Academia.eduAcademia.edu

Single Bit Comparator

https://doi.org/10.1016/J.IJLEO.2009.12.003

Abstract

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues.

Key takeaways
sparkles

AI

  1. The proposed all-optical single bit comparator utilizes four wave mixing (FWM) for efficient operation.
  2. Frequency encoding represents binary states, enhancing robustness against signal degradation during transmission.
  3. The system achieves ultra-fast operation, potentially exceeding Giga bits per second throughput.
  4. Utilizing semiconductor optical amplifiers (SOAs) allows for compact and cost-effective implementations of optical comparators.
  5. The single bit comparator can be cascaded for multibit comparisons, extending its application scope.
This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/copyright Author's personal copy Optik 121 (2010) 2230–2233 Contents lists available at ScienceDirect Optik journal homepage: www.elsevier.de/ijleo Method of developing an all optical wavelength encoded single bit comparator exploiting four wave mixing and wavelength filtering character of nonlinear semiconductor optical amplifiers Parimal Ghosh a,n, Sisir Kumar Garai b, Sourangshu Mukhopadhyay c a Department of Physics (PG & UG), B.B. College, Asansol, West Bengal 713 303, India b Department of Physics, M.U.C Women’s College, Burdwan, West Bengal 713 104, India c Department of Physics, The University of Burdwan, Burdwan, West Bengal 713 104, India a r t i c l e in f o a b s t r a c t Article history: Data Comparator is the integral part of arithmetic and logical unit of any electronic or optical data Received 3 May 2009 processor. Due to some inherent limitations of electronics it is not possible to obtain super fast data Accepted 15 September 2009 processing (over Terahertz limit) from electronic comparator. Again frequency encoding technique has been established as an excellent one over the other optical data encoding techniques. Semiconductor Keywords: optical amplifiers (SOA) technologies have shown its potentiality of realising many all-optical systems. Four wave mixing In this communication the authors have proposed a new all-optical wavelength encoded single bit Add/drop multiplexer binary comparator exploiting the four wave mixing and wavelength filtering property of nonlinear Semiconductor optical amplifier semiconductor optical amplifier. Comparator & 2009 Elsevier GmbH. All rights reserved. Wavelength conversion 1. Introduction mechanisms are adopted for representing the bits or digits with optics. In this connection the intensity encoding, polarization In digital electronic comparators the magnitudes of two binary encoding, phase encoding may be mentioned [1–7]. But these digits are compared where the binary bits are encoded by the coding processes have some inherent problems. To avoid those presence (say logic 1) or absence (say logic 0) of an electrical problems very recently frequency encoding scheme has been quantity (either voltage or current). The conventional electronic established very efficiently [10–12]. The prime beauty of the digital systems have already shown their limitations so far as the scheme is that as frequency is the fundamental property of the speed and parallel processing with a wide range of data are wave so it can preserve its identity irrespective of the absorption, concerned. To overcome these limitations optics can put its strong reflection, transmission during its propagation through different potential role in digital information processing, networking, systems as well as communicating media. On the other hand, in image processing, etc. because of its high quality of inherent last few decades there were found a large number of proposals parallelism. Being of its charge neutrality and no rest mass where optics was successfully used in many logical, arithmetic character photon has proved its superiority as information carrier and algebraic operations. In this connection we can also refer than electron in the data processors and communication many works on all optical optoelectronic logic gates and flip-flops networks. using spatial light modulators and optoelectronic transducers. But Again it is also established that in the conventional ways of a number of significant drawbacks of electronic/optoelectronic digital electronic systems, the photonic systems cannot be memory prevent the scientists and researchers to build-up a real implemented without any modification. To make a compatible life super fast optical/optoelectronic computer. We can also recall all optical system scientists and researchers from all over the the polarization encoding technique for implementing some world proposed several direct and indirect schemes for develop- optical processors because of some basic advantages of the ing the optical information processing system over the last three scheme compared to others. Also it may be exploited to decades. To implement those schemes several optical coding implement binary, tri-state or even multi-valued logical opera- tions as well with polarization encoding [13]. In comparison to these schemes, the proposal for representing binary logic states n by different frequencies of a wave removes a lot of difficulties Corresponding author. E-mail addresses: [email protected] (P. Ghosh), [email protected] what one found in other proposed schemes [10–12]. In a similar (S. Kumar Garai), [email protected] (S. Mukhopadhyay). manner, in this communication we have chosen two different and 0030-4026/$ - see front matter & 2009 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijleo.2009.12.003 Author's personal copy P. Ghosh et al. / Optik 121 (2010) 2230–2233 2231 Table 1 probe laser beam of wavelength ls and two modulated probe Excitation table of an optical comparator. input laser signals of wavelengths l1 and l2 respectively injected into an SOA. The FWM process in the amplifier gives rise to a new Inputs Intermediate outputs Final output conjugate signal that is absolutely a spectrally inverted replica A B L E G Y of the input probe signal [8,9]. For an efficient FWM in an SOA, the polarization states of the pump and the probe beams l1(0) l1(0) ls ls essentially be identical. There are two different FWM depending l1(0) l2(1) l1(0) l1 upon the co-polarized and orthogonal polarized dual pump l2(1) l1(0) l2(1) l2 l2(1) l2(1) ls ls schemes relative to the probe signal. In the former case, the two co-polarized pumps interact with the input data signal in an SOA to generate a new non-conjugate signal at the wavelength specific wavelengths l1 and l2 to represent the binary informa- indicated in Fig. 1. tion ‘0’ and ‘1’, respectively. We also exploited the highly efficient cross gain modulation (XGM) and wavelength conversion prop- 2.2. ADD/DROP multiplexing (ADM) erty of bulk nonlinear semiconductor optical amplifier (SOA), properly known as four wave mixing (FWM). Many approaches The principle of frequency routing can be achieved using the have been proposed to achieve all optical logic functions, based on special filtering properties of the reflecting SOA. By exploiting this the nonlinear effects in semiconductor optical amplifier, in optical property add/drop multiplexer (ADM) can be developed which fibres or in waveguides. Particularly, all optical logic gates based can be used to send different wavelength signals to separate on the nonlinear effects of SOAs such as cross gain modulation paths. The main function of a wavelength ADM is to separate and (XGM), cross phase modulation (XPM), four wave mixing (FWM) to identify a particular wavelength channel without interference and cross polarization modulation are promising due to SOA’s with the adjacent channels [8,9,16,17]. The advantage of using high gain in optical power, strong change of refractive index and this method is that in this case there is no loss at all encountered suitable for photonic up gradation [19]. It should be parallel during the passage of the signal through the SOA rather it is mentioned here that such properties are independent of polariza- boosted by the amplifying property of the RSOA. The filtering can tion and also insensitive to the wavelength of the input data, be achieved at any desired frequency in a range by controlling the provided if it is conducted within the SOA gain bandwidth limit bias current of the SOA. The desired wavelength channel is [8–10,14–17]. This can be controlled by the intensity of the pump reflected back by the filter after amplifying a second time by the beam. Moreover, since the FWM effect is used, our propose MQW section and directed towards the drop port by means of an scheme of binary comparator can provide an ultra fast operation optical circulator (OC). The remaining channels pass through the [18]. Input states of a single bit binary comparator are ‘1’ and ‘0’. filter section for further processing. In our system two laser pump beams of wavelength l1 and l2 are used as logic 0 and logic 1, respectively. In addition to these two pump beams a signal beam of wavelength ls is also applied to the 3. Operational scheme of the optical binary data comparator input of the nonlinear SOA. The conventional excitation table of the single bit comparator is given in Table 1, where A and B The functional truth table of an all optical wavelength encoded denote the inputs which are to be compared and L, E and G binary comparator is given in Table 1. The proposed all optical represent three possible intermediate outputs (analogous to system is shown in Fig. 2. The three beams are allowed to pass electronic counterpart). In our system we have combine the through the nonlinear SOA which produces a signal at a three different intermediate outputs to get the final output. The wavelength of lA lB + ls at its output due to the application of final output wavelengths have the meaning as follows: the signal proper bias current of SOA. This output signal is allowed to pass wavelength ls represents the equality condition, the wavelength through three successive add/drop multiplexers ADM1, ADM2 l1(0) represents the less than condition and finally the and ADM3 which are tuned for reflecting optical signals of wavelength l2(1) represents the greater than condition of the wavelengths ls ,lA lB + ls and lB lA + ls, respectively [17]. The two binary bits. This is advantageous compared to electronic reflecting wave from each ADM’s is dropped out by means of comparator which have three different outputs channels. Again three optical circulators C1, C2 and C3, respectively as shown in this single bit comparator can be used to cascade several single bit Fig. 2. The drop beams act as the pump beams for the RSOAs and comparators to achieve an all optical wavelength encoded multi- transfers their energy to the probe beams of respective RSOA bit comparator. through wavelength up or down conversion process [16]. The tuning of ADM can be made possible by varying the bias current of SOA properly used to construct the ADM. 2. Working principle Now let us explain the operation of the comparator for three different input data conditions, i e, Equality, Greater than and Less The basic principle of operation of our proposed binary optical than, with reference to our proposed all optical system as shown system of single bit comparator depends mainly on different in Fig. 2. nonlinear properties of the SOA. The two main principles involve Case-I: When input probe beams (PA and PB) are of same are (i) the Four Wave Mixing and (ii) the ADD/DROP multiplexing. wavelength say either l1 or l2, i.e., both of them contain the same optical information. And in both the conditions, at the output of 2.1. Four wave mixing the nonlinear SOA the converted signal becomes lc = ls, since lA = lB. In this case ADM1 reflects the signal having wavelength ls Four-wave mixing (FWM) is basically a third order nonlinear as it is tuned at this particular wavelength by adjusting the bias coherent process where the dielectric polarization in a medium current of the SOA constituting the wavelength ADM and the depends on the product of three electric fields. The induced reflected beam is directed towards the drop port by means of an polarization leads to the creation of new frequency components of OC marked as C1. And at the output labelled E one should obtain the electric field. SOA controlled FWM can be used to construct the signal of wavelength ls, indicating that same optical infor- wavelength converters. The basic scheme is comprised of one CW mation present at the two data inputs of the comparator (E). Author's personal copy 2232 P. Ghosh et al. / Optik 121 (2010) 2230–2233 TM Probe Probe Probe Signal Converted Signal Converted Pump A Pump B Signal output Signal S Pump A O θ A Pump B TE λ Pump A Pump B λs λA-λB+λs λA λB Fig. 1. Co-polarized dual pump wavelength converters using SOA [20]: (a) character of probe signal and pump beams, (b) schematic diagram of SOA, (c) spectrum of the output beams. Signal ADM 1 ADM 2 ADM 3 λs λA-λB+λs λB-λA+λs Probe PA SOA C1 C2 C3 Probe PB R R S S O O λA A λB A 1 2 E = λs L = λA G = λB (λA = λB) (λA < λB) (λA > λB) Final Comparator Output (OP) Fig. 2. Proposed all optical system of single bit data comparator. Table 2 PA PB Output of the SOA lc = lA lB + ls Output of the comparator at OP Remarks l1(0) l1(0) ls ls PA = PB l1(0) l2(1) l1 l2 + ls l1(0) PA o PB l2(1) l1(0) l2 l1 + ls l2(1) PA 4 PB l2(1) l2(1) ls ls PA = PB In this way the equality condition can be achieved through our easily passed through ADM1 and ADM2 but it is reflected by the proposed all optical data comparator. final member in the array (ADM3) tuned at l2 l1 + ls to move Case II: When input probe beam PA is of wavelength l1 and towards the drop port with the help of another OC marked as C3. input probe beam PB is of wavelength l2 representing PA = 0 and After this the signal reaches the RSOA2 as pump beam where it PB = 1, then at the output of the nonlinear SOA we have will transfer its energy to the weak probe beam of wavelength lc = l1 l2 + ls. Now this particular wave can easily pass through lB = l2 by the nonlinear XGM process. Hence at the output one ADM1 but it gets reflected by the 2nd ADM, here ADM2 in the obtains an optical signal of wavelength l2(G). Now as the array is tuned to reflect this particular wavelength. This wave is wavelength l2 is encoded as ‘1’ in our scheme so this will dropped with the help of another OC marked as C2 to enter to the correspond to the fact that optical data at the probe input PA is reflective SOA (RSOA1) as pump beam. RSOA1 will transfer its greater than that at other input PB. In this way the greater than power to the weak probe beam of wavelength lA = l1 by the condition can also be established in our system, along with the nonlinear cross gain modulation (XGM) process of the SOA. Hence two other conditions. Now the three outputs are connected to a at the output one obtains the optical signal of wavelength l1 (L). single comparator output (OP). Now as the wavelength l1 is encoded as ‘0’ in our scheme so this The whole operation is summarised in Table 2. will correspond to the fact that optical data at the probe input PA is less than that at other input PB. In this manner the less than condition can be checked in our system. 4. Conclusion Case-III: When input probe beam PA is of wavelength l2 and input probe beam PB is of wavelength l1 representing PA =1 and The sole operation can be performed using all optical signals PB = 0, then at the output of the nonlinear SOA we have and one might expect a throughput far above the Giga bits lc = l2 l1 + ls. Now this particular wavelength signal can be per second speed from such a device since the SOA based FWM Author's personal copy P. Ghosh et al. / Optik 121 (2010) 2230–2233 2233 operation executes an ultra high switching speed. The system is [6] S. Mukhopadhyay, Binary optical data subtraction using ternary digital all optical and hence a high degree of parallelism can be obtained representation technique in optical arithmetic problems, Appl. Opt. (USA) 31 (1992) 4622–4623. from the proposed system. As this is an SOA based system it [7] S. Mukhopadhyay, D.N. Das, P.P. Das, P. Ghosh, Implementation of all optical might be cost effective as well as compact in size and robust. digital matrix multiplication scheme with non-linear material, Opt. Eng. 40 Apart from this our proposed system exploits the potential (9) (2001) 1998–2002. advantages of the wavelength encoding techniques over the [8] M.J. Connelly, in: Semiconductor Optical Amplifiers, Kluwer Academic Publishers, Dordrecht, 2002 Chapter 7. other conventional encoding schemes used in the optical non- [9] N.K. Dutta, Q. Wang, in: Semiconductor Optical Amplifiers, World Scientific, linear material based systems. One may use the laser beams of Singapore, 2006 Chapter 8. ls = 1540 nm, l1 = 1545 nm and l2 = 1550 nm to obtain a standard [10] S.K. Garai, A. Pal, S. Mukhopadhyay, All optical frequency encoded inversion result. Otherwise the encoded signals may be chosen from the operation with tristate logic using reflecting semiconductor optical ampli- fiers, Opt. Int. J. Light Electron. Opt. (2009), doi:10.1016/j.ijleo.2009.02.011. conventional C band (1536–1570 nm) to perform the experiment. [11] S.K. Garai, S. Mukhopadhyay, A method of optical implementation of This is due to the fact that within C band the performance of the frequency encoded different logic operations using second harmonic and system will be independent of the wavelength of the signals used. difference frequency generation techniques in non-linear material, Opt. Int. J. In RSOA both shorter and longer wavelength conversions are used Light Electron. Opt. (2008), doi:10.1016/j.ijleo.2008.10.011. [12] S.K. Garai, D. Samanta, S. Mukhopadhyay, All optical implementation of but it should be noted that the extinction ratio of a signal inversion logic operation by second harmonic generation and wave mixing converted to shorter wavelength is always better [16]. The single character of some nonlinear material, Opt. Optoelectron. Technol. China 6 (4) bit comparator can be extended to compare the data of higher (2008) 43–46. number of bits and we believe that these will be included in our [13] S.K. Garai, A method of implementing polarization encoded AND and NAND operations with tristate logic, IConTOP (2009). future communication. RSOA based wavelength converters can [14] M Asghari, I.H. White, R.V. Penty, Wavelength conversion using semicon- show very good performance at probe power level as low as ductor optical amplifiers, J. Lightwave Technol. 15 (17) (1997) 1181–1190. 8 dBm compared to that of SOA based converters, which fails to [15] H. Soto, D. Erasme, G. Guekos, Cross-polarization modulation in perform below 4 dBm. semiconductor optical amplifier, IEEE Photon Technol. Lett. 11 (1999) 902–970. [16] L.Q. Guo, M.J. Connelly, A novel approach to all optical wavelength conversion References by utilizing a reflective semiconductor optical amplifier in co-polarization scheme, Opt. Commun. 281 (17) (2008) 4470–4473. [17] S. Yu, W. Gu., A tunable wavelength conversion and wavelength add/drop [1] W. Wu, S. Campbell, S. Zhou, P. Yeh, Polarisation encoded optical logic scheme based on cascaded second order nonlinearity with double pass operations in photorefractive media, Opt. Lett. (1993) 1742–1744. configuration, IEEE J. Quantum Electron. 41 (2005) 1007–1012. [2] A.A. Sawchuck, T.C. Strand, Digital optical computing, Proc. IEEE 72 (1984) [18] Li Pei-Li, Huang De-xiu, Zhang Xin-liang, Zhu Guang-xi, Ultrahigh-speed all 758–779. [3] S.K. Chandra, An all optical approach of utilizing four wave-mixing for optical half adder based on four-wave mixing in semiconductor optical developing an all optical XOR logic operation by phase encoding mechanism, amplifier, Opt. Express 14 (24) (2006) 11839–11847. IConTOP (2009). [19] H.J.S. Dorren, X. Yang, A.K. Mishra, Z. Li, H. Ju, H. de Waardt, G.D. Khoe, [4] S. Dhar, S. Mukhopdhyay, An all-optical decoding method for ASCII coded T. Simoyama, H. Ishikawa, H. Kawashima, T. Hasama, All optical logic based data using non-linear material based switching, Opt. Eng. (USA) 45 (11) on ultrafast gain and index dynamics in a semiconductor optical amplifier, (2006) 115201–115204. IEEE J. Select. Top. Quantum Electron. 10 (2004) 1079–1092. [5] S. Mukhopadhyay, D.N. Das, N. Pahari, An all optical method for the addition [20] J.P.R. Lacey, M.A. Summerfeld, S.J. Madden, Tunability of polarization of binary data by non-linear material, Appl. Opt. (USA) 43 (33) (2004) insensitive wavelength converters based on four-wave mixing in semicon- 6147–6150. ductor optical amplifiers, J Lightwave Technol. 16 (1998) 2419–2427.

References (20)

  1. W. Wu, S. Campbell, S. Zhou, P. Yeh, Polarisation encoded optical logic operations in photorefractive media, Opt. Lett. (1993) 1742-1744.
  2. A.A. Sawchuck, T.C. Strand, Digital optical computing, Proc. IEEE 72 (1984) 758-779.
  3. S.K. Chandra, An all optical approach of utilizing four wave-mixing for developing an all optical XOR logic operation by phase encoding mechanism, IConTOP (2009).
  4. S. Dhar, S. Mukhopdhyay, An all-optical decoding method for ASCII coded data using non-linear material based switching, Opt. Eng. (USA) 45 (11) (2006) 115201-115204.
  5. S. Mukhopadhyay, D.N. Das, N. Pahari, An all optical method for the addition of binary data by non-linear material, Appl. Opt. (USA) 43 (33) (2004) 6147-6150.
  6. S. Mukhopadhyay, Binary optical data subtraction using ternary digital representation technique in optical arithmetic problems, Appl. Opt. (USA) 31 (1992) 4622-4623.
  7. S. Mukhopadhyay, D.N. Das, P.P. Das, P. Ghosh, Implementation of all optical digital matrix multiplication scheme with non-linear material, Opt. Eng. 40 (9) (2001) 1998-2002.
  8. M.J. Connelly, in: Semiconductor Optical Amplifiers, Kluwer Academic Publishers, Dordrecht, 2002 Chapter 7.
  9. N.K. Dutta, Q. Wang, in: Semiconductor Optical Amplifiers, World Scientific, Singapore, 2006 Chapter 8.
  10. S.K. Garai, A. Pal, S. Mukhopadhyay, All optical frequency encoded inversion operation with tristate logic using reflecting semiconductor optical ampli- fiers, Opt. Int. J. Light Electron. Opt. (2009), doi:10.1016/j.ijleo.2009.02.011.
  11. S.K. Garai, S. Mukhopadhyay, A method of optical implementation of frequency encoded different logic operations using second harmonic and difference frequency generation techniques in non-linear material, Opt. Int. J. Light Electron. Opt. (2008), doi:10.1016/j.ijleo.2008.10.011.
  12. S.K. Garai, D. Samanta, S. Mukhopadhyay, All optical implementation of inversion logic operation by second harmonic generation and wave mixing character of some nonlinear material, Opt. Optoelectron. Technol. China 6 (4) (2008) 43-46.
  13. S.K. Garai, A method of implementing polarization encoded AND and NAND operations with tristate logic, IConTOP (2009).
  14. M Asghari, I.H. White, R.V. Penty, Wavelength conversion using semicon- ductor optical amplifiers, J. Lightwave Technol. 15 (17) (1997) 1181-1190.
  15. H. Soto, D. Erasme, G. Guekos, Cross-polarization modulation in semiconductor optical amplifier, IEEE Photon Technol. Lett. 11 (1999) 902-970.
  16. L.Q. Guo, M.J. Connelly, A novel approach to all optical wavelength conversion by utilizing a reflective semiconductor optical amplifier in co-polarization scheme, Opt. Commun. 281 (17) (2008) 4470-4473.
  17. S. Yu, W. Gu., A tunable wavelength conversion and wavelength add/drop scheme based on cascaded second order nonlinearity with double pass configuration, IEEE J. Quantum Electron. 41 (2005) 1007-1012.
  18. Li Pei-Li, Huang De-xiu, Zhang Xin-liang, Zhu Guang-xi, Ultrahigh-speed all optical half adder based on four-wave mixing in semiconductor optical amplifier, Opt. Express 14 (24) (2006) 11839-11847.
  19. H.J.S. Dorren, X. Yang, A.K. Mishra, Z. Li, H. Ju, H. de Waardt, G.D. Khoe, T. Simoyama, H. Ishikawa, H. Kawashima, T. Hasama, All optical logic based on ultrafast gain and index dynamics in a semiconductor optical amplifier, IEEE J. Select. Top. Quantum Electron. 10 (2004) 1079-1092.
  20. J.P.R. Lacey, M.A. Summerfeld, S.J. Madden, Tunability of polarization insensitive wavelength converters based on four-wave mixing in semicon- ductor optical amplifiers, J Lightwave Technol. 16 (1998) 2419-2427.

FAQs

sparkles

AI

What advances does frequency encoding offer in optical data comparison?add

The study reveals that frequency encoding preserves identity during propagation, outperforming intensity and polarization encoding methods, which had significant drawbacks.

How does the proposed comparator leverage Four Wave Mixing for operation?add

The research demonstrates that Four Wave Mixing in the SOA produces new frequency components by interacting with two laser pumps and a signal beam, enabling effective wavelength conversion.

What are the implications of using RSOA in the proposed comparator system?add

The implementation of RSOA allows for cost-effective, compact, and robust designs, capable of high-speed operations exceeding gigabits per second.

How does the proposed system ensure high parallel processing capability?add

The research indicates that the all-optical nature of the comparator utilizes nonlinear interactions, facilitating significant parallelism and reduced operational constraints compared to traditional electronic systems.

What were the findings regarding the extinction ratio in wavelength conversion?add

The study finds that RSOA can achieve better performance for shorter wavelength conversions, showcasing an extinction ratio superiority, especially at lower probe power levels around -8 dBm.

About the author
Papers
10
Followers
6
View all papers from Parimal Ghosharrow_forward