially coherent light by a circular aperture,” IEEE J. Quantum Electron. 36, 1407–1411 (2000). L. Pan and B. Lu, “The spectral switch of partially coherent light in Young’s experiment,” IEEE J. Quantum Electron. 37, 1377–1381 (2001). J. Pu and S. Nemoto, “Spectral changes and 1 ⫻ N spectral switches in the diffraction of partially coherent light by an aperture,” J. Opt. Soc. Am. A 19, 339–344 (2002). J. T. Foley and E. Wolf, “Phenomenon of spectral switches as a new effect in singular optics with polychromatic light,” J. Opt. Soc. Am. A 19, 2510–2516 (2002). H. C. Kandpal, “Experimental observation of the phenomenon of spectral switch” J. Opt. A, Pure Appl. Opt. 3, 296–299 (2001). H. C. Kandpal, S. Anand, and J. S. Vaishya, “Experimental observation of the phenomenon of spectral switching for a class of partially coherent light,” IEEE J. Quantum Electron. 38, 336–339 (2002). S. Anand, B. K. Yadav, and H. C. Kandpal, “Experimental study of the phenomenon of 1 ⫻ N spectral switch due to diffraction of partially coherent light,” J. Opt. Soc. Am. A 19, 2223–2228 (2002). B. K. Yadav, S. A. M. Rizvi, and H. C. Kandpal, “Experimental observation of spectral changes of partially coherent light in Young’s experiment,” J. Opt. A, Pure Appl. Opt. 8, 72–76 (2006). 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 2959 M. M. Brundavanam, N. K. Viswanathan, and N. R. Desai, “Spectral anomalies due to temporal correlation in a whitelight interferometer,” Opt. Lett. 32, 2279–2281 (2007). J. Pu, C. Cai, and S. Nemoto, “Spectral anomalies in Young’s double-slit interference experiment,” Opt. Express 12, 5131–5139 (2004). B. K. Yadav, S. A. M. Rizvi, S. Raman, R. Mehrotra, and H. C. Kandpal, “Information encoding by spectral anamolies of spatially coherent light diffracted by an annular aperture,” Opt. Commun. 269, 253–260 (2007). B. K. Yadav, N. S. Bisht, R. Mehrotra, and H. C. Kandpal, “Diffraction-induced spectral anomalies for information encoding and information hiding—possibilities and limitations,” Opt. Commun. 277, 24–32 (2007). L. Pan and B. Lu, “Spectral switches of polychromatic Gaussian beams passing through an astigmatic aperture lens,” Opt. Commun. 234, 13–22 (2004). J. Pu, C. Cai, and S. Nemoto, “Spectral switches of partially coherent light focused by a filter-lens system with chromatic aberration,” J. Opt. Soc. Am. A 21, 994–999 (2004). X. Hu and J. Pu, “The effect of spherical aberration on singularities and spectral changes of focused beams,” New J. Phys. 8, 93 (2006). S. Raman, N. S. Bisht, B. K. Yadav, R. Mehrotra, M. Husain, and H. C. Kandpal, “Experimental observation of the effect of astigmatic aperture lens on the spectral switches of polychromatic Gaussian beam,” J. Mod. Opt. 55, 1629–1638 (2008). E. Wolf, “Invariance of the spectrum of light on propagation,” Phys. Rev. Lett. 56, 1370–1372 (1986). G. Gbur, T. D. Visser, and E. Wolf, “Anomalous behavior of spectra near phase singularities of focused waves,” Phys. Rev. Lett. 88, 013901 (2001). S. A. Ponomarenko and E. Wolf, “Spectral anomalies in a Fraunhofer diffraction pattern,” Opt. Lett. 27, 1211–1213 (2002). H. Zhang, G. Wu, and H. Guo, “Spectral anomalies of focused hollow Gaussian beams at the geometrical focal plane,” Opt. Commun. 281, 4169–4172 (2008). Y. Cai, X. Lu, and Q. Lin, “Hollow Gaussian beams and their propagation properties,” Opt. Lett. 28, 1084–1086 (2003). H. T. Eyyuboglu, Y. Cai, and B. K. Baykal, “Spectral shifts of general beams in turbulent media,” J. Opt. A, Pure Appl. Opt. 10, 015005 (2008). T. Yatagai, S. Kawai, and H. Huang, “Optical computing and interconnects,” Proc. IEEE 84, 828–852 (1996). Dominic O’Brien, “Free-space optical links for broadband wireless communication,” http://www.acreo.se/upload/ Publications/Proceedings/OE03/FREESPACE-OE003.pdf. J. Turunen, E. Tervonen, and A. T. Friberg, “Acousto-optic control and modulation of optical coherence by electronically synthesized holographic grating,” J. Appl. Phys. 67, 49–59 (1990).